fix: harden tool-call protocol and fallback handling

* fix: surface real exception class+message in SSE error events

* fix: tighten SSE error patch scope and key redaction

* fix: redact base64-style secret suffixes fully

* style: remove notes/ reference from the dosctring

* fix: rebuild env cache on each error call

* fix: route BaseException through serde.default on SSE/webhook paths

* fix: distinguish routed providers by request URL host

* feat: normalize provider-SDK exceptions via ErrorNormalizationMiddleware

* refactor: drop json_dumpb dataclass-bypass wrappers, superseded by middleware

* fix: guard _extract_host against SDK properties that raise

* refactor: derive provider tag from ModelRequest.model, not the exception

* refactor: drop serde.default patch and exception-based inference; ProviderStreamError.model_dump handles the emit

* refactor: move envelope helpers from patches.py to errors.py

* feat: extend ErrorNormalizationMiddleware coverage to every model-call path

* chore: clean up review findings from middleware pivot

* fix: pass through all langgraph.errors

* fix: move langgraph.errors pass-through into _normalize

* fix: pass through ContextOverflowError in _normalize

.github/assets/badge-pypi-dark.svg

@@ -5,5 +5,5 @@
   <rect x="54" y="5" width="62" height="24" rx="6" fill="#1565c0"/>
   <text x="85" y="22" text-anchor="middle"
         font-family="Inter, -apple-system, system-ui, sans-serif"
-        font-size="13" font-weight="700" fill="#ffffff">v0.2.1</text>
+        font-size="13" font-weight="700" fill="#ffffff">v0.2.2</text>
 </svg>

.github/assets/badge-pypi-light.svg

@@ -5,5 +5,5 @@
   <rect x="54" y="5" width="62" height="24" rx="6" fill="#2563eb"/>
   <text x="85" y="22" text-anchor="middle"
         font-family="Inter, -apple-system, system-ui, sans-serif"
-        font-size="13" font-weight="700" fill="#ffffff">v0.2.1</text>
+        font-size="13" font-weight="700" fill="#ffffff">v0.2.2</text>
 </svg>

EvoScientist/EvoScientist.py

@@ -19,6 +19,7 @@ Usage:
 import json
 import logging
 import os
+from collections.abc import Sequence
 from pathlib import Path
 from typing import TYPE_CHECKING
 
@@ -304,8 +305,12 @@ def _inject_subagent_middleware(
     path doesn't fall back to the global-writing ``_ensure_chat_model()``.
     """
     from .middleware import (
+        DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD,
         ContextOverflowMapperMiddleware,
+        ErrorNormalizationMiddleware,
+        RepetitiveToolCallGuardMiddleware,
         ToolErrorHandlerMiddleware,
+        ToolProtocolGuardMiddleware,
         create_context_editing_middleware,
         create_memory_lifecycle_middleware,
         create_memory_middleware,
@@ -314,6 +319,16 @@ def _inject_subagent_middleware(
     )
 
     cfg = cfg if cfg is not None else _ensure_config()
+    repetitive_tool_call_threshold = getattr(
+        cfg,
+        "repetitive_tool_call_threshold",
+        DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD,
+    )
+    if not isinstance(repetitive_tool_call_threshold, int):
+        repetitive_tool_call_threshold = DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD
+    max_consecutive_tool_errors = getattr(cfg, "max_consecutive_tool_errors", 3)
+    if not isinstance(max_consecutive_tool_errors, int):
+        max_consecutive_tool_errors = 3
     memory_controls = MemoryControls.from_config(cfg)
     memory_dir = str(_paths_mod.MEMORIES_DIR)
     memory_scheduler = default_memory_scheduler()
@@ -333,6 +348,16 @@ def _inject_subagent_middleware(
             memory_scheduler=memory_scheduler,
         )
         middleware = [
+            # Outermost — catches provider-SDK exceptions from the
+            # model call (including inner middlewares) and normalizes
+            # them into a non-dataclass envelope wrapper before
+            # anything downstream sees them.
+            ErrorNormalizationMiddleware(),
+            RepetitiveToolCallGuardMiddleware(
+                threshold=repetitive_tool_call_threshold,
+                max_consecutive_errors=max_consecutive_tool_errors,
+            ),
+            ToolProtocolGuardMiddleware(),
             # Subagents share the main agent's model: use the threaded
             # ``chat_model`` on the pure path, else defer to the factory's
             # ``_ensure_chat_model()`` fallback (when ``chat_model=None``).
@@ -641,9 +666,14 @@ def _get_default_middleware(
     *,
     for_async_subagent: bool = False,
     workspace_dir: str | Path | None = None,
+    memory_dir: str | Path | None = None,
     cfg=None,
     chat_model=None,
     memory_source_agent: str = "EvoScientist",
+    tool_selector_threshold: int | None = None,
+    memory_max_inline_profile_chars: int | None = None,
+    enable_background_execution: bool = True,
+    enable_legacy_model_fallback: bool = True,
 ):
     """Build the default middleware list.
 
@@ -665,10 +695,14 @@ def _get_default_middleware(
             Async sub-agent factories pass their deployed agent name here.
     """
     from .middleware import (
+        DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD,
         ConfigurableModelMiddleware,
         ContextOverflowMapperMiddleware,
+        ErrorNormalizationMiddleware,
         ModelFallbackMiddleware,
+        RepetitiveToolCallGuardMiddleware,
         ToolErrorHandlerMiddleware,
+        ToolProtocolGuardMiddleware,
         create_code_interpreter_middleware,
         create_context_editing_middleware,
         create_memory_lifecycle_middleware,
@@ -681,10 +715,20 @@ def _get_default_middleware(
     )
 
     cfg = cfg if cfg is not None else _ensure_config()
+    repetitive_tool_call_threshold = getattr(
+        cfg,
+        "repetitive_tool_call_threshold",
+        DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD,
+    )
+    if not isinstance(repetitive_tool_call_threshold, int):
+        repetitive_tool_call_threshold = DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD
+    max_consecutive_tool_errors = getattr(cfg, "max_consecutive_tool_errors", 3)
+    if not isinstance(max_consecutive_tool_errors, int):
+        max_consecutive_tool_errors = 3
     if cfg.model_fallbacks:
         load_fallback_chain(cfg.model_fallbacks)
     model = chat_model if chat_model is not None else _ensure_chat_model()
-    memory_dir = str(_paths_mod.MEMORIES_DIR)
+    memory_dir = str(memory_dir or _paths_mod.MEMORIES_DIR)
     source_type = (
         MemorySourceType.SUBAGENT if for_async_subagent else MemorySourceType.TURN
     )
@@ -699,18 +743,20 @@ def _get_default_middleware(
     # ``ModelFallbackMiddleware``: a configurable.model override sets the
     # PRIMARY model only, leaving the fallback chain free to try its own
     # alternatives instead of re-overriding every retry to the same model.
-    memory_middleware = create_memory_middleware(
-        memory_dir,
-        workspace_dir=workspace_dir,
-        source_type=source_type,
-        source_agent=memory_source_agent,
-        enable_profile_memory=memory_controls.profile_enabled,
-        enable_observation_memory=memory_controls.observations_enabled,
-        enable_observation_tool=memory_controls.observation_tool_enabled(
+    memory_kwargs = {
+        "workspace_dir": workspace_dir,
+        "source_type": source_type,
+        "source_agent": memory_source_agent,
+        "enable_profile_memory": memory_controls.profile_enabled,
+        "enable_observation_memory": memory_controls.observations_enabled,
+        "enable_observation_tool": memory_controls.observation_tool_enabled(
             MemoryObservationTarget.AGENT
         ),
-        memory_scheduler=memory_scheduler,
-    )
+        "memory_scheduler": memory_scheduler,
+    }
+    if memory_max_inline_profile_chars is not None:
+        memory_kwargs["max_inline_profile_chars"] = memory_max_inline_profile_chars
+    memory_middleware = create_memory_middleware(memory_dir, **memory_kwargs)
     # Main-agent tool selection may use the auxiliary model; async sub-agents
     # keep the main model (they do real work, not a one-off helper call).
     # context_editing stays on the main model — its model only sizes the
@@ -728,16 +774,32 @@ def _get_default_middleware(
             from .llm import get_chat_model
 
             tool_selector_model = get_chat_model(model=aux_model, provider=aux_provider)
+    selector_middlewares = create_tool_selector_middleware(
+        **(
+            {"threshold": tool_selector_threshold}
+            if tool_selector_threshold is not None
+            else {}
+        ),
+        model=tool_selector_model,
+        track_stream_selection=not for_async_subagent,
+    )
     mw = [
+        # Outermost — catches provider-SDK exceptions from the model
+        # call (including exceptions surfaced through inner
+        # middlewares) and normalizes them into a non-dataclass
+        # envelope wrapper before anything downstream sees them.
+        ErrorNormalizationMiddleware(),
         ConfigurableModelMiddleware(),
         create_context_editing_middleware(model),
-        ModelFallbackMiddleware(),
+        *([ModelFallbackMiddleware()] if enable_legacy_model_fallback else []),
+        RepetitiveToolCallGuardMiddleware(
+            threshold=repetitive_tool_call_threshold,
+            max_consecutive_errors=max_consecutive_tool_errors,
+        ),
         ContextOverflowMapperMiddleware(),
         ToolErrorHandlerMiddleware(),
-        *create_tool_selector_middleware(
-            model=tool_selector_model,
-            track_stream_selection=not for_async_subagent,
-        ),
+        *selector_middlewares,
+        ToolProtocolGuardMiddleware(),
         # Interpreter prompt must land before runtime/memory context, so this
         # middleware sits ahead of runtime_context in the stack.
         create_code_interpreter_middleware(
@@ -770,7 +832,7 @@ def _get_default_middleware(
     # Background-process tools (run_in_background / check_process / stop_process /
     # list_processes) — main agent only. Async sub-agents run on langgraph-dev and
     # must not spawn local OS processes.
-    if not for_async_subagent:
+    if not for_async_subagent and enable_background_execution:
         from .middleware.background import BackgroundExecutionMiddleware
 
         mw.append(BackgroundExecutionMiddleware())
@@ -868,6 +930,14 @@ def create_cli_agent(
     chat_model=None,
     *,
     on_mcp_progress=None,
+    workspace_backend=None,
+    memory_dir: str | Path | None = None,
+    tool_selector_threshold: int | None = None,
+    memory_max_inline_profile_chars: int | None = None,
+    enable_subagents: bool = True,
+    enable_background_execution: bool = True,
+    main_agent_outer_middlewares: Sequence[AgentMiddleware] | None = None,
+    main_agent_route_middleware: AgentMiddleware | None = None,
 ) -> "CompiledStateGraph":
     """Create agent with checkpointer for CLI multi-turn support.
 
@@ -894,6 +964,22 @@ def create_cli_agent(
         chat_model: Optional pre-built chat model.  Only triggers the pure
             path when ``config`` is also explicit; otherwise it is ignored in
             favor of the ``_ensure_chat_model()`` fallback.
+        workspace_backend: Optional host-provided backend for the workspace
+            route. The default remains ``CustomSandboxBackend``.
+        memory_dir: Optional memory root used by both the backend route and
+            memory middleware.
+        tool_selector_threshold: Optional adaptive tool-selection threshold.
+        memory_max_inline_profile_chars: Optional memory profile injection cap.
+        enable_subagents: Whether configured subagents are available to the agent.
+        enable_background_execution: Whether local background-process tools are
+            installed. Embedding hosts should disable this when process execution
+            is provided by an external backend.
+        main_agent_outer_middlewares: Optional host-owned middleware installed
+            only on the top-level agent, outside EvoScientist's default chain.
+        main_agent_route_middleware: Optional host-owned route middleware placed
+            after ConfigurableModelMiddleware and before tool selection. When
+            provided, EvoScientist's legacy model fallback is disabled for the
+            top-level agent so the host is the only fallback authority.
     """
     import os as _os
 
@@ -935,19 +1021,21 @@ def create_cli_agent(
         workspace_dir = str(_paths.WORKSPACE_ROOT)
 
     # Read paths dynamically so runtime set_workspace_root() changes are picked up
-    _mem_dir = str(_paths.MEMORIES_DIR)
+    _mem_dir = str(memory_dir or _paths.MEMORIES_DIR)
     _usr_skills_dir = str(_paths.USER_SKILLS_DIR)
     _global_skills_dir = str(_paths.GLOBAL_SKILLS_DIR)
 
     # Always construct fresh backends from current paths (avoids stale
     # module-level backend when workspace root changed at runtime).
     set_active_workspace(workspace_dir)
-    ws_backend = CustomSandboxBackend(
-        root_dir=workspace_dir,
-        virtual_mode=True,
-        timeout=cfg.sandbox_execute_timeout,
-        dangerous=cfg.dangerous_mode,
-    )
+    ws_backend = workspace_backend
+    if ws_backend is None:
+        ws_backend = CustomSandboxBackend(
+            root_dir=workspace_dir,
+            virtual_mode=True,
+            timeout=cfg.sandbox_execute_timeout,
+            dangerous=cfg.dangerous_mode,
+        )
     sk_backend = MergedSkillsBackend(
         primary_dir=_usr_skills_dir,
         global_dir=_global_skills_dir,
@@ -969,8 +1057,29 @@ def create_cli_agent(
     # CLI agent never drifts from the default chain. Anything CLI-specific
     # (e.g. ``HumanInTheLoopMiddleware``) is appended below.
     mw: list[AgentMiddleware] = _get_default_middleware(
-        workspace_dir=workspace_dir, cfg=cfg, chat_model=chat_model
+        workspace_dir=workspace_dir,
+        memory_dir=_mem_dir,
+        cfg=cfg,
+        chat_model=chat_model,
+        tool_selector_threshold=tool_selector_threshold,
+        memory_max_inline_profile_chars=memory_max_inline_profile_chars,
+        enable_background_execution=enable_background_execution,
+        enable_legacy_model_fallback=main_agent_route_middleware is None,
     )
+    if main_agent_route_middleware is not None:
+        configurable_index = next(
+            (
+                index
+                for index, middleware in enumerate(mw)
+                if getattr(middleware, "name", "") == "configurable_model"
+            ),
+            None,
+        )
+        if configurable_index is None:
+            raise RuntimeError("ConfigurableModelMiddleware route slot is unavailable")
+        mw.insert(configurable_index + 1, main_agent_route_middleware)
+    if main_agent_outer_middlewares:
+        mw = [*main_agent_outer_middlewares, *mw]
 
     # HITL on main agent only — passing `interrupt_on=` to create_deep_agent
     # would propagate it to every subagent, breaking parallel execute calls
@@ -995,6 +1104,8 @@ def create_cli_agent(
         chat_model=chat_model,
         workspace_dir=workspace_dir,
     )
+    if not enable_subagents:
+        kwargs = {**kwargs, "subagents": []}
 
     return create_deep_agent(
         **kwargs,

EvoScientist/__init__.py

@@ -9,6 +9,8 @@ from __future__ import annotations
 
 from importlib import import_module
 
+__version__ = "0.2.2"
+
 _EXPORTS: dict[str, tuple[str, str]] = {
     # Agent graph (lazy to avoid expensive initialization at import time)
     "EvoScientist_agent": (".EvoScientist", "EvoScientist_agent"),

EvoScientist/ccproxy_manager.py

@@ -20,6 +20,9 @@ from EvoScientist.config import EvoScientistConfig
 
 logger = logging.getLogger(__name__)
 
+_CCPROXY_AUTH_TIMEOUT_SECONDS = 30
+_CCPROXY_HEALTH_TIMEOUT_SECONDS = 180
+
 
 # =============================================================================
 # Availability & auth checks
@@ -127,7 +130,11 @@ def check_ccproxy_auth(provider: str = "claude_api") -> tuple[bool, str]:
             [exe, "auth", "status", provider],
             capture_output=True,
             text=True,
-            timeout=10,
+            # ccproxy's CLI initializes its full plugin system on every
+            # invocation — a cold start takes ~10s on Apple Silicon, so a
+            # 10s timeout made OAuth startup fail intermittently with
+            # "Auth check timed out".
+            timeout=_CCPROXY_AUTH_TIMEOUT_SECONDS,
         )
         import re as _re
 
@@ -176,6 +183,33 @@ def is_ccproxy_running(port: int) -> bool:
         return False
 
 
+def write_ccproxy_config() -> str:
+    """Write the ccproxy config file EvoScientist passes to ``serve --config``.
+
+    Disables ccproxy's default Codex model mappings, which rewrite any
+    ``gpt-*``/``o1-*``/``o3-*``/``claude-*`` model to ``gpt-5.3-codex``
+    before forwarding — silently overriding the model the user configured
+    (and failing outright on accounts where ``gpt-5.3-codex`` is not
+    served). With no mappings, the requested model reaches the Codex
+    backend unmodified.
+
+    Returns:
+        Absolute path to the generated config file.
+    """
+    from EvoScientist.config import get_config_dir
+
+    path = get_config_dir() / "ccproxy.toml"
+    path.parent.mkdir(parents=True, exist_ok=True)
+    path.write_text(
+        "# Generated by EvoScientist (ccproxy_manager) — do not edit;\n"
+        "# regenerated on every ccproxy start.\n"
+        "[plugins.codex]\n"
+        "model_mappings = []\n",
+        encoding="utf-8",
+    )
+    return str(path)
+
+
 def start_ccproxy(port: int) -> subprocess.Popen:
     """Start ccproxy serve as a background process.
 
@@ -186,18 +220,32 @@ def start_ccproxy(port: int) -> subprocess.Popen:
         The Popen handle for the ccproxy process.
 
     Raises:
-        RuntimeError: If ccproxy fails to become healthy within 30 seconds.
+        RuntimeError: If ccproxy fails to become healthy within
+            ``_CCPROXY_HEALTH_TIMEOUT_SECONDS``.
         FileNotFoundError: If ccproxy binary is not found.
     """
     exe = _ccproxy_exe() or "ccproxy"
+    cmd = [exe, "serve", "--port", str(port)]
+    try:
+        cmd += ["--config", write_ccproxy_config()]
+    except (OSError, UnicodeError) as exc:
+        logger.warning(
+            "Could not write ccproxy config (%s); starting with defaults — "
+            "Codex model mappings will rewrite gpt-* models to gpt-5.3-codex",
+            exc,
+        )
+    logger.warning(
+        "Starting ccproxy on port %d; first startup may take up to %d seconds",
+        port,
+        _CCPROXY_HEALTH_TIMEOUT_SECONDS,
+    )
     proc = subprocess.Popen(
-        [exe, "serve", "--port", str(port)],
+        cmd,
         stdout=subprocess.DEVNULL,
         stderr=subprocess.DEVNULL,
     )
 
-    # Wait for health (ccproxy can take up to ~11s on first start)
-    deadline = time.monotonic() + 30
+    deadline = time.monotonic() + _CCPROXY_HEALTH_TIMEOUT_SECONDS
     while time.monotonic() < deadline:
         if proc.poll() is not None:
             raise RuntimeError(
@@ -213,7 +261,10 @@ def start_ccproxy(port: int) -> subprocess.Popen:
         proc.wait(timeout=3)
     except subprocess.TimeoutExpired:
         proc.kill()
-    raise RuntimeError("ccproxy did not become healthy within 30 seconds")
+    raise RuntimeError(
+        "ccproxy did not become healthy within "
+        f"{_CCPROXY_HEALTH_TIMEOUT_SECONDS} seconds"
+    )
 
 
 def stop_ccproxy(proc: subprocess.Popen | None) -> None:

EvoScientist/channels/middleware.py

@@ -75,11 +75,13 @@ class DedupCache:
         max_size: int = _DEDUP_MAX,
         trim_to: int = _DEDUP_TRIM,
         ttl_seconds: float = _DEDUP_TTL,
+        clock: Callable[[], float] | None = None,
     ) -> None:
         self._seen: OrderedDict[str, float] = OrderedDict()
         self._max = max_size
         self._trim = trim_to
         self._ttl = ttl_seconds
+        self._clock = clock or time.monotonic
 
     # ── public API ──────────────────────────────────────────────────
 
@@ -93,15 +95,16 @@ class DedupCache:
         if not msg_id:
             return False
 
-        self._prune()
+        now = self._clock()
+        self._prune(now)
 
         if msg_id in self._seen:
             # LRU: refresh position and timestamp
             self._seen.move_to_end(msg_id)
-            self._seen[msg_id] = time.monotonic()
+            self._seen[msg_id] = now
             return True
 
-        self._seen[msg_id] = time.monotonic()
+        self._seen[msg_id] = now
         if len(self._seen) > self._max:
             while len(self._seen) > self._trim:
                 self._seen.popitem(last=False)
@@ -118,9 +121,9 @@ class DedupCache:
 
     # ── internal ────────────────────────────────────────────────────
 
-    def _prune(self) -> None:
+    def _prune(self, now: float | None = None) -> None:
         """Remove entries older than *ttl_seconds*."""
-        cutoff = time.monotonic() - self._ttl
+        cutoff = (self._clock() if now is None else now) - self._ttl
         # OrderedDict is insertion-ordered; oldest entries are first.
         while self._seen:
             _key, ts = next(iter(self._seen.items()))
@@ -428,11 +431,13 @@ class DedupMiddleware(InboundMiddleware):
         max_size: int = 1000,
         trim_to: int = 500,
         ttl_seconds: float = 3600.0,
+        clock: Callable[[], float] | None = None,
     ) -> None:
         self._cache = DedupCache(
             max_size=max_size,
             trim_to=trim_to,
             ttl_seconds=ttl_seconds,
+            clock=clock,
         )
 
     async def process_inbound(

EvoScientist/config/onboard/steps.py

@@ -371,14 +371,31 @@ def _step_minimax_region(config: EvoScientistConfig) -> str:
     return _MINIMAX_REGIONS[region]
 
 
-def _step_anthropic_auth_mode(config: EvoScientistConfig) -> str:
-    """Step 2a: Select Anthropic authentication mode (API key vs OAuth).
+def _step_oauth_auth_mode(
+    config: EvoScientistConfig,
+    *,
+    provider_label: str,
+    ccproxy_provider: str,
+    config_attr: str,
+    prompt_login_label: str,
+    oauth_choice_label: str | None = None,
+    status_label: str | None = None,
+    question_label: str | None = None,
+) -> str:
+    """Select API-key vs ccproxy OAuth authentication for a provider.
 
     Args:
         config: Current configuration.
+        provider_label: Provider display name for direct API-key access.
+        ccproxy_provider: ccproxy auth provider name.
+        config_attr: Config attribute storing this provider's auth mode.
+        prompt_login_label: Label used in "Log in to ..." prompts.
+        oauth_choice_label: Optional display label for the OAuth choice.
+        status_label: Optional display label for status messages.
+        question_label: Optional prompt label override.
 
     Returns:
-        Selected auth mode: "api_key", "oauth", or "auto".
+        Selected auth mode: "api_key" or "oauth".
     """
     from ...ccproxy_manager import check_ccproxy_auth, is_ccproxy_available
 
@@ -386,10 +403,14 @@ def _step_anthropic_auth_mode(config: EvoScientistConfig) -> str:
 
     from .prompter import BACK_SENTINEL, GoBack, install_navigation_keys
 
+    oauth_label = oauth_choice_label or f"{prompt_login_label} OAuth"
+    auth_status_label = status_label or oauth_label
+    auth_question_label = question_label or f"{provider_label} authentication mode"
+
     choices = [
-        Choice(title="API Key (direct Anthropic access)", value="api_key"),
+        Choice(title=f"API Key (direct {provider_label} access)", value="api_key"),
         Choice(
-            title="Claude Code OAuth (via ccproxy — no API key needed)"
+            title=f"{oauth_label} (via ccproxy — no API key needed)"
             + (
                 ""
                 if ccproxy_available
@@ -401,12 +422,12 @@ def _step_anthropic_auth_mode(config: EvoScientistConfig) -> str:
         Choice(title="← Back (re-pick provider)", value=BACK_SENTINEL),
     ]
 
-    current = config.anthropic_auth_mode
+    current = getattr(config, config_attr)
     if current not in ("api_key", "oauth"):
         current = "api_key"
 
     question = questionary.select(
-        "Authentication mode  [Esc/← to go back]:",
+        f"{auth_question_label}  [Esc/← to go back]:",
         choices=choices,
         default=current,
         style=WIZARD_STYLE,
@@ -448,11 +469,9 @@ def _step_anthropic_auth_mode(config: EvoScientistConfig) -> str:
     if auth_mode == "oauth":
         _prompt_ccproxy_port(config)
 
-    # If OAuth selected, check auth status and offer login
-    if auth_mode in ("oauth", "auto"):
-        authed, msg = check_ccproxy_auth()
+        authed, msg = check_ccproxy_auth(ccproxy_provider)
         if authed:
-            console.print(f"  [green]✓ OAuth: {msg}[/green]")
+            console.print(f"  [green]✓ {auth_status_label}: {msg}[/green]")
             relogin = questionary.confirm(
                 "Re-authenticate to refresh credentials?",
                 default=False,
@@ -462,11 +481,13 @@ def _step_anthropic_auth_mode(config: EvoScientistConfig) -> str:
             if relogin is None:
                 raise KeyboardInterrupt()
             if relogin:
-                _run_ccproxy_login("claude_api", "OAuth")
+                _run_ccproxy_login(ccproxy_provider, auth_status_label)
         else:
-            console.print(f"  [yellow]OAuth not authenticated: {msg}[/yellow]")
+            console.print(
+                f"  [yellow]{auth_status_label} not authenticated: {msg}[/yellow]"
+            )
             login = questionary.confirm(
-                "Log in to Claude now?",
+                f"Log in to {prompt_login_label} now?",
                 default=True,
                 style=CONFIRM_STYLE,
                 qmark=QMARK,
@@ -474,11 +495,32 @@ def _step_anthropic_auth_mode(config: EvoScientistConfig) -> str:
             if login is None:
                 raise KeyboardInterrupt()
             if login:
-                _run_ccproxy_login("claude_api", "OAuth")
+                _run_ccproxy_login(ccproxy_provider, auth_status_label)
 
     return auth_mode
 
 
+def _step_anthropic_auth_mode(config: EvoScientistConfig) -> str:
+    """Step 2a: Select Anthropic authentication mode (API key vs OAuth).
+
+    Args:
+        config: Current configuration.
+
+    Returns:
+        Selected auth mode: "api_key" or "oauth".
+    """
+    return _step_oauth_auth_mode(
+        config,
+        provider_label="Anthropic",
+        ccproxy_provider="claude_api",
+        config_attr="anthropic_auth_mode",
+        prompt_login_label="Claude",
+        oauth_choice_label="Claude Code OAuth",
+        status_label="OAuth",
+        question_label="Authentication mode",
+    )
+
+
 def _step_openai_auth_mode(config: EvoScientistConfig) -> str:
     """Step 2b: Select OpenAI authentication mode (API key vs Codex OAuth).
 
@@ -488,101 +530,16 @@ def _step_openai_auth_mode(config: EvoScientistConfig) -> str:
     Returns:
         Selected auth mode: "api_key" or "oauth".
     """
-    from ...ccproxy_manager import check_ccproxy_auth, is_ccproxy_available
-
-    ccproxy_available = is_ccproxy_available()
-
-    from .prompter import BACK_SENTINEL, GoBack, install_navigation_keys
-
-    choices = [
-        Choice(title="API Key (direct OpenAI access)", value="api_key"),
-        Choice(
-            title="Codex OAuth (via ccproxy — no API key needed)"
-            + (
-                ""
-                if ccproxy_available
-                else " [requires: pip install evoscientist[oauth]]"
-            ),
-            value="oauth",
-        ),
-        questionary.Separator(),
-        Choice(title="← Back (re-pick provider)", value=BACK_SENTINEL),
-    ]
-
-    current = config.openai_auth_mode
-    if current not in ("api_key", "oauth"):
-        current = "api_key"
-
-    question = questionary.select(
-        "OpenAI authentication mode  [Esc/← to go back]:",
-        choices=choices,
-        default=current,
-        style=WIZARD_STYLE,
-        qmark=QMARK,
-        use_indicator=True,
+    return _step_oauth_auth_mode(
+        config,
+        provider_label="OpenAI",
+        ccproxy_provider="codex",
+        config_attr="openai_auth_mode",
+        prompt_login_label="Codex",
+        oauth_choice_label="Codex OAuth",
+        status_label="Codex OAuth",
+        question_label="OpenAI authentication mode",
     )
-    install_navigation_keys(question, with_back=True)
-    auth_mode = question.ask()
-
-    if auth_mode is None:
-        raise KeyboardInterrupt()
-    if auth_mode == BACK_SENTINEL:
-        raise GoBack()
-
-    if auth_mode == "oauth" and not ccproxy_available:
-        console.print("  [yellow]✗ ccproxy not installed[/yellow]")
-        console.print()
-        install = questionary.confirm(
-            'Install ccproxy now? (pip install "evoscientist[oauth]")',
-            default=True,
-            style=WIZARD_STYLE,
-            qmark=f"  {QMARK}",
-        ).ask()
-        if install is None:
-            raise KeyboardInterrupt()
-        if install:
-            console.print()
-            if _install_ccproxy():
-                console.print("  [green]✓ ccproxy installed successfully.[/green]")
-            else:
-                console.print("  [yellow]Falling back to API key mode.[/yellow]")
-                return "api_key"
-        else:
-            console.print(
-                '  [dim]Skipped. Install manually: pip install "evoscientist[oauth]"[/dim]'
-            )
-            return "api_key"
-
-    # If OAuth selected, prompt for port and check auth status
-    if auth_mode == "oauth":
-        _prompt_ccproxy_port(config)
-        authed, msg = check_ccproxy_auth("codex")
-        if authed:
-            console.print(f"  [green]✓ Codex OAuth: {msg}[/green]")
-            relogin = questionary.confirm(
-                "Re-authenticate to refresh credentials?",
-                default=False,
-                style=CONFIRM_STYLE,
-                qmark=QMARK,
-            ).ask()
-            if relogin is None:
-                raise KeyboardInterrupt()
-            if relogin:
-                _run_ccproxy_login("codex", "Codex OAuth")
-        else:
-            console.print(f"  [yellow]Codex OAuth not authenticated: {msg}[/yellow]")
-            login = questionary.confirm(
-                "Log in to Codex now?",
-                default=True,
-                style=CONFIRM_STYLE,
-                qmark=QMARK,
-            ).ask()
-            if login is None:
-                raise KeyboardInterrupt()
-            if login:
-                _run_ccproxy_login("codex", "Codex OAuth")
-
-    return auth_mode
 
 
 def _step_provider_api_key(

EvoScientist/config/onboard/wizard.py

@@ -129,6 +129,12 @@ _PROVIDER_KEY_ATTR = {
     "custom-anthropic": "custom_anthropic_api_key",
 }
 
+_MINIMAX_GLOBAL_BASE_URL = "https://api.minimax.io/anthropic"
+_CUSTOM_PROVIDER_BASE_URL = {
+    "custom-openai": ("custom_openai_base_url", "CUSTOM_OPENAI_BASE_URL"),
+    "custom-anthropic": ("custom_anthropic_base_url", "CUSTOM_ANTHROPIC_BASE_URL"),
+}
+
 
 def _autosave(config: EvoScientistConfig) -> None:
     """Persist current config to disk between phases.
@@ -142,6 +148,201 @@ def _autosave(config: EvoScientistConfig) -> None:
         pass
 
 
+def _configure_provider_base_url(
+    config: EvoScientistConfig,
+    provider: str,
+    *,
+    strict: bool,
+) -> list[str]:
+    """Configure provider-specific base URL/region and return Ollama models."""
+    if provider in _CUSTOM_PROVIDER_BASE_URL:
+        attr_name, env_name = _CUSTOM_PROVIDER_BASE_URL[provider]
+        current_base_url = getattr(config, attr_name) or os.environ.get(env_name, "")
+        if strict:
+            if not current_base_url:
+                raise RuntimeError(
+                    f"--non-interactive: {provider} provider needs a base URL. "
+                    f"Set the {env_name} env var or run without --non-interactive."
+                )
+            setattr(config, attr_name, current_base_url)
+        else:
+            setattr(
+                config,
+                attr_name,
+                _step_base_url(config, current_value=current_base_url),
+            )
+    elif provider == "minimax":
+        if strict:
+            config.minimax_base_url = (
+                config.minimax_base_url or _MINIMAX_GLOBAL_BASE_URL
+            )
+        else:
+            config.minimax_base_url = _step_minimax_region(config)
+    elif provider == "ollama":
+        if strict:
+            config.ollama_base_url = (
+                config.ollama_base_url
+                or os.environ.get("OLLAMA_BASE_URL", "")
+                or "http://localhost:11434"
+            )
+        else:
+            ollama_url, ollama_detected_models = _step_ollama_base_url(config)
+            config.ollama_base_url = ollama_url
+            return ollama_detected_models
+    return []
+
+
+def _configure_provider_auth_mode(
+    config: EvoScientistConfig,
+    provider: str,
+    *,
+    strict: bool,
+) -> None:
+    """Configure Anthropic/OpenAI auth mode for the selected provider."""
+    if provider == "anthropic":
+        if strict:
+            config.anthropic_auth_mode = "api_key"
+        else:
+            config.anthropic_auth_mode = _step_anthropic_auth_mode(config)
+    elif provider == "openai":
+        if strict:
+            config.openai_auth_mode = "api_key"
+        else:
+            config.openai_auth_mode = _step_openai_auth_mode(config)
+
+
+def _active_llm_providers(config: EvoScientistConfig) -> set[str]:
+    """Return providers currently selected by the main and auxiliary models."""
+    providers = {config.provider}
+    if config.auxiliary_provider:
+        providers.add(config.auxiliary_provider)
+    return providers
+
+
+def _reconcile_oauth_modes(config: EvoScientistConfig) -> None:
+    """Clear OAuth flags for providers no selected model uses."""
+    active_providers = _active_llm_providers(config)
+    if "anthropic" not in active_providers:
+        config.anthropic_auth_mode = "api_key"
+    if "openai" not in active_providers:
+        config.openai_auth_mode = "api_key"
+
+
+def _provider_uses_oauth(config: EvoScientistConfig, provider: str) -> bool:
+    return (provider == "anthropic" and config.anthropic_auth_mode == "oauth") or (
+        provider == "openai" and config.openai_auth_mode == "oauth"
+    )
+
+
+def _apply_preset_provider_api_key(
+    config: EvoScientistConfig,
+    provider: str,
+    preset_api_key: str,
+    *,
+    skip_validation: bool,
+) -> None:
+    """Validate and store a CLI-supplied provider API key."""
+    if not skip_validation:
+        from .helpers import _provider_key_info
+
+        _info = _provider_key_info(config, provider)
+        validate_fn = _info[2] if _info else None
+        if validate_fn is not None:
+            console.print("  [dim]Validating preset API key...[/dim]", end="")
+            valid, msg = validate_fn(preset_api_key)
+            if valid:
+                console.print(f"\r  [green]✓ {msg}[/green]      ")
+            else:
+                console.print(f"\r  [red]✗ {msg}[/red]      ")
+                raise RuntimeError(
+                    f"--api-key rejected by {provider} validator: {msg}. "
+                    "Pass --skip-validation to override."
+                )
+
+    key_attr = _PROVIDER_KEY_ATTR.get(provider, "openai_api_key")
+    setattr(config, key_attr, preset_api_key)
+    console.print(
+        f"  [green]✓ API key: ***{preset_api_key[-4:]}[/green]   [dim](--api-key)[/dim]"
+    )
+
+
+def _configure_provider_api_key(
+    config: EvoScientistConfig,
+    provider: str,
+    *,
+    skip_validation: bool,
+    preset_api_key: str | None = None,
+    require_api_key=None,
+) -> None:
+    """Configure provider API key unless the provider does not need one."""
+    if provider == "ollama" or _provider_uses_oauth(config, provider):
+        return
+
+    key_attr = _PROVIDER_KEY_ATTR.get(provider, "openai_api_key")
+    if preset_api_key is not None:
+        _apply_preset_provider_api_key(
+            config,
+            provider,
+            preset_api_key,
+            skip_validation=skip_validation,
+        )
+        return
+
+    if require_api_key is not None:
+        require_api_key()
+    new_key = _step_provider_api_key(config, provider, skip_validation)
+    if new_key is not None:
+        setattr(config, key_attr, new_key)
+    elif not getattr(config, key_attr):
+        _print_step_skipped("API Key", "not set")
+
+
+def _provider_connection_configured(config: EvoScientistConfig, provider: str) -> bool:
+    """Return True when provider-level setup can be safely reused."""
+    if provider == "ollama":
+        return bool(config.ollama_base_url)
+    if provider == "custom-openai" and not config.custom_openai_base_url:
+        return False
+    if provider == "custom-anthropic" and not config.custom_anthropic_base_url:
+        return False
+    if provider == "minimax" and not config.minimax_base_url:
+        return False
+    if _provider_uses_oauth(config, provider):
+        return True
+    key_attr = _PROVIDER_KEY_ATTR.get(provider, "openai_api_key")
+    return bool(getattr(config, key_attr))
+
+
+def _configure_provider_connection(
+    config: EvoScientistConfig,
+    provider: str,
+    *,
+    strict: bool,
+    skip_validation: bool,
+    preset_api_key: str | None = None,
+    require_api_key=None,
+) -> list[str]:
+    """Configure provider base URL/region, auth mode, and API key."""
+    ollama_detected_models = _configure_provider_base_url(
+        config,
+        provider,
+        strict=strict,
+    )
+    _configure_provider_auth_mode(
+        config,
+        provider,
+        strict=strict,
+    )
+    _configure_provider_api_key(
+        config,
+        provider,
+        skip_validation=skip_validation,
+        preset_api_key=preset_api_key,
+        require_api_key=require_api_key,
+    )
+    return ollama_detected_models
+
+
 # Sections offered in Keep/Modify/Reset → which step labels they enable.
 _SECTION_LABELS: list[tuple[str, str]] = [
     ("ui", "UI backend"),
@@ -479,102 +680,17 @@ def run_onboard(
                         provider = _step_provider(config)
                         config.provider = provider
 
-                    # Step 2a: Base URL (custom-openai, custom-anthropic,
-                    # minimax, ollama). In strict non-interactive mode we
-                    # never call the interactive _step_base_url /
-                    # _step_minimax_region / _step_ollama_base_url helpers —
-                    # fall back to the existing config value or the
-                    # CUSTOM_*_BASE_URL / OLLAMA_BASE_URL env var instead.
-                    # If neither is set for a provider that needs it, raise
-                    # so the user sees the same "missing required answer"
-                    # error as for other required prompts.
-                    if provider == "custom-openai":
-                        current_base_url = (
-                            config.custom_openai_base_url
-                            or os.environ.get("CUSTOM_OPENAI_BASE_URL", "")
-                        )
-                        if strict:
-                            if not current_base_url:
-                                raise RuntimeError(
-                                    "--non-interactive: custom-openai provider "
-                                    "needs a base URL. Set the "
-                                    "CUSTOM_OPENAI_BASE_URL env var or run "
-                                    "without --non-interactive."
-                                )
-                            config.custom_openai_base_url = current_base_url
-                        else:
-                            config.custom_openai_base_url = _step_base_url(
-                                config, current_value=current_base_url
-                            )
-                    elif provider == "custom-anthropic":
-                        current_base_url = (
-                            config.custom_anthropic_base_url
-                            or os.environ.get("CUSTOM_ANTHROPIC_BASE_URL", "")
-                        )
-                        if strict:
-                            if not current_base_url:
-                                raise RuntimeError(
-                                    "--non-interactive: custom-anthropic "
-                                    "provider needs a base URL. Set the "
-                                    "CUSTOM_ANTHROPIC_BASE_URL env var or run "
-                                    "without --non-interactive."
-                                )
-                            config.custom_anthropic_base_url = current_base_url
-                        else:
-                            config.custom_anthropic_base_url = _step_base_url(
-                                config, current_value=current_base_url
-                            )
-                    elif provider == "minimax":
-                        if strict:
-                            # MiniMax has 2 region URLs; default to whatever
-                            # is already in config, else the Global endpoint.
-                            config.minimax_base_url = (
-                                config.minimax_base_url
-                                or "https://api.minimax.io/anthropic"
-                            )
-                        else:
-                            config.minimax_base_url = _step_minimax_region(config)
-                    elif provider == "ollama":
-                        if strict:
-                            # Ollama: existing config value > env var >
-                            # localhost default. Skip the live connection
-                            # validation under strict — model discovery
-                            # happens at runtime anyway.
-                            config.ollama_base_url = (
-                                config.ollama_base_url
-                                or os.environ.get("OLLAMA_BASE_URL", "")
-                                or "http://localhost:11434"
-                            )
-                            # ollama_detected_models stays [] — model picker
-                            # will fall back to free-text or the preset.
-                        else:
-                            ollama_url, ollama_detected_models = _step_ollama_base_url(
-                                config
-                            )
-                            config.ollama_base_url = ollama_url
-
-                    # Step 2b: Auth mode (Anthropic or OpenAI — API key vs OAuth).
-                    # In strict non-interactive mode we assume "api_key".
-                    # The prompt offers a `← Back` choice that raises GoBack so
-                    # the user can re-pick the provider without exiting the wizard.
                     try:
-                        if provider == "anthropic":
-                            if strict:
-                                config.anthropic_auth_mode = "api_key"
-                            else:
-                                config.anthropic_auth_mode = _step_anthropic_auth_mode(
-                                    config
-                                )
-                        elif provider == "openai":
-                            if strict:
-                                config.openai_auth_mode = "api_key"
-                            else:
-                                config.openai_auth_mode = _step_openai_auth_mode(config)
-                        else:
-                            # Non-Anthropic/OpenAI provider: reset OAuth modes to
-                            # avoid stale oauth config triggering ccproxy at startup.
-                            config.anthropic_auth_mode = "api_key"
-                            config.openai_auth_mode = "api_key"
+                        ollama_detected_models = _configure_provider_connection(
+                            config,
+                            provider,
+                            strict=strict,
+                            skip_validation=skip_validation,
+                            preset_api_key=_preset("api_key"),
+                            require_api_key=lambda provider=provider: _require(
+                                "api_key", f"{provider} API key"
+                            ),
+                        )
                     except GoBack:
                         # User picked "← Back" — restore config to its state at the
                         # top of this iteration (drops any base_url / region /
@@ -594,60 +710,9 @@ def run_onboard(
                         ollama_detected_models = []
                         console.print("  [dim]↩ Returning to provider selection.[/dim]")
                         continue
-                    break  # auth_mode succeeded — exit sub-loop
+                    break  # Provider setup succeeded — exit sub-loop
 
-                # Step 2c: Provider API Key (skip for Ollama and pure OAuth)
-                _skip_api_key = (
-                    provider == "ollama"
-                    or (
-                        provider == "anthropic"
-                        and config.anthropic_auth_mode == "oauth"
-                    )
-                    or (provider == "openai" and config.openai_auth_mode == "oauth")
-                )
-                if not _skip_api_key:
-                    key_attr = _PROVIDER_KEY_ATTR.get(provider, "openai_api_key")
-                    preset_api_key = _preset("api_key")
-                    if preset_api_key is not None:
-                        # Validate the preset key against the same validator
-                        # the interactive path uses, unless --skip-validation
-                        # was passed. Interactive flow shows a "Save anyway?"
-                        # confirm on failure; the non-interactive path has no
-                        # way to ask, so a failed validation is fatal.
-                        if not skip_validation:
-                            from .helpers import _provider_key_info
-
-                            _info = _provider_key_info(config, provider)
-                            validate_fn = _info[2] if _info else None
-                            if validate_fn is not None:
-                                console.print(
-                                    "  [dim]Validating preset API key...[/dim]",
-                                    end="",
-                                )
-                                valid, msg = validate_fn(preset_api_key)
-                                if valid:
-                                    console.print(f"\r  [green]✓ {msg}[/green]      ")
-                                else:
-                                    console.print(f"\r  [red]✗ {msg}[/red]      ")
-                                    raise RuntimeError(
-                                        f"--api-key rejected by {provider} "
-                                        f"validator: {msg}. Pass "
-                                        "--skip-validation to override."
-                                    )
-                        setattr(config, key_attr, preset_api_key)
-                        console.print(
-                            f"  [green]✓ API key: ***{preset_api_key[-4:]}[/green]"
-                            "   [dim](--api-key)[/dim]"
-                        )
-                    else:
-                        _require("api_key", f"{provider} API key")
-                        new_key = _step_provider_api_key(
-                            config, provider, skip_validation
-                        )
-                        if new_key is not None:
-                            setattr(config, key_attr, new_key)
-                        elif not getattr(config, key_attr):
-                            _print_step_skipped("API Key", "not set")
+                _reconcile_oauth_modes(config)
                 _autosave(config)
             else:
                 # Provider section skipped — keep prior provider value to drive
@@ -680,44 +745,55 @@ def run_onboard(
                         "kept current" if config.auxiliary_model else "not set",
                     )
                 elif _step_auxiliary_enable(config):
-                    # Assemble: pick provider -> base URL (custom) -> key -> model,
-                    # mirroring the main flow's order. Keys/base URLs are stored
-                    # per provider, so when the auxiliary provider matches the main
-                    # one they're already set and the user just keeps them (Enter).
-                    # Ollama needs no key. Re-runs default to the saved auxiliary
-                    # provider/model rather than the main ones.
-                    aux_provider = _step_provider(
-                        config,
-                        label="co-pilot",
-                        default_value=config.auxiliary_provider,
-                    )
-                    config.auxiliary_provider = aux_provider
-                    if aux_provider == "custom-openai":
-                        config.custom_openai_base_url = _step_base_url(
+                    from .prompter import GoBack
+
+                    aux_ollama_detected_models: list[str] = []
+                    while True:
+                        loop_snapshot = copy.deepcopy(config)
+                        aux_provider = _step_provider(
                             config,
-                            current_value=config.custom_openai_base_url
-                            or os.environ.get("CUSTOM_OPENAI_BASE_URL", ""),
+                            label="co-pilot",
+                            default_value=config.auxiliary_provider,
                         )
-                    elif aux_provider == "custom-anthropic":
-                        config.custom_anthropic_base_url = _step_base_url(
-                            config,
-                            current_value=config.custom_anthropic_base_url
-                            or os.environ.get("CUSTOM_ANTHROPIC_BASE_URL", ""),
-                        )
-                    elif aux_provider == "minimax":
-                        config.minimax_base_url = _step_minimax_region(config)
-                    if aux_provider != "ollama":
-                        aux_key_attr = _PROVIDER_KEY_ATTR.get(
-                            aux_provider, "openai_api_key"
-                        )
-                        new_aux_key = _step_provider_api_key(
-                            config, aux_provider, skip_validation
-                        )
-                        if new_aux_key is not None:
-                            setattr(config, aux_key_attr, new_aux_key)
+                        config.auxiliary_provider = aux_provider
+                        if (
+                            aux_provider == config.provider
+                            and _provider_connection_configured(config, aux_provider)
+                        ):
+                            if aux_provider == "ollama":
+                                aux_ollama_detected_models = ollama_detected_models
+                            _print_step_skipped(
+                                "Co-pilot credentials",
+                                "reusing main provider settings",
+                            )
+                        else:
+                            try:
+                                aux_ollama_detected_models = (
+                                    _configure_provider_connection(
+                                        config,
+                                        aux_provider,
+                                        strict=False,
+                                        skip_validation=skip_validation,
+                                    )
+                                )
+                            except GoBack:
+                                for field_name in vars(loop_snapshot):
+                                    setattr(
+                                        config,
+                                        field_name,
+                                        getattr(loop_snapshot, field_name),
+                                    )
+                                aux_ollama_detected_models = []
+                                console.print(
+                                    "  [dim]↩ Returning to co-pilot provider "
+                                    "selection.[/dim]"
+                                )
+                                continue
+                        break
                     config.auxiliary_model = _step_model(
                         config,
                         aux_provider,
+                        ollama_detected_models=aux_ollama_detected_models,
                         label="co-pilot",
                         default_value=config.auxiliary_model,
                     )
@@ -725,6 +801,7 @@ def run_onboard(
                     # Skip: single driver — clear any prior auxiliary config.
                     config.auxiliary_provider = ""
                     config.auxiliary_model = ""
+                _reconcile_oauth_modes(config)
                 _autosave(config)
 
             if "tavily" in sections_to_run:

EvoScientist/config/settings.py

@@ -106,11 +106,23 @@ def _normalize_hhmm(value: Any) -> str | None:
 def get_config_dir() -> Path:
     """Get the configuration directory path.
 
-    Uses XDG_CONFIG_HOME if set, otherwise ~/.config/evoscientist/
+    Priority:
+        1. EVOSCIENTIST_CONFIG_DIR
+        2. EVOSCIENTIST_HOME/config
+        3. XDG_CONFIG_HOME/evoscientist
+        4. ~/.config/evoscientist
     """
+    configured = os.environ.get("EVOSCIENTIST_CONFIG_DIR")
+    if configured:
+        return Path(configured).expanduser().resolve()
+
+    home = os.environ.get("EVOSCIENTIST_HOME")
+    if home:
+        return Path(home).expanduser().resolve() / "config"
+
     xdg_config = os.environ.get("XDG_CONFIG_HOME")
     if xdg_config:
-        return Path(xdg_config) / "evoscientist"
+        return Path(xdg_config).expanduser() / "evoscientist"
     return Path.home() / ".config" / "evoscientist"
 
 
@@ -123,6 +135,16 @@ def get_config_path() -> Path:
 # Configuration dataclass
 # =============================================================================
 
+# OpenRouter app-attribution defaults (issue #339). Single source of truth: the
+# EvoScientistConfig fields below default to these, and llm/models.py imports
+# them for its env-fallback, so the values never drift across the two layers.
+OPENROUTER_DEFAULT_HTTP_REFERER = "https://github.com/EvoScientist/EvoScientist"
+OPENROUTER_DEFAULT_APP_TITLE = "EvoScientist"
+# OpenRouter honors only the first 2 categories per request (server-side limit)
+# and silently ignores the rest, so keep the two most relevant ones. Chosen per
+# maintainer review — creative-writing is a less competitive marketplace group.
+OPENROUTER_DEFAULT_APP_CATEGORIES = "creative-writing,personal-agent"
+
 
 @dataclass
 class EvoScientistConfig:
@@ -240,6 +262,13 @@ class EvoScientistConfig:
     # Lower (e.g., 5000) if you want a tighter safety net against runaway loops.
     recursion_limit: int = 1_000_000
 
+    # Number of consecutive model rounds with the same structured tool name and
+    # arguments that activates provider-facing loop repair. Set 0 to disable.
+    repetitive_tool_call_threshold: int = 2
+    # Number of consecutive deterministic tool errors allowed before the next
+    # model call is blocked. Transient provider/network errors are not counted.
+    max_consecutive_tool_errors: int = 3
+
     # Memory Settings
     # Profile memory injects and maintains `/memories/profile/...` files.
     memory_profile_enabled: bool = True
@@ -278,10 +307,21 @@ class EvoScientistConfig:
     # a deploy-style langgraph server instead of the in-terminal CLI/TUI.
     ui_backend: Literal["cli", "tui", "webui"] = "tui"
     log_level: str = "warning"
-    reasoning_effort: str = "high"
+    # Empty means use the provider/model default. A non-empty value is an
+    # explicit user override exported as EVOSCIENTIST_REASONING_EFFORT.
+    reasoning_effort: str = ""
     # Anthropic prompt caching for OpenRouter anthropic/* models. Opt out if
     # cache-write costs outweigh the benefit for a workflow.
     openrouter_anthropic_prompt_cache: bool = True
+    # OpenRouter app attribution (issue #339). Sent only for the openrouter
+    # provider; identifies EvoScientist in OpenRouter's app rankings/analytics.
+    # Override (e.g. a private fork) via these fields or their env vars.
+    # Defaults live in the module constants above (also imported by llm/models.py).
+    openrouter_http_referer: str = OPENROUTER_DEFAULT_HTTP_REFERER
+    openrouter_app_title: str = OPENROUTER_DEFAULT_APP_TITLE
+    # Comma-separated; split into a list before being passed to
+    # langchain-openrouter (its app_categories kwarg expects list[str]).
+    openrouter_app_categories: str = OPENROUTER_DEFAULT_APP_CATEGORIES
 
     # Channel Settings
     channel_enabled: str = ""  # "imessage" | "telegram" | "discord" | "slack" | "wechat" | "dingtalk" | "feishu" | "email" | "qq" | "signal" | "" (comma-separated for multiple)
@@ -440,6 +480,14 @@ class EvoScientistConfig:
     stt_compute_type: str = "int8"  # "int8" | "float16" | "float32"
 
     def __post_init__(self) -> None:
+        for field_name in (
+            "repetitive_tool_call_threshold",
+            "max_consecutive_tool_errors",
+        ):
+            value = getattr(self, field_name)
+            if not isinstance(value, int) or isinstance(value, bool) or value < 0:
+                raise ValueError(f"{field_name} must be a non-negative integer")
+
         # A non-positive or non-int sandbox_execute_timeout (e.g. a hand-edited
         # config file value — load_config does not coerce file values — or a
         # 0/negative env value) would raise inside CustomSandboxBackend.__init__
@@ -548,6 +596,10 @@ def save_config(config: EvoScientistConfig) -> None:
     """
     config_path = get_config_path()
     config_path.parent.mkdir(parents=True, exist_ok=True)
+    try:
+        config_path.parent.chmod(0o700)
+    except OSError:
+        pass
 
     data = _config_to_dict(config)
 
@@ -560,6 +612,10 @@ def save_config(config: EvoScientistConfig) -> None:
             sort_keys=False,
             allow_unicode=True,
         )
+    try:
+        config_path.chmod(0o600)
+    except OSError:
+        pass
 
 
 def reset_config() -> None:
@@ -694,6 +750,11 @@ def set_config_value(key: str, value: Any) -> bool:
 
     if key == "sandbox_execute_timeout" and value <= 0:
         return False
+    if key in {
+        "repetitive_tool_call_threshold",
+        "max_consecutive_tool_errors",
+    } and (isinstance(value, bool) or value < 0):
+        return False
     if key == "memory_skill_synthesis_time":
         value = _normalize_hhmm(value)
         if value is None:
@@ -753,6 +814,9 @@ _ENV_MAPPINGS = {
     "openrouter_anthropic_prompt_cache": (
         "EVOSCIENTIST_OPENROUTER_ANTHROPIC_PROMPT_CACHE"
     ),
+    "openrouter_http_referer": "EVOSCIENTIST_OPENROUTER_HTTP_REFERER",
+    "openrouter_app_title": "EVOSCIENTIST_OPENROUTER_APP_TITLE",
+    "openrouter_app_categories": "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES",
     "dangerous_mode": "EVOSCIENTIST_DANGEROUS_MODE",
     "channel_debug_tracing": "EVOSCIENTIST_CHANNEL_DEBUG_TRACING",
     "ccproxy_port": "EVOSCIENTIST_CCPROXY_PORT",
@@ -769,6 +833,10 @@ _ENV_MAPPINGS = {
     "langgraph_dev_file_persistence": "EVOSCIENTIST_LANGGRAPH_DEV_FILE_PERSISTENCE",
     "langgraph_dev_jobs_per_worker": "EVOSCIENTIST_LANGGRAPH_DEV_JOBS_PER_WORKER",
     "recursion_limit": "EVOSCIENTIST_RECURSION_LIMIT",
+    "repetitive_tool_call_threshold": (
+        "EVOSCIENTIST_REPETITIVE_TOOL_CALL_THRESHOLD"
+    ),
+    "max_consecutive_tool_errors": "EVOSCIENTIST_MAX_CONSECUTIVE_TOOL_ERRORS",
     "memory_profile_enabled": "EVOSCIENTIST_MEMORY_PROFILE_ENABLED",
     "memory_observations_enabled": "EVOSCIENTIST_MEMORY_OBSERVATIONS_ENABLED",
     "memory_observation_writer": "EVOSCIENTIST_MEMORY_OBSERVATION_WRITER",
@@ -884,6 +952,22 @@ def apply_config_to_env(config: EvoScientistConfig) -> None:
         os.environ["TAVILY_API_KEY"] = config.tavily_api_key
     if config.reasoning_effort and not os.environ.get("EVOSCIENTIST_REASONING_EFFORT"):
         os.environ["EVOSCIENTIST_REASONING_EFFORT"] = config.reasoning_effort
+    if config.openrouter_http_referer and not os.environ.get(
+        "EVOSCIENTIST_OPENROUTER_HTTP_REFERER"
+    ):
+        os.environ["EVOSCIENTIST_OPENROUTER_HTTP_REFERER"] = (
+            config.openrouter_http_referer
+        )
+    if config.openrouter_app_title and not os.environ.get(
+        "EVOSCIENTIST_OPENROUTER_APP_TITLE"
+    ):
+        os.environ["EVOSCIENTIST_OPENROUTER_APP_TITLE"] = config.openrouter_app_title
+    if config.openrouter_app_categories and not os.environ.get(
+        "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES"
+    ):
+        os.environ["EVOSCIENTIST_OPENROUTER_APP_CATEGORIES"] = (
+            config.openrouter_app_categories
+        )
     if not config.openrouter_anthropic_prompt_cache and not os.environ.get(
         "EVOSCIENTIST_OPENROUTER_ANTHROPIC_PROMPT_CACHE"
     ):

EvoScientist/langgraph_dev/http.py

@@ -43,7 +43,7 @@ async def get_models(_request: Request) -> JSONResponse:
     ``discover_ollama_models()`` call, same 1.5-s timeout, same
     fail-soft semantics (the probe returns ``[]`` on any error, never
     raises). The TUI's "Custom Ollama model…" sentinel is intentionally
-    omitted: that's a widget-specific input affordance, not part of
+    omitted — that's a widget-specific input affordance, not part of
     the registry surface.
 
     ``default`` reflects the deployment's currently-configured fallback

EvoScientist/langgraph_dev/main_graph.py

@@ -5,8 +5,244 @@ in ``EvoScientist/EvoScientist.py`` so it doesn't construct on plain
 ``import EvoScientist``. ``langgraph dev`` 's symbol resolver inspects
 module attributes directly and doesn't trigger ``__getattr__``, so we
 re-export here to make it visible.
+
+Before re-export we upgrade the compiled graph's class in place to
+``_EvoFilteredGraph``, which strips ``PrivateStateAttr``-marked fields
+(currently just ``_quickjs_snapshot_payload``) from ``get_state`` /
+``get_state_history`` responses. Upstream ``langchain_quickjs`` annotates
+the field ``PrivateStateAttr = OmitFromSchema(input=True, output=True)``,
+but LangGraph's ``_prepare_state_snapshot`` doesn't honor that on
+checkpoint reads — every ``getState`` materializes the delta chain back
+into a full ~1.4 MB blob, which the WebUI then downloads. The subclass
+closes the gap without touching the middleware's write path, preserving
+cross-turn REPL persistence as ``langchain-ai/deepagents#3064`` shipped it.
 """
 
-from EvoScientist.EvoScientist import EvoScientist_agent
+from langgraph.graph.state import CompiledStateGraph
+from langgraph.types import PregelTask, StateSnapshot
+
+from EvoScientist.EvoScientist import EvoScientist_agent as _agent
+
+_PRIVATE_STATE_FIELDS = frozenset({"_quickjs_snapshot_payload"})
+
+# Sanity check on the LangGraph internals ``_strip_private`` scrubs. If any
+# of these attributes disappear or get renamed in a future upstream bump,
+# the assertion fires at import time — the deployment refuses to start,
+# instead of silently degrading (the filter would ``.get()`` its way to a
+# no-op and the private-field payload would come back on the wire without
+# anyone noticing until a user reports slow thread switches again).
+#
+# Doesn't cover every internal we depend on — ``metadata["writes"]`` /
+# ``metadata["counters_since_delta_snapshot"]`` dict keys aren't a canary
+# target because ``dict.get`` already tolerates their absence. What we
+# canary here is the ``NamedTuple`` field set: renames there would be the
+# highest-impact silent regression.
+_EXPECTED_SNAPSHOT_FIELDS = frozenset({"values", "metadata", "tasks"})
+_EXPECTED_TASK_FIELDS = frozenset({"result", "state"})
+
+_missing_snap = _EXPECTED_SNAPSHOT_FIELDS - set(StateSnapshot._fields)
+_missing_task = _EXPECTED_TASK_FIELDS - set(PregelTask._fields)
+if _missing_snap or _missing_task:
+    raise RuntimeError(
+        "LangGraph state shape drifted from the version _strip_private was "
+        f"written against. Missing StateSnapshot fields: {_missing_snap or set()}. "
+        f"Missing PregelTask fields: {_missing_task or set()}. Review "
+        "_strip_private and re-verify against the current upstream shape "
+        "before removing this assertion."
+    )
+
+
+def _strip_private(snap):
+    """Strip ``PrivateStateAttr``-marked fields from a ``StateSnapshot``.
+
+    Empirically verified against a live history response for a thread with
+    a single touched turn: the private field leaks on four surfaces — three
+    trivial, one heavy:
+
+    * ``snap.values`` — the materialized channel state exposed as the main
+      payload. For DeltaChannels this is the delta chain replayed into full
+      bytes (~1.4 MB for the quickjs snapshot). ``get_state`` and every
+      history entry.
+    * ``snap.metadata['writes']`` — ``{node_name: {channel: value}}`` map of
+      the raw writes that produced each checkpoint. On the ``after_agent``
+      step that first snapshots the REPL, ``value`` is the encoded write
+      record ``("snap", full_bytes)`` ≈ 1.4 MB.
+    * ``snap.tasks[*].result`` — the return dict of each completed
+      ``PregelTask``. ``after_agent`` returns
+      ``{"_quickjs_snapshot_payload": ("snap", bytes)}``; this dict becomes
+      the task's ``result`` field, which the API surfaces verbatim under
+      ``tasks[*].result`` (``langgraph_api.state:106``). This is the
+      dominant leak: 1.7 MB in the last history entry of any thread whose
+      most-recent-in-window checkpoint had a snapshot anchor.
+    * ``snap.metadata['counters_since_delta_snapshot']`` — DeltaChannel's
+      snapshot cadence bookkeeping ``{channel: [count, superstep]}``. Tiny
+      (~20 B) but exposes the private field name; strip for cleanliness.
+    * ``snap.tasks[*].state`` (nested ``StateSnapshot``) — populated when the
+      caller passes ``subgraphs=True``. Repeats all of the above surfaces
+      for each subgraph task, so recurse into it. Not exercised by the
+      current WebUI (which doesn't pass ``subgraphs=True`` on REST reads),
+      but SDK / curl / gRPC callers can.
+    """
+    if snap is None:
+        return snap
+    values = {k: v for k, v in snap.values.items() if k not in _PRIVATE_STATE_FIELDS}
+    metadata = snap.metadata
+    if metadata:
+        new_metadata = metadata
+        if new_metadata.get("writes"):
+            scrubbed_writes = {
+                node: {
+                    k: v for k, v in ch_writes.items() if k not in _PRIVATE_STATE_FIELDS
+                }
+                for node, ch_writes in new_metadata["writes"].items()
+            }
+            new_metadata = {**new_metadata, "writes": scrubbed_writes}
+        if new_metadata.get("counters_since_delta_snapshot"):
+            scrubbed_counters = {
+                k: v
+                for k, v in new_metadata["counters_since_delta_snapshot"].items()
+                if k not in _PRIVATE_STATE_FIELDS
+            }
+            new_metadata = {
+                **new_metadata,
+                "counters_since_delta_snapshot": scrubbed_counters,
+            }
+        metadata = new_metadata
+    tasks = snap.tasks
+    if tasks:
+        new_tasks = []
+        changed = False
+        for t in tasks:
+            replace_kwargs: dict = {}
+            result = getattr(t, "result", None)
+            if isinstance(result, dict) and any(
+                k in result for k in _PRIVATE_STATE_FIELDS
+            ):
+                replace_kwargs["result"] = {
+                    k: v for k, v in result.items() if k not in _PRIVATE_STATE_FIELDS
+                }
+            # ``t.state`` is a ``RunnableConfig | StateSnapshot | None`` per
+            # ``PregelTask``'s typing. When ``subgraphs=True`` on the caller,
+            # this holds the subgraph's fully-materialized ``StateSnapshot`` —
+            # which repeats the same four leak surfaces (``values``,
+            # ``metadata.writes``, ``metadata.counters_since_delta_snapshot``,
+            # ``tasks[*].result/state``). Recurse so the whole tree is clean.
+            nested_state = getattr(t, "state", None)
+            if isinstance(nested_state, StateSnapshot):
+                scrubbed_state = _strip_private(nested_state)
+                if scrubbed_state is not nested_state:
+                    replace_kwargs["state"] = scrubbed_state
+            if replace_kwargs:
+                new_tasks.append(t._replace(**replace_kwargs))
+                changed = True
+            else:
+                new_tasks.append(t)
+        if changed:
+            tasks = tuple(new_tasks)
+    return snap._replace(values=values, metadata=metadata, tasks=tasks)
+
+
+class _EvoFilteredGraph(CompiledStateGraph):
+    """Filters ``PrivateStateAttr``-marked state fields from checkpoint reads.
+
+    ``Pregel.copy`` uses ``self.__class__(**attrs)`` so this subclass
+    survives the ``graph_obj.copy(update=...)`` call in
+    ``langgraph_api.graph.get_graph`` that binds the checkpointer / store
+    before yielding to endpoint handlers.
+
+    **Known gap — streaming paths.** The overrides only cover ``get_state``
+    / ``get_state_history``. On this compiled graph,
+    ``self.output_channels`` correctly excludes ``_quickjs_snapshot_payload``
+    (respects ``OmitFromSchema(output=True)``), but
+    ``self.stream_channels_asis`` includes it alongside other private
+    fields (``jump_to``, ``_summarization_event``) — the two lists are
+    built by ``langgraph.graph.state``'s graph builder and only the first
+    checks the output schema. So a client streaming with
+    ``stream_mode="values"`` or ``stream_mode="events"`` (which fall back
+    to ``stream_channels_asis`` when ``output_keys`` is ``None``) can pull
+    the anchor blob in per-run event data. Empirically the WebUI's
+    ``stream_mode=["updates"]`` path is clean, so this is transient per-run
+    rather than the persistent per-getState download this PR targets.
+    Filter here first; extend into the stream layer if a client relying on
+    ``values`` / ``events`` reports it.
+    """
+
+    async def aget_state(self, config, *, subgraphs=False):
+        return _strip_private(await super().aget_state(config, subgraphs=subgraphs))
+
+    def get_state(self, config, *, subgraphs=False):
+        return _strip_private(super().get_state(config, subgraphs=subgraphs))
+
+    async def aget_state_history(self, config, **kw):
+        async for snap in super().aget_state_history(config, **kw):
+            yield _strip_private(snap)
+
+    def get_state_history(self, config, **kw):
+        for snap in super().get_state_history(config, **kw):
+            yield _strip_private(snap)
+
+
+# In-place ``__class__`` swap: the subclass adds only methods (no new
+# instance attributes) so the memory layout is identical and the swap is
+# safe. Constructing a fresh ``_EvoFilteredGraph`` via ``.copy()`` would
+# require reproducing the deep-agent build pipeline; the swap avoids that.
+_agent.__class__ = _EvoFilteredGraph
+EvoScientist_agent = _agent
+
+
+def _apply_filter_to_all_registered_graphs() -> None:
+    """Extend the class swap to every graph registered in ``langgraph.json``.
+
+    ``EvoScientist.py:_build_middleware_stack`` installs
+    ``create_code_interpreter_middleware`` unconditionally — it's not gated
+    on the ``for_async_subagent`` flag — so every subagent (sync ``task``
+    dispatch and async ``start_async_task``) carries the QuickJS REPL and
+    can produce ``_quickjs_snapshot_payload`` writes on its own checkpoint
+    namespace.
+
+    Async subagents get their own ``thread_id`` and their ``/threads/{id}/state``
+    endpoint is served by their own compiled graph. Without swapping the
+    class on those graphs, the filter we applied to ``EvoScientist_agent``
+    doesn't reach that endpoint and any real code_interpreter touch inside
+    a subagent leaks the anchor snapshot verbatim.
+
+    Reads the graph registry straight from ``langgraph.json`` so a new
+    subagent added to the config picks up the swap automatically — no
+    hardcoded list to keep in sync.
+
+    Idempotent (skips graphs already swapped) and safe on graphs that don't
+    use the middleware — ``_strip_private`` returns snapshots unchanged when
+    the private field is absent. Best-effort: if the config is unreadable
+    or an entry can't be resolved, the deployment still starts — only the
+    unresolvable subagents remain unfiltered.
+    """
+    import json
+    from importlib import import_module
+    from pathlib import Path
+
+    config_path = Path(__file__).parent / "langgraph.json"
+    try:
+        config = json.loads(config_path.read_text())
+    except (OSError, json.JSONDecodeError):
+        return
+
+    for path in config.get("graphs", {}).values():
+        # Format: "module.dotted.path:attr_name"
+        if ":" not in path:
+            continue
+        module_path, attr = path.rsplit(":", 1)
+        try:
+            module = import_module(module_path)
+        except ImportError:
+            continue
+        graph = getattr(module, attr, None)
+        if isinstance(graph, CompiledStateGraph) and not isinstance(
+            graph, _EvoFilteredGraph
+        ):
+            graph.__class__ = _EvoFilteredGraph
+
+
+_apply_filter_to_all_registered_graphs()
+
 
 __all__ = ["EvoScientist_agent"]

EvoScientist/langgraph_dev/manager.py

@@ -306,14 +306,19 @@ def is_async_subagents_available() -> bool:
 
 def _langgraph_exe() -> str | None:
     """Return the path to the langgraph CLI binary, or None if not found."""
+    import sys
+
+    executable_dir = os.path.dirname(sys.executable)
+    candidate_names = (
+        ["langgraph.exe", "langgraph"] if os.name == "nt" else ["langgraph"]
+    )
+    for candidate_name in candidate_names:
+        candidate = os.path.join(executable_dir, candidate_name)
+        if os.path.isfile(candidate) and os.access(candidate, os.X_OK):
+            return candidate
     found = shutil.which("langgraph")
     if found:
         return found
-    import sys as _sys
-
-    candidate = os.path.join(os.path.dirname(_sys.executable), "langgraph")
-    if os.path.isfile(candidate) and os.access(candidate, os.X_OK):
-        return candidate
     return None
 
 
@@ -709,6 +714,7 @@ def start_langgraph_dev(
     sub_env["EVOSCIENTIST_DEPLOY_MODE"] = "full" if deploy_mode else "stripped"
 
     try:
+        logger.info("Starting langgraph dev with CLI: %s", exe)
         proc = subprocess.Popen(
             [
                 exe,

EvoScientist/llm/context_window.py

@@ -20,9 +20,9 @@ _KNOWN_MODEL_CONTEXT_WINDOWS: dict[str, int] = {
     # Qwen 3.7 closed-source tiers — Max flagship and Plus (1M).
     "qwen3.7-max": 1_000_000,
     "qwen3.7-plus": 1_000_000,
-    # xAI Grok — per-model windows (build-0.1: 256K, 4.3: 1M).
+    # xAI Grok — per-model windows (build-0.1: 256K, 4.5: 500K).
     "grok-build-0.1": 256_000,
-    "grok-4.3": 1_000_000,
+    "grok-4.5": 500_000,
     # Claude Haiku 4.5 — exception to the ``claude-`` family (200K, not 1M).
     "claude-haiku-4-5": 200_000,
     # MiniMax M3 — 1M context (M2.x variants stay at provider default ~204K).
@@ -32,6 +32,8 @@ _KNOWN_MODEL_CONTEXT_WINDOWS: dict[str, int] = {
     # Zhipu GLM-5.2 — 1M context, an exception to the ``glm-5`` family (203K).
     # Matches OpenRouter ``z-ai/glm-5.2`` via split('/')[-1].
     "glm-5.2": 1_000_000,
+    # Tencent Hunyuan HY3 — 262K context (OpenRouter ``tencent/hy3``).
+    "hy3": 262_000,
 }
 
 # Family-level fallbacks: tried only after exact-name lookup misses.
@@ -40,6 +42,8 @@ _KNOWN_MODEL_CONTEXT_WINDOWS: dict[str, int] = {
 _KNOWN_MODEL_FAMILIES: list[tuple[str, int]] = [
     # All Claude — 1M via the ``context-1m-2025-08-07`` beta header.
     ("claude-", 1_000_000),
+    # OpenAI GPT-5.6 family — sol, terra, luna variants
+    ("gpt-5.6", 1_050_000),
     # OpenAI GPT-5.5 family — base, pro, future variants
     ("gpt-5.5", 1_050_000),
     # Google Gemini 3.x family — flash, flash-lite, pro (1.05M). Excludes 2.5.

EvoScientist/llm/errors.py

@@ -0,0 +1,386 @@
+"""Provider-error surface for langgraph SSE frames.
+
+Provides :class:`ProviderStreamError` — a normalized, non-dataclass
+exception raised by ``ErrorNormalizationMiddleware`` in place of the
+provider SDK exception that a chat model call raised. Non-dataclass on
+purpose: since orjson 3.0, dataclass instances are serialized natively
+via their field enumeration, skipping the ``default=`` hook that
+would otherwise build our SSE envelope. Some provider SDKs (openrouter
+today) decorate their exceptions with ``@dataclass``, so their errors
+emerge on the wire as raw dataclass fields — no envelope, no way for
+the WebUI to distinguish quota / auth / rate-limit. Wrapping them in
+a plain ``Exception`` subclass here keeps orjson on the ``default=``
+path, which then calls :meth:`ProviderStreamError.model_dump`
+(upstream ``langgraph_api.serde.default`` checks that hook before its
+``BaseException`` branch) — no serde monkey-patch needed.
+
+Also lives here: the pure-function helpers the middleware uses to
+build the envelope (provider tag from ``ModelRequest.model``, SDK
+field extractors, env-driven API-key redaction). They stay next to
+:class:`ProviderStreamError` because the middleware is their only
+consumer.
+"""
+
+from __future__ import annotations
+
+import os
+import re
+from typing import Any
+
+# ---------------------------------------------------------------------------
+# ProviderStreamError
+# ---------------------------------------------------------------------------
+
+
+class AgentControlError(Exception):
+    """Host-defined terminal control error that must bypass model fallback."""
+
+    non_fallbackable = True
+
+    def __init__(
+        self,
+        code: str,
+        message: str,
+        *,
+        status_code: int = 403,
+        retryable: bool = False,
+    ) -> None:
+        super().__init__(message)
+        self.code = code
+        self.message = message
+        self.status_code = status_code
+        self.retryable = retryable
+
+    def model_dump(self) -> dict[str, Any]:
+        return {
+            "error": type(self).__name__,
+            "code": self.code,
+            "message": self.message,
+            "status_code": self.status_code,
+            "retryable": self.retryable,
+        }
+
+
+class ModelToolProtocolError(AgentControlError):
+    """A completed model response contained an invalid tool-call protocol."""
+
+    def __init__(
+        self,
+        reason: str,
+        *,
+        provider: str | None = None,
+        model: str | None = None,
+        route_key: str | None = None,
+        config_generation: int | None = None,
+        api_mode: str | None = None,
+        endpoint: str | None = None,
+        tool_call_transport: str | None = None,
+        call_id: str | None = None,
+        call_diagnostic: dict[str, Any] | None = None,
+    ) -> None:
+        super().__init__(
+            "MODEL_TOOL_PROTOCOL_INVALID",
+            "The model returned an invalid structured tool call.",
+            status_code=502,
+            retryable=False,
+        )
+        self.reason = reason
+        self.provider = provider
+        self.model = model
+        self.route_key = route_key
+        self.config_generation = config_generation
+        self.api_mode = api_mode
+        self.endpoint = endpoint
+        self.tool_call_transport = tool_call_transport
+        self.call_id = call_id
+        # Internal-only, redacted structure for server logs. Deliberately omitted
+        # from model_dump() so it never becomes part of the public SSE contract.
+        self.call_diagnostic = dict(call_diagnostic or {})
+        self.fallbackable = True
+        self.recoverable = True
+
+    def model_dump(self) -> dict[str, Any]:
+        payload = super().model_dump()
+        payload.update(
+            {
+                "reason": self.reason,
+                "fallbackable": self.fallbackable,
+                "recoverable": self.recoverable,
+            }
+        )
+        for key in (
+            "provider",
+            "model",
+            "route_key",
+            "config_generation",
+            "api_mode",
+            "endpoint",
+            "tool_call_transport",
+            "call_id",
+        ):
+            value = getattr(self, key)
+            if value is not None:
+                payload[key] = value
+        return payload
+
+
+class ProviderStreamError(Exception):
+    """Envelope-shaped wrapper for a provider SDK exception raised
+    inside a chat model call.
+
+    Attributes mirror the SSE envelope one-for-one:
+
+    - ``provider`` — concrete provider tag (``openai`` / ``anthropic``
+      / ``deepseek`` / ``openrouter`` / ``openai_compat`` / …)
+    - ``class_qualname`` — fully qualified name of the underlying
+      exception's class (e.g. ``openrouter.errors.…``)
+    - ``message`` — API-key-redacted ``str(exc)``
+    - ``status_code`` — HTTP status if the SDK exposed one
+    - ``code`` — provider error code (``insufficient_quota``, …)
+    - ``err_type`` — provider error type label (openai's ``.type``)
+    - ``request_id`` — SDK-provided correlation id
+
+    The underlying exception is available via ``__cause__`` (set by
+    ``raise ProviderStreamError(...) from exc`` in the middleware).
+    """
+
+    def __init__(
+        self,
+        provider: str,
+        class_qualname: str,
+        message: str,
+        *,
+        status_code: int | None = None,
+        code: str | None = None,
+        err_type: str | None = None,
+        request_id: str | None = None,
+    ) -> None:
+        super().__init__(message)
+        self.provider = provider
+        self.class_qualname = class_qualname
+        self.message = message
+        self.status_code = status_code
+        self.code = code
+        self.err_type = err_type
+        self.request_id = request_id
+
+    def as_envelope(self) -> dict[str, Any]:
+        """Return the SSE envelope dict — the shape the WebUI consumes."""
+        payload: dict[str, Any] = {
+            "error": self.class_qualname.rsplit(".", 1)[-1],
+            "class": self.class_qualname,
+            "message": self.message,
+            "provider": self.provider,
+        }
+        if self.status_code is not None:
+            payload["status_code"] = self.status_code
+        if self.code is not None:
+            payload["code"] = self.code
+        if self.err_type is not None:
+            payload["type"] = self.err_type
+        if self.request_id:
+            payload["request_id"] = self.request_id
+        return payload
+
+    def model_dump(self) -> dict[str, Any]:
+        """Serialization hook consumed by ``langgraph_api.serde.default``.
+
+        Upstream's dispatch checks ``hasattr(obj, 'model_dump')`` BEFORE
+        the ``isinstance(obj, BaseException)`` branch, so exposing this
+        method lets upstream emit our envelope with no monkey-patch on
+        its ``default`` callable. The name matches Pydantic's
+        convention deliberately — it's the hook upstream is looking
+        for.
+        """
+        return self.as_envelope()
+
+
+# ---------------------------------------------------------------------------
+# API-key redaction — env-driven, prefix-only
+# ---------------------------------------------------------------------------
+#
+# Redaction is built from credentials actually deployed via env vars,
+# not from generic key shapes. Rationale: (a) zero false positives —
+# we only scrub strings we know are secrets, (b) defense-in-depth —
+# the compiled regex holds only the first 8 chars of each key, so a
+# leak of the regex object itself (traceback locals, process dump)
+# can't expose the secret. Suffix-greedy match consumes the rest of
+# the key shape at runtime. The table is rebuilt on every
+# ``_redact_api_keys`` call so credentials loaded after import
+# (typically ``load_dotenv`` in a main entry point) still get
+# scrubbed. ``re.compile`` caches by source string internally, so an
+# unchanged env costs a dict lookup.
+
+_API_KEY_ENV_SUFFIXES = ("_API_KEY", "_TOKEN", "_SECRET")
+_API_KEY_MIN_LEN = 12
+_API_KEY_PREFIX_LEN = 8
+
+
+def _build_env_key_redaction_re() -> re.Pattern[str] | None:
+    prefixes: list[str] = []
+    for k, v in os.environ.items():
+        if not k.endswith(_API_KEY_ENV_SUFFIXES):
+            continue
+        if not isinstance(v, str) or len(v) < _API_KEY_MIN_LEN:
+            continue
+        prefixes.append(re.escape(v[:_API_KEY_PREFIX_LEN]))
+    if not prefixes:
+        return None
+    alternation = "|".join(f"{p}[A-Za-z0-9_+/=.-]*" for p in prefixes)
+    return re.compile(alternation)
+
+
+def _redact_api_keys(message: str) -> str:
+    """Replace any deployed key prefix in *message* with ``<redacted>``.
+
+    Defensive; provider error messages occasionally echo the
+    authorization header back. Rebuilt per call so credentials loaded
+    after import (typical ``load_dotenv`` pattern) are still redacted.
+    """
+    pattern = _build_env_key_redaction_re()
+    if pattern is None:
+        return message
+    return pattern.sub("<redacted>", message)
+
+
+# ---------------------------------------------------------------------------
+# Provider inference from ModelRequest.model
+# ---------------------------------------------------------------------------
+#
+# Host → concrete provider. Hand-maintained snapshot mirroring the
+# routed-provider tables in ``llm/models.py``
+# (``_OPENAI_ROUTED_PROVIDERS`` + ``_ANTHROPIC_ROUTED_PROVIDERS``).
+# Kept here rather than imported from ``models.py`` to keep the
+# import surface of ``errors.py`` minimal — importing ``models.py``
+# would pull in every langchain chat-model client at first
+# middleware access. Consumed by ``_lookup_host_or_compat``; unknown
+# hosts fall back to ``<module>_compat`` so the WebUI knows
+# "openai/anthropic SDK, but not native" instead of getting a
+# misleading concrete tag. Update when a new routed provider is
+# added to ``models.py``.
+#
+# Related sibling: ``_PROVIDER_EXC_MODULE_PREFIXES`` in
+# ``middleware/error_normalization.py`` — the exception-side
+# provider allow-list. Adding a whole new provider SDK (not just a
+# new base_url routed through an existing one) means updating that
+# list too.
+
+_HOST_TO_PROVIDER: dict[str, str] = {
+    "api.openai.com": "openai",
+    "api.anthropic.com": "anthropic",
+    "api.deepseek.com": "deepseek",
+    "api.moonshot.cn": "moonshot",
+    "api.siliconflow.cn": "siliconflow",
+    "open.bigmodel.cn": "zhipu",  # zhipu + zhipu-code share this host
+    "ark.cn-beijing.volces.com": "volcengine",
+    "dashscope.aliyuncs.com": "dashscope",
+    "coding.dashscope.aliyuncs.com": "dashscope",
+    "api.minimaxi.com": "minimax",
+    "api.kimi.com": "kimi",  # kimi-coding shares this host
+    "openrouter.ai": "openrouter",
+}
+
+
+def _provider_from_model(model: Any) -> str | None:
+    """Derive the concrete provider tag from a chat model instance.
+
+    Class-based dispatch for unambiguous providers (``ChatOpenRouter``,
+    ``ChatGoogleGenerativeAI``); ``openai_api_base`` /
+    ``anthropic_api_url`` looked up in ``_HOST_TO_PROVIDER`` for
+    openai/anthropic-shape clients (native + routed). Returns ``None``
+    when the model isn't from a recognized provider SDK — the caller
+    (``ErrorNormalizationMiddleware``) then passes the exception
+    through unchanged.
+    """
+    cls_module = type(model).__module__ or ""
+    if cls_module.startswith("langchain_openrouter"):
+        return "openrouter"
+    if cls_module.startswith("langchain_google_genai"):
+        return "google_genai"
+    if cls_module.startswith("langchain_openai"):
+        return _lookup_host_or_compat(
+            getattr(model, "openai_api_base", None), module_tag="openai"
+        )
+    if cls_module.startswith("langchain_anthropic"):
+        return _lookup_host_or_compat(
+            getattr(model, "anthropic_api_url", None), module_tag="anthropic"
+        )
+    return None
+
+
+def _lookup_host_or_compat(base_url: str | None, module_tag: str) -> str:
+    """Extract host from *base_url* and look up in ``_HOST_TO_PROVIDER``.
+
+    Falls back to *module_tag* when no ``base_url`` is set (native SDK
+    default endpoint) or ``<module_tag>_compat`` for an unrecognized
+    host — the honest "openai SDK shape but unknown upstream" tag.
+    """
+    if not base_url:
+        return module_tag
+    try:
+        from urllib.parse import urlparse
+
+        host = urlparse(base_url).hostname
+    except Exception:
+        host = None
+    if not host:
+        return module_tag
+    return _HOST_TO_PROVIDER.get(host.lower(), f"{module_tag}_compat")
+
+
+# ---------------------------------------------------------------------------
+# SDK-field extractors — populate the envelope's optional fields
+# ---------------------------------------------------------------------------
+
+
+def _extract_status_code(exc: BaseException) -> int | None:
+    """Best-effort HTTP status code from a provider SDK exception.
+
+    Order matters: openai/anthropic store it on ``.status_code``;
+    httpx-wrappers expose it via ``.response.status_code``;
+    ``google.genai.errors.APIError`` (unusually) stores it as an
+    integer ``.code`` — type-disambiguated from openai/anthropic's
+    string ``.code`` (provider error code, surfaced separately).
+    """
+    status_code = getattr(exc, "status_code", None)
+    if isinstance(status_code, int):
+        return status_code
+    response = getattr(exc, "response", None)
+    if response is not None:
+        rsc = getattr(response, "status_code", None)
+        if isinstance(rsc, int):
+            return rsc
+    code = getattr(exc, "code", None)
+    if isinstance(code, int):
+        return code
+    return None
+
+
+def _extract_provider_code(exc: BaseException) -> str | None:
+    """Provider error code (e.g. ``insufficient_quota``,
+    ``invalid_api_key``). Distinct from HTTP status; higher signal for
+    a WebUI toast than the integer alone.
+    """
+    code = getattr(exc, "code", None)
+    if isinstance(code, str) and code:
+        return code
+    return None
+
+
+def _extract_error_type(exc: BaseException) -> str | None:
+    """Provider error type label.
+
+    - openai exposes this as ``.type`` (``rate_limit_error`` etc.)
+    - ``google.genai.errors.APIError`` stores a string label at
+      ``.status`` (``"NOT_FOUND"``, ``"RESOURCE_EXHAUSTED"``, …) — a
+      good fit for the same field.
+
+    ``.type`` takes precedence when both are set.
+    """
+    err_type = getattr(exc, "type", None)
+    if isinstance(err_type, str) and err_type:
+        return err_type
+    status = getattr(exc, "status", None)
+    if isinstance(status, str) and status:
+        return status
+    return None

EvoScientist/llm/models.py

@@ -10,11 +10,19 @@ endpoints) and convenient short names for common models.
 from __future__ import annotations
 
 import os
+import re
+import subprocess
 import warnings
+from functools import lru_cache
 from typing import Any
 
 from langchain.chat_models import init_chat_model
 
+from ..config.settings import (
+    OPENROUTER_DEFAULT_APP_CATEGORIES,
+    OPENROUTER_DEFAULT_APP_TITLE,
+    OPENROUTER_DEFAULT_HTTP_REFERER,
+)
 from .context_window import apply_known_context_window
 from .patches import (
     _is_ccproxy_codex,
@@ -37,6 +45,50 @@ _DEEPSEEK_BASE_URL = "https://api.deepseek.com"
 _MOONSHOT_BASE_URL = "https://api.moonshot.cn/v1"
 _KIMI_CODING_BASE_URL = "https://api.kimi.com/coding/"
 
+# Minimum Codex CLI version advertised when no explicit override is set. Newer
+# installed versions are advertised automatically.
+_CODEX_CLIENT_VERSION_FALLBACK = "0.144.1"
+
+
+@lru_cache(maxsize=1)
+def _installed_codex_client_version() -> str:
+    """Return the installed Codex CLI version, or an empty string."""
+    try:
+        result = subprocess.run(
+            ["codex", "--version"],
+            capture_output=True,
+            text=True,
+            timeout=2,
+            check=False,
+        )
+    except (OSError, subprocess.TimeoutExpired):
+        return ""
+
+    if result.returncode != 0:
+        return ""
+    match = re.search(r"\b(\d+\.\d+\.\d+)\b", result.stdout + result.stderr)
+    return match.group(1) if match else ""
+
+
+def _resolve_codex_client_version() -> str:
+    """Resolve an explicit override or the newer of installed and minimum versions."""
+    override = os.environ.get("EVOSCIENTIST_CODEX_CLIENT_VERSION", "").strip()
+    if override:
+        return override
+
+    installed = _installed_codex_client_version()
+    if installed and tuple(map(int, installed.split("."))) >= tuple(
+        map(int, _CODEX_CLIENT_VERSION_FALLBACK.split("."))
+    ):
+        return installed
+    return _CODEX_CLIENT_VERSION_FALLBACK
+
+
+def _resolve_reasoning_effort(default: str) -> str:
+    """Return the configured reasoning effort or a provider-specific default."""
+    return os.environ.get("EVOSCIENTIST_REASONING_EFFORT", "").strip() or default
+
+
 # Providers routed through the OpenAI provider with a custom base_url.
 # Maps provider name → (base_url or None, env var for API key).
 _OPENAI_ROUTED_PROVIDERS: dict[str, tuple[str | None, str]] = {
@@ -68,6 +120,19 @@ _THINKING_CAPABLE_PROVIDERS: set[str] = {"minimax"}
 _TRUTHY_ENV_VALUES = {"1", "true", "yes", "on"}
 _FALSEY_ENV_VALUES = {"0", "false", "no", "off"}
 
+# OpenRouter app attribution (issue #339). Default values are the single source
+# of truth in config/settings.py (imported above); langchain-openrouter maps
+# app_url → HTTP-Referer, app_title → X-Title, app_categories →
+# X-OpenRouter-Categories. OpenRouter honors at most this many categories per
+# request (server-side limit) and silently ignores the rest, so the sent list is
+# capped to this many below. https://openrouter.ai/docs/app-attribution
+_OPENROUTER_MAX_CATEGORIES_PER_REQUEST = 2
+
+# Legacy/provider-specific options that are not accepted by the installed
+# LangChain chat model constructors. Leaving them at the top level makes
+# LangChain move them into model_kwargs and can later leak them into SDK calls.
+_UNSUPPORTED_CHAT_MODEL_KWARGS = frozenset({"sanitize_openai_sdk_headers"})
+
 # Model registry: list of (short_name, model_id, provider)
 # Allows same short_name across different providers.
 _MODEL_ENTRIES: list[tuple[str, str, str]] = [
@@ -89,6 +154,9 @@ _MODEL_ENTRIES: list[tuple[str, str, str]] = [
     ("claude-sonnet-4-6", "claude-sonnet-4-6", "anthropic"),
     ("claude-haiku-4-5", "claude-haiku-4-5", "anthropic"),
     # OpenAI
+    ("gpt-5.6-sol", "gpt-5.6-sol", "openai"),
+    ("gpt-5.6-terra", "gpt-5.6-terra", "openai"),
+    ("gpt-5.6-luna", "gpt-5.6-luna", "openai"),
     ("gpt-5.5-pro", "gpt-5.5-pro", "openai"),
     ("gpt-5.5", "gpt-5.5", "openai"),
     ("gpt-5.4", "gpt-5.4", "openai"),
@@ -145,6 +213,9 @@ _MODEL_ENTRIES: list[tuple[str, str, str]] = [
     ("claude-opus-4.8-fast", "anthropic/claude-opus-4.8-fast", "openrouter"),
     ("claude-sonnet-5", "anthropic/claude-sonnet-5", "openrouter"),
     ("claude-sonnet-4.6", "anthropic/claude-sonnet-4.6", "openrouter"),
+    ("gpt-5.6-sol", "openai/gpt-5.6-sol", "openrouter"),
+    ("gpt-5.6-terra", "openai/gpt-5.6-terra", "openrouter"),
+    ("gpt-5.6-luna", "openai/gpt-5.6-luna", "openrouter"),
     ("gpt-5.5-pro", "openai/gpt-5.5-pro", "openrouter"),
     ("gpt-5.5", "openai/gpt-5.5", "openrouter"),
     ("gpt-5.4", "openai/gpt-5.4", "openrouter"),
@@ -159,7 +230,8 @@ _MODEL_ENTRIES: list[tuple[str, str, str]] = [
     ("mimo-v2.5-pro", "xiaomi/mimo-v2.5-pro", "openrouter"),
     ("mimo-v2.5", "xiaomi/mimo-v2.5", "openrouter"),
     ("grok-build-0.1", "x-ai/grok-build-0.1", "openrouter"),
-    ("grok-4.3", "x-ai/grok-4.3", "openrouter"),
+    ("grok-4.5", "x-ai/grok-4.5", "openrouter"),
+    ("hy3", "tencent/hy3", "openrouter"),
     ("qwen3.7-max", "qwen/qwen3.7-max", "openrouter"),
     ("qwen3.7-plus", "qwen/qwen3.7-plus", "openrouter"),
     ("qwen3.6-flash", "qwen/qwen3.6-flash", "openrouter"),
@@ -257,6 +329,15 @@ def _env_flag_disabled(name: str) -> bool:
     return value is not None and value.strip().lower() in _FALSEY_ENV_VALUES
 
 
+def _drop_unsupported_chat_model_kwargs(kwargs: dict[str, Any]) -> None:
+    for key in _UNSUPPORTED_CHAT_MODEL_KWARGS:
+        kwargs.pop(key, None)
+    model_kwargs = kwargs.get("model_kwargs")
+    if isinstance(model_kwargs, dict):
+        for key in _UNSUPPORTED_CHAT_MODEL_KWARGS:
+            model_kwargs.pop(key, None)
+
+
 def _supports_openrouter_anthropic_prompt_cache(provider: str, model_id: str) -> bool:
     """Return whether EvoScientist should declare OpenRouter Claude caching."""
     return provider == "openrouter" and model_id.startswith(
@@ -314,8 +395,16 @@ def _apply_auto_config(
     Mutates *kwargs* in place.  Only sets keys that the caller hasn't already
     provided, so explicit user settings are never overridden.
     """
+    disable_reasoning = bool(kwargs.pop("_disable_reasoning", False))
+    disable_thinking = bool(kwargs.pop("_disable_thinking", False))
+    if disable_reasoning:
+        kwargs.pop("reasoning", None)
+        kwargs.pop("include_thoughts", None)
+    if disable_thinking:
+        kwargs.pop("thinking", None)
+
     # Anthropic: extended thinking
-    if provider == "anthropic" and "thinking" not in kwargs:
+    if provider == "anthropic" and not disable_thinking and "thinking" not in kwargs:
         _supports_thinking = original_provider in _THINKING_CAPABLE_PROVIDERS
         # Detect local proxy (e.g. ccproxy): thinking blocks in conversation
         # history cause 422 errors because the proxy doesn't accept 'thinking'
@@ -334,24 +423,31 @@ def _apply_auto_config(
             kwargs["thinking"] = {"type": "enabled", "budget_tokens": 10000}
 
     # OpenAI (native, not third-party routed): reasoning
-    if provider == "openai" and not is_third_party and "reasoning" not in kwargs:
-        if _is_ccproxy_codex():
-            # ccproxy uses Chat Completions which doesn't support reasoning.
-            pass
-        else:
-            _eff = (
-                "xhigh"
-                if ("5.4" in model_id or "5.5" in model_id or "codex" in model_id)
-                else "high"
+    if (
+        provider == "openai"
+        and not is_third_party
+        and not disable_reasoning
+        and "reasoning" not in kwargs
+    ):
+        _default_effort = (
+            "xhigh"
+            if (
+                "5.4" in model_id
+                or "5.5" in model_id
+                or "5.6" in model_id
+                or "codex" in model_id
             )
-            kwargs["reasoning"] = {"effort": _eff, "summary": "auto"}
+            else "high"
+        )
+        _eff = _resolve_reasoning_effort(_default_effort)
+        kwargs["reasoning"] = {"effort": _eff, "summary": "auto"}
 
     # Google GenAI: surface thinking traces
-    if provider == "google-genai":
+    if provider == "google-genai" and not disable_reasoning:
         kwargs.setdefault("include_thoughts", True)
 
     # Ollama: separate reasoning content from response for thinking models
-    if provider == "ollama" and "reasoning" not in kwargs:
+    if provider == "ollama" and not disable_reasoning and "reasoning" not in kwargs:
         kwargs["reasoning"] = True
 
 
@@ -378,7 +474,46 @@ def get_chat_model(
         >>> model = get_chat_model("gpt-4o")  # OpenAI model
         >>> model = get_chat_model("claude-3-opus-20240229", provider="anthropic")  # Full ID
     """
-    model = model or DEFAULT_MODEL
+    skip_runtime_resolver = bool(kwargs.pop("_skip_runtime_model_resolver", False))
+    runtime_provider_name: str | None = None
+    runtime_supports_reasoning: bool | None = None
+    runtime_resolved = None
+    if not skip_runtime_resolver:
+        from EvoScientist.runtime_integrations import resolve_runtime_model
+
+        runtime_resolved = resolve_runtime_model(model, provider)
+
+    if runtime_resolved is not None:
+        resolved_params = dict(getattr(runtime_resolved, "params", {}) or {})
+        extra_body = resolved_params.pop("_extra_body", None)
+        default_headers = resolved_params.pop("_default_headers", None)
+        if extra_body:
+            resolved_params["extra_body"] = extra_body
+        if default_headers:
+            resolved_params["default_headers"] = default_headers
+        resolved_params.update(kwargs)
+        kwargs = resolved_params
+
+        resolved_api_key = str(getattr(runtime_resolved, "api_key", "") or "")
+        resolved_base_url = str(getattr(runtime_resolved, "base_url", "") or "")
+        if resolved_api_key:
+            kwargs.setdefault("api_key", resolved_api_key)
+        if resolved_base_url:
+            kwargs.setdefault("base_url", resolved_base_url.rstrip("/"))
+
+        runtime_provider_name = str(
+            getattr(runtime_resolved, "provider_name", "") or ""
+        )
+        runtime_supports_reasoning = bool(
+            getattr(runtime_resolved, "supports_reasoning", False)
+        )
+        if not runtime_supports_reasoning:
+            kwargs.setdefault("_disable_reasoning", True)
+            kwargs.setdefault("_disable_thinking", True)
+        model = str(runtime_resolved.model_id)
+        provider = str(runtime_resolved.protocol)
+    else:
+        model = model or DEFAULT_MODEL
 
     # Look up short name in registry (provider-aware)
     model_id = None
@@ -413,22 +548,35 @@ def get_chat_model(
     _is_third_party = (
         provider in _OPENAI_ROUTED_PROVIDERS or provider in _ANTHROPIC_ROUTED_PROVIDERS
     )
+    if runtime_provider_name and runtime_provider_name != provider:
+        _is_third_party = True
+    if (
+        runtime_resolved is not None
+        and provider == "openai"
+        and resolved_base_url
+        and "api.openai.com" not in resolved_base_url.lower()
+    ):
+        _is_third_party = True
     _is_openai_proxy = False
-    _original_provider: str | None = None
+    _original_provider: str | None = (
+        runtime_provider_name if runtime_provider_name != provider else None
+    )
     if provider == "anthropic":
         base_url = os.environ.get("ANTHROPIC_BASE_URL", "")
         if base_url:
-            kwargs["base_url"] = base_url
+            kwargs.setdefault("base_url", base_url)
         api_key = os.environ.get("ANTHROPIC_API_KEY", "")
         if api_key:
-            kwargs["api_key"] = api_key
+            kwargs.setdefault("api_key", api_key)
 
     # Native OpenAI base_url override (e.g. ccproxy Codex at localhost:8000/codex/v1)
     elif provider == "openai":
         base_url = os.environ.get("OPENAI_BASE_URL", "")
         if base_url:
-            kwargs["base_url"] = base_url
-            _is_openai_proxy = _is_ccproxy_codex()
+            kwargs.setdefault("base_url", base_url)
+            _is_openai_proxy = _is_ccproxy_codex(
+                kwargs.get("base_url"), kwargs.get("api_key")
+            )
             if _is_openai_proxy:
                 # Use Responses API for ccproxy: bypasses the format chain
                 # converter (Chat→Responses→Chat) which returns 502 on
@@ -441,9 +589,23 @@ def get_chat_model(
                 # for Chat Completions tool_call duplication — not an issue
                 # with the Responses API SSE format.)
                 kwargs.pop("streaming", None)  # remove if set elsewhere
+                # ccproxy forwards client headers upstream and only
+                # gap-fills its own, so the Codex backend sees this
+                # client's identity. Without Codex-CLI-shaped headers it
+                # rejects current models ("The '<model>' model requires
+                # a newer version of Codex").
+                _codex_ver = _resolve_codex_client_version()
+                _headers = kwargs.get("default_headers") or {}
+                kwargs["default_headers"] = _headers
+                _headers.setdefault("originator", "codex_cli_rs")
+                _headers.setdefault("version", _codex_ver)
+                _headers.setdefault(
+                    "User-Agent",
+                    f"codex_cli_rs/{_headers['version']} (EvoScientist)",
+                )
         api_key = os.environ.get("OPENAI_API_KEY", "")
         if api_key:
-            kwargs["api_key"] = api_key
+            kwargs.setdefault("api_key", api_key)
 
     # OpenAI-routed providers → route through OpenAI provider with base_url
     elif provider in _OPENAI_ROUTED_PROVIDERS:
@@ -461,10 +623,10 @@ def get_chat_model(
         else:
             base_url = base_url_default
         if base_url:
-            kwargs["base_url"] = base_url
+            kwargs.setdefault("base_url", base_url)
         api_key = os.environ.get(api_key_env, "")
         if api_key:
-            kwargs["api_key"] = api_key
+            kwargs.setdefault("api_key", api_key)
         # SiliconFlow: disable thinking — LangChain drops reasoning_content
         # from history, causing error 20015 on multi-turn requests.
         if provider == "siliconflow":
@@ -481,15 +643,61 @@ def get_chat_model(
         _is_third_party = True
         api_key = os.environ.get("OPENROUTER_API_KEY", "")
         if api_key:
-            kwargs["api_key"] = api_key
+            kwargs.setdefault("api_key", api_key)
         # Reasoning via `effort` + `summary: "auto"` so a readable reasoning
         # summary is returned for display. OpenAI-Responses also emits encrypted
         # reasoning items (`rs_*` id) that can't be replayed on multi-turn
         # passback (OpenRouter's `/responses` beta is stateless, store=false —
         # "Item with id 'rs_...' not found"); the patch strips them on passback,
         # so enabling `summary` is safe. See langchain-ai/langchain#37777.
-        effort = os.environ.get("EVOSCIENTIST_REASONING_EFFORT", "").strip() or "high"
+        effort = _resolve_reasoning_effort("high")
         kwargs.setdefault("reasoning", {"effort": effort, "summary": "auto"})
+        # App attribution (issue #339): identify EvoScientist to OpenRouter so
+        # usage is credited to the project (app rankings, model app tabs,
+        # analytics) rather than langchain-openrouter's LangChain-branded
+        # defaults. setdefault so an explicit caller kwarg wins; values are
+        # configurable via EVOSCIENTIST_OPENROUTER_* env (fed from the config
+        # file by apply_config_to_env). Applied only here, so no other provider
+        # ever receives these kwargs.
+        kwargs.setdefault(
+            "app_url",
+            os.environ.get("EVOSCIENTIST_OPENROUTER_HTTP_REFERER", "").strip()
+            or OPENROUTER_DEFAULT_HTTP_REFERER,
+        )
+        kwargs.setdefault(
+            "app_title",
+            os.environ.get("EVOSCIENTIST_OPENROUTER_APP_TITLE", "").strip()
+            or OPENROUTER_DEFAULT_APP_TITLE,
+        )
+        # app_categories must be a list[str] (langchain-openrouter joins it into
+        # the X-OpenRouter-Categories header); split the comma-separated config
+        # value and drop blanks so a stray comma/space can't emit an empty one.
+        _app_categories_raw = (
+            os.environ.get("EVOSCIENTIST_OPENROUTER_APP_CATEGORIES", "").strip()
+            or OPENROUTER_DEFAULT_APP_CATEGORIES
+        )
+        _app_categories = [
+            c.strip() for c in _app_categories_raw.split(",") if c.strip()
+        ]
+        # Cap to the per-request limit and warn, so a misconfigured extra is
+        # dropped predictably here (and surfaced to the user) rather than being
+        # silently truncated server-side.
+        _limit = _OPENROUTER_MAX_CATEGORIES_PER_REQUEST
+        if len(_app_categories) > _limit:
+            warnings.warn(
+                f"OpenRouter accepts at most {_limit} app categories per "
+                f"request, so only the first {_limit} are sent: "
+                f"{_app_categories[:_limit]}. Ignoring the rest: "
+                f"{_app_categories[_limit:]}. Set "
+                f"EVOSCIENTIST_OPENROUTER_APP_CATEGORIES (or the "
+                f"openrouter_app_categories config) to at most {_limit} "
+                f"categories to silence this warning.",
+                UserWarning,
+                stacklevel=2,
+            )
+            _app_categories = _app_categories[:_limit]
+        if _app_categories:
+            kwargs.setdefault("app_categories", _app_categories)
         _patch_openrouter_strip_responses_reasoning()
 
     # Anthropic-routed providers → route through Anthropic provider with base_url
@@ -510,10 +718,10 @@ def get_chat_model(
         else:
             base_url = base_url_default
         if base_url:
-            kwargs["base_url"] = base_url
+            kwargs.setdefault("base_url", base_url)
         api_key = os.environ.get(api_key_env, "")
         if api_key:
-            kwargs["api_key"] = api_key
+            kwargs.setdefault("api_key", api_key)
         # Kimi Coding Plan requires claude-code User-Agent header
         if provider == "kimi-coding":
             kwargs.setdefault("default_headers", {})["User-Agent"] = "claude-code/0.1.0"
@@ -522,8 +730,9 @@ def get_chat_model(
     elif provider == "ollama":
         base_url = os.environ.get("OLLAMA_BASE_URL", "")
         if base_url:
-            kwargs["base_url"] = base_url
+            kwargs.setdefault("base_url", base_url)
 
+    _drop_unsupported_chat_model_kwargs(kwargs)
     _apply_auto_config(provider, model_id, _is_third_party, kwargs, _original_provider)
     _apply_openrouter_anthropic_prompt_cache(provider, model_id, kwargs)
 
@@ -540,7 +749,14 @@ def get_chat_model(
         elif _responses_api_setting == "true":
             kwargs["use_responses_api"] = True
 
-    chat_model = init_chat_model(model=model_id, model_provider=provider, **kwargs)
+    anthropic_auth_token = None
+    if provider == "anthropic" and kwargs.get("api_key"):
+        anthropic_auth_token = os.environ.pop("ANTHROPIC_AUTH_TOKEN", None)
+    try:
+        chat_model = init_chat_model(model=model_id, model_provider=provider, **kwargs)
+    finally:
+        if anthropic_auth_token is not None:
+            os.environ["ANTHROPIC_AUTH_TOKEN"] = anthropic_auth_token
 
     # Flatten list content to strings for strict OpenAI-compatible providers
     # (DeepSeek, SiliconFlow, OpenRouter, custom-openai, etc.) and

EvoScientist/llm/patches.py

@@ -25,6 +25,7 @@ Utilities:
 
 from __future__ import annotations
 
+import hashlib
 import os
 from typing import Any
 
@@ -178,15 +179,20 @@ _patch_ccproxy_codex_compat()
 # ---------------------------------------------------------------------------
 # Utility: detect ccproxy's Codex adapter (as opposed to generic localhost).
 # ---------------------------------------------------------------------------
-def _is_ccproxy_codex() -> bool:
+def _is_ccproxy_codex(
+    base_url: str | None = None,
+    api_key: str | None = None,
+) -> bool:
     """Return True if the OpenAI endpoint is ccproxy's Codex adapter.
 
     Checks for the ccproxy-specific markers set by ``setup_codex_env()``
     in ``ccproxy_manager.py``: the sentinel API key and the ``/codex/v1``
     path.  Plain localhost endpoints (vLLM, Ollama, etc.) are not affected.
     """
-    base_url = os.environ.get("OPENAI_BASE_URL", "")
-    api_key = os.environ.get("OPENAI_API_KEY", "")
+    if base_url is None:
+        base_url = os.environ.get("OPENAI_BASE_URL", "")
+    if api_key is None:
+        api_key = os.environ.get("OPENAI_API_KEY", "")
     return (
         ("127.0.0.1" in base_url or "localhost" in base_url)
         and api_key == "ccproxy-oauth"
@@ -267,6 +273,299 @@ def _flatten_message_content(content: Any) -> str | list[Any] | Any:
     return "\n\n".join(parts) if parts else ""
 
 
+def _stable_tool_call_id(message: Any, message_index: int, call_index: int) -> str:
+    seed = ":".join(
+        (
+            str(getattr(message, "id", "") or "message"),
+            str(message_index),
+            str(call_index),
+        )
+    )
+    return "call_" + hashlib.sha256(seed.encode("utf-8")).hexdigest()[:24]
+
+
+def _tool_message_match_index(
+    tool_messages: list[Any],
+    used_indexes: set[int],
+    *,
+    call_id: str,
+    call_name: str,
+) -> int | None:
+    """Find the best unused result for one assistant tool call."""
+
+    def _matches(index: int, *, require_id: bool, require_name: bool) -> bool:
+        if index in used_indexes:
+            return False
+        message = tool_messages[index]
+        result_id = str(getattr(message, "tool_call_id", "") or "")
+        result_name = str(getattr(message, "name", "") or "")
+        if require_id and result_id != call_id:
+            return False
+        if not require_id and result_id:
+            return False
+        return not require_name or not result_name or result_name == call_name
+
+    if call_id:
+        for require_name in (True, False):
+            for index in range(len(tool_messages)):
+                if _matches(index, require_id=True, require_name=require_name):
+                    return index
+        for require_name in (True, False):
+            for index in range(len(tool_messages)):
+                if _matches(index, require_id=False, require_name=require_name):
+                    return index
+        return None
+
+    # A result-side identifier is more authoritative than a generated fallback.
+    for require_name in (True, False):
+        for index, message in enumerate(tool_messages):
+            if index in used_indexes:
+                continue
+            result_id = str(getattr(message, "tool_call_id", "") or "")
+            result_name = str(getattr(message, "name", "") or "")
+            if result_id and (
+                not require_name or not result_name or result_name == call_name
+            ):
+                return index
+    for require_name in (True, False):
+        for index in range(len(tool_messages)):
+            if _matches(index, require_id=False, require_name=require_name):
+                return index
+    return None
+
+
+def _copy_ai_message_with_tool_pairs(
+    message: Any,
+    message_index: int,
+    tool_messages: list[Any],
+) -> tuple[Any | None, list[Any]]:
+    """Return a replay-safe assistant message and its matched tool results."""
+    import copy
+
+    copied = copy.copy(message)
+    additional_kwargs = dict(getattr(message, "additional_kwargs", None) or {})
+    # Parsed tool_calls are canonical. Raw copies can otherwise re-introduce an
+    # invalid call after invalid_tool_calls has been cleared.
+    additional_kwargs.pop("tool_calls", None)
+    copied.additional_kwargs = additional_kwargs
+    copied.invalid_tool_calls = []
+
+    original_calls = list(getattr(message, "tool_calls", None) or [])
+    used_results: set[int] = set()
+    matched_calls: list[dict[str, Any]] = []
+    matched_result_indexes: list[int] = []
+    original_to_matched_call: dict[int, tuple[str, str]] = {}
+
+    for call_index, original_call in enumerate(original_calls):
+        call = dict(original_call)
+        call_id = str(call.get("id") or "")
+        call_name = str(call.get("name") or "").strip()
+        # A missing name is structurally unreplayable. Never infer it from
+        # arguments or retain its paired ToolMessage in provider history.
+        if not call_name:
+            continue
+        call["name"] = call_name
+        result_index = _tool_message_match_index(
+            tool_messages,
+            used_results,
+            call_id=call_id,
+            call_name=call_name,
+        )
+        # A historical client-side function call is only replayable together
+        # with its result. Incomplete calls are discarded instead of asking the
+        # provider to continue a broken tool turn.
+        if result_index is None:
+            continue
+        if not call_id:
+            result_id = str(
+                getattr(tool_messages[result_index], "tool_call_id", "") or ""
+            )
+            call_id = result_id or _stable_tool_call_id(
+                message, message_index, call_index
+            )
+        call["id"] = call_id
+        matched_calls.append(call)
+        matched_result_indexes.append(result_index)
+        original_to_matched_call[call_index] = (call_id, call_name)
+        used_results.add(result_index)
+
+    copied.tool_calls = matched_calls
+    if isinstance(copied.content, list):
+        original_call_index = 0
+        blocks: list[Any] = []
+        for original_block in copied.content:
+            if not isinstance(original_block, dict):
+                blocks.append(original_block)
+                continue
+            block = dict(original_block)
+            if block.get("type") in {"tool_call", "function_call"}:
+                matched_call = original_to_matched_call.get(original_call_index)
+                original_call_index += 1
+                if matched_call is None:
+                    continue
+                call_id, call_name = matched_call
+                # LangChain content blocks use id; the Responses converter later
+                # maps it to call_id.
+                block["id"] = call_id
+                block["name"] = call_name
+                if isinstance(block.get("function"), dict):
+                    block["function"] = {**block["function"], "name": call_name}
+            blocks.append(block)
+        copied.content = blocks
+
+    matched_results: list[Any] = []
+    result_to_call_id = {
+        result_index: matched_calls[index]["id"]
+        for index, result_index in enumerate(matched_result_indexes)
+    }
+    for result_index, result in enumerate(tool_messages):
+        call_id = result_to_call_id.get(result_index)
+        if call_id is None:
+            continue
+        copied_result = copy.copy(result)
+        copied_result.tool_call_id = call_id
+        matched_results.append(copied_result)
+
+    had_tool_protocol = bool(original_calls) or bool(
+        getattr(message, "invalid_tool_calls", None)
+    )
+    if not matched_calls and had_tool_protocol:
+        replayable_content = _flatten_message_content(copied.content)
+        if not replayable_content:
+            return None, matched_results
+
+    return copied, matched_results
+
+
+def _sanitize_openai_tool_history(messages: list[Any]) -> list[Any]:
+    """Copy history while retaining only complete, replayable tool turns."""
+
+    normalized: list[Any] = []
+    index = 0
+    while index < len(messages):
+        message = messages[index]
+        message_type = getattr(message, "type", None)
+        if message_type == "tool":
+            # A tool result without its immediately preceding assistant call is
+            # invalid for both Chat Completions and Responses APIs.
+            index += 1
+            continue
+        if message_type != "ai":
+            normalized.append(message)
+            index += 1
+            continue
+
+        next_index = index + 1
+        tool_messages: list[Any] = []
+        while (
+            next_index < len(messages)
+            and getattr(messages[next_index], "type", None) == "tool"
+        ):
+            tool_messages.append(messages[next_index])
+            next_index += 1
+        copied, matched_results = _copy_ai_message_with_tool_pairs(
+            message,
+            index,
+            tool_messages,
+        )
+        if copied is not None:
+            normalized.append(copied)
+        normalized.extend(matched_results)
+        index = next_index
+
+    return normalized
+
+
+def _ensure_openai_tool_call_ids(messages: list[Any]) -> list[Any]:
+    """Backward-compatible alias for replay-safe tool history normalization."""
+
+    return _sanitize_openai_tool_history(messages)
+
+
+def _has_assistant_tool_protocol(messages: list[Any]) -> bool:
+    """Return whether history contains assistant-side tool protocol state."""
+
+    for message in messages:
+        if getattr(message, "type", None) != "ai":
+            continue
+        if getattr(message, "tool_calls", None) or getattr(
+            message, "invalid_tool_calls", None
+        ):
+            return True
+        additional_kwargs = getattr(message, "additional_kwargs", None) or {}
+        if additional_kwargs.get("tool_calls"):
+            return True
+        content = getattr(message, "content", None)
+        if isinstance(content, list) and any(
+            isinstance(block, dict)
+            and block.get("type") in {"tool_call", "function_call"}
+            for block in content
+        ):
+            return True
+    return False
+
+
+def _validate_openai_tool_history(messages: list[Any]) -> None:
+    """Raise when sanitized history still contains an invalid tool protocol."""
+
+    available_call_ids: set[str] = set()
+    for message in messages:
+        message_type = getattr(message, "type", None)
+        if message_type == "ai":
+            if getattr(message, "invalid_tool_calls", None):
+                raise ValueError("invalid_tool_calls must not be replayed")
+            response_call_ids: set[str] = set()
+            response_calls: dict[str, str] = {}
+            for call in getattr(message, "tool_calls", None) or []:
+                call_name = str(call.get("name") or "").strip()
+                if not call_name:
+                    raise ValueError("assistant tool call is missing a name")
+                call_id = str(call.get("id") or "").strip()
+                if not call_id:
+                    raise ValueError("assistant tool call is missing an id")
+                if call_id in response_call_ids or call_id in available_call_ids:
+                    raise ValueError(
+                        "assistant tool call id is duplicated while outstanding"
+                    )
+                response_call_ids.add(call_id)
+                available_call_ids.add(call_id)
+                response_calls[call_id] = call_name
+            content = getattr(message, "content", None)
+            content_call_ids: set[str] = set()
+            if isinstance(content, list):
+                for block in content:
+                    if not isinstance(block, dict) or block.get("type") not in {
+                        "tool_call",
+                        "function_call",
+                    }:
+                        continue
+                    block_id = str(
+                        block.get("id") or block.get("call_id") or ""
+                    ).strip()
+                    block_name = block.get("name") or block.get("tool_name")
+                    function = block.get("function")
+                    if not block_name and isinstance(function, dict):
+                        block_name = function.get("name")
+                    block_name = str(block_name or "").strip()
+                    if (
+                        not block_id
+                        or block_id in content_call_ids
+                        or response_calls.get(block_id) != block_name
+                    ):
+                        raise ValueError(
+                            "assistant content block does not match parsed tool call"
+                        )
+                    content_call_ids.add(block_id)
+        elif message_type == "tool":
+            call_id = str(getattr(message, "tool_call_id", "") or "")
+            if not call_id or call_id not in available_call_ids:
+                raise ValueError("tool result does not match a prior tool call")
+            available_call_ids.remove(call_id)
+
+    if available_call_ids:
+        raise ValueError("assistant tool call is missing its tool result")
+
+
 def _sanitize_messages(messages: list[Any], hoist_tool_media: bool = True) -> list[Any]:
     """Flatten list content for OpenAI-compatible APIs, preserving media.
 
@@ -282,6 +581,9 @@ def _sanitize_messages(messages: list[Any], hoist_tool_media: bool = True) -> li
 
     from langchain_core.messages import HumanMessage
 
+    sanitize_tool_history = _has_assistant_tool_protocol(messages)
+    if sanitize_tool_history:
+        messages = _sanitize_openai_tool_history(messages)
     out: list[Any] = []
     pending_media: list[Any] = []  # media hoisted out of a run of tool messages
 
@@ -320,6 +622,8 @@ def _sanitize_messages(messages: list[Any], hoist_tool_media: bool = True) -> li
             msg.content = flat
             out.append(msg)
     _flush()  # conversation may end with tool messages
+    if sanitize_tool_history:
+        _validate_openai_tool_history(out)
     return out
 
 
@@ -729,6 +1033,66 @@ def _patch_openai_capture_reasoning_content() -> None:
 _patch_openai_capture_reasoning_content()
 
 
+# ---------------------------------------------------------------------------
+# Patch (module-level): silence langgraph_api's OpenAPI schema-generation
+# warnings for endpoints whose docstrings aren't valid YAML.
+#
+# Upstream ``langgraph_api.utils.SchemaGenerator.get_schema`` calls
+# ``parse_docstring`` (inherited from Starlette's ``BaseSchemaGenerator``)
+# on every registered endpoint. When the docstring is prose with stray
+# ``:`` characters, ``yaml.safe_load`` raises and upstream logs the
+# failure + full traceback at WARNING level. It then falls back to
+# ``{"description": docstring}`` — the endpoint still ends up in the
+# schema with its prose as the description, just without structured
+# ``parameters``/``responses``/``tags`` fields.
+#
+# The fallback path is fine; the warning + traceback is just noise. And
+# it's only triggered for our deploy because mounting any custom Starlette
+# app (``EvoScientist/langgraph_dev/http.py``) makes upstream call
+# ``update_openapi_spec`` at startup — which iterates EVERY route,
+# including upstream's own endpoints whose prose docstrings predate the
+# YAML convention.
+#
+# Fix: wrap ``parse_docstring`` itself and absorb ``yaml.YAMLError`` by
+# returning the same fallback shape upstream's except branch produces.
+# Non-YAML exceptions are deliberately left to propagate — upstream's
+# ``get_schema`` already catches them and logs WARNING + traceback, so
+# unexpected failures remain debuggable. Patching ``parse_docstring`` (a
+# small, stable method) instead of ``get_schema`` (the larger loop body)
+# minimizes our exposure to upstream churn.
+# ---------------------------------------------------------------------------
+_langgraph_schema_silenced_patched = False
+
+
+def _patch_langgraph_schema_generator_silence_warnings() -> None:
+    global _langgraph_schema_silenced_patched
+    if _langgraph_schema_silenced_patched:
+        return
+    try:
+        import langgraph_api.utils as _lgapi_utils
+        import yaml
+
+        _SchemaGenerator = _lgapi_utils.SchemaGenerator
+        _orig_parse_docstring = _SchemaGenerator.parse_docstring
+
+        def _patched_parse_docstring(self: Any, func: Any) -> dict[str, Any]:
+            try:
+                return _orig_parse_docstring(self, func)
+            except yaml.YAMLError:
+                return {"description": getattr(func, "__doc__", None) or ""}
+
+        _SchemaGenerator.parse_docstring = _patched_parse_docstring
+        _langgraph_schema_silenced_patched = True
+    except Exception:
+        # Patches are loader-safe: never crash the import. Silent failure
+        # here just leaves the upstream warnings visible in deploy logs,
+        # which is a benign fallback.
+        pass
+
+
+_patch_langgraph_schema_generator_silence_warnings()
+
+
 # ---------------------------------------------------------------------------
 # Patch (lazy, OpenRouter only): strip OpenAI-Responses encrypted reasoning
 # items from outgoing assistant messages.

EvoScientist/logging_config.py

@@ -0,0 +1,302 @@
+"""Shared logging configuration helpers."""
+
+from __future__ import annotations
+
+import logging
+import os
+import sys
+from datetime import UTC, datetime
+from pathlib import Path
+from typing import Any, TextIO
+
+DEFAULT_LOG_RETENTION_DAYS = 30
+DEFAULT_LOG_FORMAT = "%(asctime)s [%(levelname)s] %(name)s: %(message)s"
+DEFAULT_LOG_DATE_FORMAT = "%Y-%m-%d %H:%M:%S"
+MANAGED_HANDLER_ATTR = "_evoscientist_managed_handler"
+
+
+def resolve_log_level(level: int | str | None, default: int = logging.INFO) -> int:
+    """Resolve a logging level from config or environment input."""
+    if isinstance(level, int):
+        return level
+    raw = str(level or "").strip()
+    if not raw:
+        return default
+    if raw.isdigit():
+        return int(raw)
+    normalized = raw.upper()
+    if normalized == "WARN":
+        normalized = "WARNING"
+    resolved = logging.getLevelNamesMapping().get(normalized)
+    return resolved if isinstance(resolved, int) else default
+
+
+def _mark_managed(handler: logging.Handler, kind: str) -> logging.Handler:
+    setattr(handler, MANAGED_HANDLER_ATTR, kind)
+    return handler
+
+
+def _managed_kind(handler: logging.Handler) -> str | None:
+    kind = getattr(handler, MANAGED_HANDLER_ATTR, None)
+    return kind if isinstance(kind, str) else None
+
+
+def remove_managed_handlers(
+    logger: logging.Logger | None = None,
+    *,
+    kinds: set[str] | None = None,
+) -> None:
+    """Remove handlers installed by this module without touching external ones."""
+    target = logger or logging.getLogger()
+    for handler in target.handlers[:]:
+        kind = _managed_kind(handler)
+        if kind and (kinds is None or kind in kinds):
+            target.removeHandler(handler)
+            handler.close()
+
+
+def _standard_formatter() -> logging.Formatter:
+    return logging.Formatter(DEFAULT_LOG_FORMAT, datefmt=DEFAULT_LOG_DATE_FORMAT)
+
+
+class DailyLogFileHandler(logging.FileHandler):
+    """File handler that writes the active log to a date-based filename."""
+
+    def __init__(
+        self,
+        log_dir: str | Path,
+        *,
+        prefix: str = "evoscientist",
+        retention_days: int = DEFAULT_LOG_RETENTION_DAYS,
+        encoding: str = "utf-8",
+        utc: bool = False,
+    ) -> None:
+        self.log_dir = Path(log_dir).expanduser()
+        self.prefix = prefix
+        self.retention_days = max(1, retention_days)
+        self.utc = utc
+        self.log_dir.mkdir(parents=True, exist_ok=True)
+        super().__init__(self._dated_log_path(), encoding=encoding, delay=True)
+
+    @property
+    def active_log_path(self) -> Path:
+        """Return the active log path for the current date."""
+        return self._dated_log_path()
+
+    def _dated_log_path(self) -> Path:
+        now = datetime.now(UTC if self.utc else None)
+        return self.log_dir / f"{self.prefix}-{now:%Y-%m-%d}.log"
+
+    def emit(self, record: logging.LogRecord) -> None:
+        try:
+            expected = str(self.active_log_path)
+            if self.baseFilename != expected:
+                if self.stream:
+                    self.stream.close()
+                    self.stream = None
+                self.baseFilename = expected
+            self._delete_expired_logs()
+            super().emit(record)
+        except OSError:
+            self.handleError(record)
+
+    def getFilesToDelete(self) -> list[str]:
+        candidates = sorted(self.log_dir.glob(f"{self.prefix}-????-??-??.log"))
+        if len(candidates) <= self.retention_days:
+            return []
+        return [str(path) for path in candidates[: -self.retention_days]]
+
+    def _delete_expired_logs(self) -> None:
+        for path in self.getFilesToDelete():
+            try:
+                os.remove(path)
+            except OSError:
+                pass
+
+
+def default_log_dir() -> Path:
+    """Return the default runtime log directory."""
+    env_dir = os.environ.get("EVOSCIENTIST_LOG_DIR")
+    if env_dir:
+        return Path(env_dir).expanduser()
+
+    from EvoScientist.paths import DATA_DIR
+
+    return DATA_DIR / "logs"
+
+
+def configure_daily_file_logging(
+    logger: logging.Logger | None = None,
+    *,
+    log_dir: str | Path | None = None,
+    prefix: str = "evoscientist",
+    level: int | str = logging.INFO,
+    retention_days: int = DEFAULT_LOG_RETENTION_DAYS,
+) -> DailyLogFileHandler:
+    """Attach a daily file handler, replacing older matching handlers."""
+    target = logger or logging.getLogger()
+    resolved_level = resolve_log_level(level, default=logging.INFO)
+    retention_days = max(1, int(retention_days))
+    resolved_dir = Path(log_dir).expanduser() if log_dir else default_log_dir()
+
+    for handler in target.handlers[:]:
+        if (
+            isinstance(handler, DailyLogFileHandler)
+            and handler.prefix == prefix
+            and handler.log_dir == resolved_dir
+        ):
+            target.removeHandler(handler)
+            handler.close()
+
+    handler = DailyLogFileHandler(
+        resolved_dir,
+        prefix=prefix,
+        retention_days=retention_days,
+    )
+    _mark_managed(handler, "file")
+    handler.setLevel(resolved_level)
+    handler.setFormatter(_standard_formatter())
+    target.addHandler(handler)
+    if target.level == logging.NOTSET or target.level > resolved_level:
+        target.setLevel(resolved_level)
+    return handler
+
+
+def configure_console_logging(
+    logger: logging.Logger | None = None,
+    *,
+    level: int | str | None = logging.INFO,
+    stream: TextIO | None = None,
+    replace: bool = True,
+) -> logging.StreamHandler:
+    """Attach a standard console handler for non-interactive entry points."""
+    target = logger or logging.getLogger()
+    resolved_level = resolve_log_level(level, default=logging.INFO)
+    if replace:
+        remove_managed_handlers(target, kinds={"console", "rich"})
+
+    handler = logging.StreamHandler(stream or sys.stderr)
+    _mark_managed(handler, "console")
+    handler.setLevel(resolved_level)
+    handler.setFormatter(_standard_formatter())
+    target.addHandler(handler)
+    target.setLevel(resolved_level)
+    return handler
+
+
+def configure_rich_console_logging(
+    logger: logging.Logger | None = None,
+    *,
+    level: int | str | None = logging.INFO,
+    console: Any = None,
+    replace: bool = True,
+    dim_warnings: bool = False,
+    show_time: bool | None = None,
+    show_path: bool | None = None,
+    show_level: bool | None = None,
+) -> logging.Handler:
+    """Attach a Rich console handler for interactive CLI output."""
+    from rich.logging import RichHandler
+    from rich.markup import escape
+
+    target = logger or logging.getLogger()
+    resolved_level = resolve_log_level(level, default=logging.INFO)
+    verbose = resolved_level <= logging.DEBUG
+    if replace:
+        remove_managed_handlers(target, kinds={"console", "rich"})
+
+    class DimWarningHandler(RichHandler):
+        def emit(self, record: logging.LogRecord) -> None:
+            if dim_warnings and record.levelno == logging.WARNING and console is not None:
+                console.print(
+                    "[dim yellow]\u26a0\ufe0f  Warning:[/dim yellow] "
+                    f"[dim]{escape(record.getMessage())}[/dim]"
+                )
+                return
+            super().emit(record)
+
+    handler = DimWarningHandler(
+        console=console,
+        show_time=verbose if show_time is None else show_time,
+        show_path=verbose if show_path is None else show_path,
+        show_level=verbose if show_level is None else show_level,
+    )
+    _mark_managed(handler, "rich")
+    handler.setLevel(resolved_level)
+    target.addHandler(handler)
+    target.setLevel(resolved_level)
+    return handler
+
+
+def configure_logging(
+    logger: logging.Logger | None = None,
+    *,
+    level: int | str | None = logging.INFO,
+    log_dir: str | Path | None = None,
+    retention_days: int = DEFAULT_LOG_RETENTION_DAYS,
+    prefix: str = "evoscientist",
+    console: bool = True,
+    file: bool = True,
+    replace_managed: bool = True,
+) -> list[logging.Handler]:
+    """Configure standard EvoScientist console and daily file logging."""
+    target = logger or logging.getLogger()
+    resolved_level = resolve_log_level(level, default=logging.INFO)
+    if replace_managed:
+        remove_managed_handlers(target, kinds={"console", "rich", "file"})
+
+    handlers: list[logging.Handler] = []
+    if console:
+        handlers.append(
+            configure_console_logging(target, level=resolved_level, replace=False)
+        )
+    if file:
+        handlers.append(
+            configure_daily_file_logging(
+                target,
+                log_dir=log_dir,
+                prefix=prefix,
+                level=resolved_level,
+                retention_days=retention_days,
+            )
+        )
+    target.setLevel(resolved_level)
+    return handlers
+
+
+def configure_logging_from_settings(
+    logger: logging.Logger | None = None,
+    *,
+    default_level: int = logging.INFO,
+    prefix: str = "evoscientist",
+    console: bool = True,
+    file: bool = True,
+) -> list[logging.Handler]:
+    """Configure logging from EvoScientist settings and environment overrides."""
+    level: int | str | None = os.environ.get("EVOSCIENTIST_LOG_LEVEL")
+    log_dir: str | Path | None = os.environ.get("EVOSCIENTIST_LOG_DIR") or None
+    retention_days = int(
+        os.environ.get("EVOSCIENTIST_LOG_RETENTION_DAYS", DEFAULT_LOG_RETENTION_DAYS)
+    )
+
+    try:
+        from EvoScientist.config import get_effective_config
+
+        cfg = get_effective_config()
+        level = level or getattr(cfg, "log_level", None)
+        log_dir = log_dir or getattr(cfg, "log_dir", None) or None
+        retention_days = int(
+            getattr(cfg, "log_retention_days", DEFAULT_LOG_RETENTION_DAYS)
+        )
+    except Exception:
+        level = level or default_level
+
+    return configure_logging(
+        logger,
+        level=resolve_log_level(level, default=default_level),
+        log_dir=log_dir,
+        retention_days=retention_days,
+        prefix=prefix,
+        console=console,
+        file=file,
+    )

EvoScientist/mcp/__init__.py

@@ -10,6 +10,7 @@ from .client import (
     build_mcp_add_kwargs,
     build_mcp_edit_fields,
     edit_mcp_server,
+    get_mcp_server_errors,
     load_mcp_config,
     load_mcp_tools,
     parse_mcp_add_args,
@@ -38,6 +39,7 @@ __all__ = [
     "find_server_by_name",
     "get_all_tags",
     "get_installed_names",
+    "get_mcp_server_errors",
     "install_mcp_server",
     "install_mcp_servers",
     "load_mcp_config",

EvoScientist/mcp/client.py

@@ -114,6 +114,10 @@ _URL_TRANSPORTS = {"http", "streamable_http", "sse", "websocket"}
 # still parallelizing the common 3–7 server case to completion.
 _MAX_CONCURRENT_CONNECTIONS = 8
 
+# Last connection error per configured server. This is process-local runtime
+# diagnostics for the Web/CLI status surfaces, not persisted configuration.
+_MCP_SERVER_ERRORS: dict[str, str] = {}
+
 # Env vars forwarded to stdio MCP subprocesses on top of the MCP SDK's
 # minimal default set (HOME/PATH/USER/…). Without this, servers behind
 # a proxy or with a custom CA bundle silently fail with long timeouts.
@@ -764,6 +768,9 @@ async def _load_tools(
     if not connections:
         return {}
 
+    for stale_name in set(_MCP_SERVER_ERRORS) - set(connections):
+        _MCP_SERVER_ERRORS.pop(stale_name, None)
+
     client = MultiServerMCPClient(connections)  # type: ignore[invalid-argument-type]
 
     def _report(event: str, name: str, detail: str = "") -> None:
@@ -787,10 +794,13 @@ async def _load_tools(
             _report("start", name)
             try:
                 tools = await client.get_tools(server_name=name)
+                _MCP_SERVER_ERRORS.pop(name, None)
                 logger.info("MCP server %r: loaded %d tool(s)", name, len(tools))
                 _report("success", name, str(len(tools)))
                 return name, tools
             except Exception as exc:
+                detail = str(exc) or type(exc).__name__
+                _MCP_SERVER_ERRORS[name] = detail
                 # When the caller wired up ``on_progress`` they own the
                 # user-facing display; downgrade the logger so we don't
                 # double-print.
@@ -798,7 +808,7 @@ async def _load_tools(
                     logger.warning("MCP server %r: failed to load tools: %s", name, exc)
                 else:
                     logger.debug("MCP server %r: failed to load tools: %s", name, exc)
-                _report("error", name, str(exc))
+                _report("error", name, detail)
                 return name, []
 
     # ``return_exceptions=False`` is fine because ``_fetch`` already
@@ -807,6 +817,11 @@ async def _load_tools(
     return dict(results)
 
 
+def get_mcp_server_errors() -> dict[str, str]:
+    """Return a snapshot of the most recent per-server connection errors."""
+    return dict(_MCP_SERVER_ERRORS)
+
+
 async def aload_mcp_tools(
     config: dict[str, Any] | None = None,
     *,

EvoScientist/memory/agents/memory_worker.py

@@ -428,6 +428,7 @@ def _memory_worker_middleware(
     enable_observation_memory: bool = True,
 ):
     """Build middleware for memory workers, excluding task execution tools."""
+    from ...middleware.error_normalization import ErrorNormalizationMiddleware
     from ...middleware.memory import create_memory_middleware
 
     memory_controls = MemoryControls(
@@ -439,18 +440,23 @@ def _memory_worker_middleware(
     enable_observation_tool = memory_controls.observation_tool_enabled(
         _memory_worker_observation_target(source_type)
     )
-    return memory_agent_middleware(
-        create_memory_middleware(
-            str(memory_dir),
-            workspace_dir=workspace_dir,
-            source_type=source_type,
-            source_agent=_memory_worker_agent_name(source_type),
-            enable_profile_memory=enable_profile_memory,
-            enable_observation_memory=enable_observation_memory,
-            enable_observation_tool=enable_observation_tool,
+    return [
+        # Outermost — normalize provider-SDK exceptions from the
+        # auxiliary model call before any inner middleware sees them.
+        ErrorNormalizationMiddleware(),
+        *memory_agent_middleware(
+            create_memory_middleware(
+                str(memory_dir),
+                workspace_dir=workspace_dir,
+                source_type=source_type,
+                source_agent=_memory_worker_agent_name(source_type),
+                enable_profile_memory=enable_profile_memory,
+                enable_observation_memory=enable_observation_memory,
+                enable_observation_tool=enable_observation_tool,
+            ),
+            excluded_tools=_MEMORY_WORKER_EXCLUDED_TOOLS,
         ),
-        excluded_tools=_MEMORY_WORKER_EXCLUDED_TOOLS,
-    )
+    ]
 
 
 def _build_memory_worker_agent(

EvoScientist/memory/agents/observation_linker.py

@@ -71,6 +71,8 @@ def build_observation_linker_graph(
     workspace_dir: str | Path | None = None,
 ) -> CompiledStateGraph:
     """Build the registered LangGraph observation linker."""
+    from ...middleware.error_normalization import ErrorNormalizationMiddleware
+
     agent_paths = resolve_memory_agent_paths(
         memory_dir=memory_dir,
         workspace_dir=workspace_dir,
@@ -85,5 +87,7 @@ def build_observation_linker_graph(
         tools=tools,
         memory_dir=agent_paths.memory_dir,
         workspace_dir=agent_paths.workspace_dir,
-        middleware=memory_agent_middleware(),
+        # Outermost — normalize provider-SDK exceptions from the
+        # auxiliary model call before any inner middleware sees them.
+        middleware=[ErrorNormalizationMiddleware(), *memory_agent_middleware()],
     )

EvoScientist/middleware/__init__.py

@@ -18,6 +18,7 @@ from .context_editing import (
     create_context_editing_middleware,
 )
 from .context_overflow import ContextOverflowMapperMiddleware
+from .error_normalization import ErrorNormalizationMiddleware
 from .memory import (
     EvoMemoryMiddleware,
     create_memory_middleware,
@@ -28,29 +29,42 @@ from .memory_lifecycle import (
     default_memory_scheduler,
 )
 from .model_fallback import ModelFallbackMiddleware, load_fallback_chain
+from .repetitive_tool_guard import (
+    DEFAULT_MAX_CONSECUTIVE_TOOL_ERRORS,
+    DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD,
+    RepetitiveToolCallGuardMiddleware,
+    collapse_repetitive_tool_rounds,
+)
 from .runtime_context import RuntimeContextMiddleware, create_runtime_context_middleware
 from .scheduler import (
     SchedulerMiddleware,
     create_scheduler_middleware,
 )
 from .tool_error_handler import ToolErrorHandlerMiddleware
+from .tool_protocol_guard import ToolProtocolGuardMiddleware
 from .tool_selector import create_tool_selector_middleware
 from .utils import disable_thinking
 
 __all__ = [
+    "DEFAULT_MAX_CONSECUTIVE_TOOL_ERRORS",
+    "DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD",
     "AskUserMiddleware",
     "AskUserRequest",
     "AskUserWidgetResult",
     "Choice",
     "ConfigurableModelMiddleware",
     "ContextOverflowMapperMiddleware",
+    "ErrorNormalizationMiddleware",
     "EvoMemoryLifecycleMiddleware",
     "EvoMemoryMiddleware",
     "ModelFallbackMiddleware",
     "Question",
+    "RepetitiveToolCallGuardMiddleware",
     "RuntimeContextMiddleware",
     "SchedulerMiddleware",
     "ToolErrorHandlerMiddleware",
+    "ToolProtocolGuardMiddleware",
+    "collapse_repetitive_tool_rounds",
     "compute_context_editing_trigger",
     "create_code_interpreter_middleware",
     "create_context_editing_middleware",

EvoScientist/middleware/code_interpreter.py

@@ -45,7 +45,21 @@ _MEMORY_FIRST_INTERPRETER_PROMPT = (
 
 
 class EvoCodeInterpreterMiddleware(CodeInterpreterMiddleware):
-    """Code interpreter middleware with EvoScientist's memory preflight hint."""
+    """Code interpreter middleware with EvoScientist's memory preflight hint.
+
+    ``after_agent`` / ``aafter_agent`` are intentionally NOT overridden. An
+    earlier "conditional snapshot" gate that skipped ``after_agent`` on turns
+    where ``code_interpreter`` wasn't called saved ~50 ms/turn of
+    ``create_snapshot()`` work, but also skipped the slot eviction upstream
+    performs in the same hook (``finally: self._registry.evict(thread_id)``
+    in ``langchain_quickjs.middleware.CodeInterpreterMiddleware.after_agent``).
+    ``before_agent`` restores the REPL on every turn that follows a touched
+    one via ``self._registry.get(thread_id)`` (get-or-create), so skipping
+    eviction leaked one ``ThreadWorker`` + QuickJS Runtime per persistent
+    ``thread_id`` that ever went touched → quiet. The regression test
+    ``test_after_agent_evicts_slot_on_untouched_turn`` guards against
+    reintroducing the gate.
+    """
 
     def _prepare_for_call(self, request: ModelRequest) -> str:
         return super()._prepare_for_call(request) + _MEMORY_FIRST_INTERPRETER_PROMPT

EvoScientist/middleware/error_normalization.py

@@ -0,0 +1,240 @@
+"""ErrorNormalizationMiddleware — catch provider-SDK exceptions at the
+model boundary and re-raise as a normalized non-dataclass wrapper.
+
+Some provider SDKs (openrouter.errors.* today) decorate their exception
+classes with ``@dataclass``. When langgraph_api emits an SSE error
+frame via ``json_dumpb`` → ``orjson.dumps(obj, default=default,
+option=OPT_SERIALIZE_DATACLASS)``, orjson's dataclass fast-path
+enumerates the fields directly and skips the ``default=`` hook that
+builds our envelope. The wire payload comes out as
+``{"message": …, "status_code": …, "body": …, "headers": null,
+"raw_response": null, "data": {…}}`` with no ``error`` / ``class`` /
+``provider`` envelope and no way for the WebUI to distinguish quota /
+auth / rate-limit / model-not-found.
+
+This middleware sits at the model-call boundary. It catches
+``BaseException`` from ``handler()``, and if ``request.model`` is a
+recognized provider SDK client, wraps the exception in a
+:class:`~EvoScientist.llm.errors.ProviderStreamError` (a plain
+``Exception`` subclass, not a dataclass). The wrapper carries the SSE
+envelope pre-baked on its instance attributes.
+
+Contract: the wrap decision is based on the **model**, not the
+exception, after platform and graph control signals have been excluded.
+Provider SDK exceptions, httpx errors, langchain-wrapper failures, and
+even builtins like ``RuntimeError`` get wrapped for a recognized model.
+At the middleware boundary we can tell which provider was in use, but
+not the exception's precise origin; a uniform envelope is more useful
+to the WebUI than gambling on the exception class. If the model isn't
+from a recognized provider, or the request carries no ``.model``, the
+exception re-raises unchanged and upstream's whitelist / catch-all
+behavior takes over.
+"""
+
+from __future__ import annotations
+
+from collections.abc import Awaitable, Callable
+from typing import TYPE_CHECKING
+
+from langchain.agents.middleware.types import (
+    AgentMiddleware,
+    ModelRequest,
+    ModelResponse,
+)
+
+if TYPE_CHECKING:
+    from ..llm.errors import ProviderStreamError
+
+
+def _should_pass_through(exc: BaseException) -> bool:
+    """True if *exc* is a LangGraph-level signal that must propagate
+    untouched — either a control-flow signal or a structural error
+    that isn't a provider failure.
+
+    Covers everything in ``langgraph.errors.*``:
+
+    - **Control flow** (breaking these would corrupt the interrupt /
+      resume protocol): ``GraphBubbleUp`` and its subclasses
+      ``GraphInterrupt``, ``NodeInterrupt``, ``ParentCommand``,
+      ``GraphDrained``.
+    - **Structural** (wrapping would mis-attribute a graph-level
+      issue as a provider failure): ``InvalidUpdateError``,
+      ``EmptyInputError``, ``EmptyChannelError``, ``TaskNotFound``,
+      ``GraphRecursionError``, ``NodeCancelledError``,
+      ``NodeTimeoutError``.
+
+    Symmetric with upstream ``langgraph_api.serde.default``'s
+    whitelist, which also exposes these classes' ``str(exc)`` untouched
+    rather than swallowing them behind a provider envelope.
+
+    ``KeyboardInterrupt``, ``SystemExit``, and ``asyncio.CancelledError``
+    are handled implicitly by catching ``Exception`` — they inherit
+    from ``BaseException``.
+    """
+    return (type(exc).__module__ or "").startswith("langgraph.errors")
+
+
+# Module prefixes for provider SDK exceptions. Consumed by
+# ``_is_provider_error`` to decide whether an exception raised inside
+# a model call should surface as a provider incident or gracefully
+# degrade (used by ``_ConditionalToolSelectorMiddleware``).
+#
+# Related sibling: ``_HOST_TO_PROVIDER`` in ``llm/errors.py`` — the
+# host-side allow-list. Adding a whole new provider SDK means updating
+# both; adding a new routed provider (new base_url through an existing
+# SDK) only touches ``_HOST_TO_PROVIDER``.
+_PROVIDER_EXC_MODULE_PREFIXES: tuple[str, ...] = (
+    "openai",
+    "anthropic",
+    "google.genai",
+    "google.api_core",
+    "openrouter",
+    "langchain_openai",
+    "langchain_anthropic",
+    "langchain_google_genai",
+    "langchain_openrouter",
+    "httpx",
+)
+
+
+def _is_provider_error(exc: BaseException) -> bool:
+    """True if *exc* looks like it originated inside a provider SDK
+    (openai, anthropic, google.genai, openrouter, httpx, or their
+    langchain wrappers), as opposed to a shape / config error (structured
+    output not supported, malformed schema, missing tool, …).
+
+    Used by callers that need to decide whether an exception from the
+    model call is worth surfacing to the user (provider errors) or
+    can be silently degraded around (shape errors). Cheap alternative
+    to inspecting ``status_code`` / ``request`` because some provider
+    errors — connection errors, timeouts — don't carry those attributes.
+    """
+    module = type(exc).__module__ or ""
+    return any(module.startswith(p) for p in _PROVIDER_EXC_MODULE_PREFIXES)
+
+
+def _normalize(request: ModelRequest, exc: BaseException) -> ProviderStreamError | None:
+    """Return a :class:`ProviderStreamError` wrapping *exc* if the model
+    on *request* comes from a recognized provider SDK, or ``None`` if
+    the caller should re-raise *exc* unchanged.
+
+    Provider is read from ``request.model`` — the definitive config
+    the exception was raised under, not inferred from the exception
+    class / URL. Status / code / redaction still come from the raised
+    exception because those fields are populated by the SDK at raise
+    time.
+
+    Returns ``None`` (caller re-raises unchanged) for:
+
+    - Already-normalized wrappers (would double-attribute).
+    - LangGraph control-flow / structural errors — see
+      ``_should_pass_through``. This gate lives here so every caller
+      of ``_normalize`` (not just the wrap sites of this middleware)
+      gets the protection automatically. Notably
+      ``ModelFallbackMiddleware`` also calls ``_normalize`` at the
+      raise point of its fallback chain.
+    - ``ContextOverflowError`` — a cross-layer control signal that
+      deepagents' ``SummarizationMiddleware`` catches by type from
+      **outside** the user middleware stack to compress history and
+      retry. Wrapping it here would change the type and break that
+      self-healing fallback.
+    - ``AgentControlError`` — a platform-owned typed decision. Gateway route
+      fallback and canonical error mapping depend on its concrete type and
+      structured fields, so it must never become a provider incident.
+    - Models we don't recognize as a provider SDK.
+    """
+    from langchain_core.exceptions import ContextOverflowError
+
+    from ..llm.errors import (
+        AgentControlError,
+        ProviderStreamError,
+        _extract_error_type,
+        _extract_provider_code,
+        _extract_status_code,
+        _provider_from_model,
+        _redact_api_keys,
+    )
+
+    # Already normalized (e.g. by ModelFallbackMiddleware wrapping against
+    # the actual failing model rather than the original request's model).
+    # Pass through — re-wrapping would double-attribute.
+    if isinstance(exc, ProviderStreamError):
+        return None
+
+    # Platform control errors are raised by inner middleware after the provider
+    # response has already been interpreted. Wrapping them would erase routing,
+    # retry and recovery semantics such as ModelToolProtocolError.fallbackable.
+    if isinstance(exc, AgentControlError):
+        return None
+
+    # LangGraph control-flow / structural signals must propagate
+    # untouched, regardless of which caller invoked us.
+    if _should_pass_through(exc):
+        return None
+
+    # SummarizationMiddleware sits outside our stack and catches this
+    # by exact type to trigger reactive history compression + retry.
+    if isinstance(exc, ContextOverflowError):
+        return None
+
+    provider = _provider_from_model(getattr(request, "model", None))
+    if provider is None:
+        return None
+    cls = type(exc)
+    mod = cls.__module__ or ""
+    class_qualname = f"{mod}.{cls.__qualname__}" if mod else cls.__qualname__
+
+    request_id_attr = getattr(exc, "request_id", None)
+    request_id = (
+        request_id_attr
+        if isinstance(request_id_attr, str) and request_id_attr
+        else None
+    )
+
+    return ProviderStreamError(
+        provider=provider,
+        class_qualname=class_qualname,
+        message=_redact_api_keys(str(exc)),
+        status_code=_extract_status_code(exc),
+        code=_extract_provider_code(exc),
+        err_type=_extract_error_type(exc),
+        request_id=request_id,
+    )
+
+
+class ErrorNormalizationMiddleware(AgentMiddleware):
+    """Wrap the model call in try/except and normalize provider SDK
+    exceptions into a non-dataclass envelope wrapper.
+
+    Place this middleware **outermost** in the chain (first in the
+    middleware list) so it catches exceptions raised by inner
+    middlewares as well as the model handler itself.
+    """
+
+    name = "error_normalization"
+
+    def wrap_model_call(
+        self,
+        request: ModelRequest,
+        handler: Callable[[ModelRequest], ModelResponse],
+    ) -> ModelResponse:
+        try:
+            return handler(request)
+        except Exception as exc:
+            normalized = _normalize(request, exc)
+            if normalized is None:
+                raise
+            raise normalized from exc
+
+    async def awrap_model_call(
+        self,
+        request: ModelRequest,
+        handler: Callable[[ModelRequest], Awaitable[ModelResponse]],
+    ) -> ModelResponse:
+        try:
+            return await handler(request)
+        except Exception as exc:
+            normalized = _normalize(request, exc)
+            if normalized is None:
+                raise
+            raise normalized from exc

EvoScientist/middleware/model_fallback.py

@@ -48,6 +48,8 @@ _MALFORMED_REQUEST_PATTERNS: list[str] = [
     "invalid_request_error",
     "invalid request",
     "malformed",
+    "repetitive tool calls",
+    "identical name and arguments",
 ]
 """Substrings that identify a malformed request (client-side bug)."""
 
@@ -215,6 +217,9 @@ def _is_non_fallbackable(exc: Exception) -> str | None:
     """
     from langchain_core.exceptions import ContextOverflowError
 
+    if getattr(exc, "non_fallbackable", False):
+        return f"platform control error: {getattr(exc, 'code', type(exc).__name__)}"
+
     if isinstance(exc, ContextOverflowError):
         return "context length exceeded"
 
@@ -263,7 +268,15 @@ async def _try_fallbacks(
         "Primary model failed: %s: %s", type(primary_exc).__name__, primary_exc
     )
 
+    # Track the request whose model actually raised ``last_exc`` so we
+    # can attribute the exception to the failing model, not the
+    # original ``request.model``. Without this, a fallback chain
+    # ``deepseek → moonshot`` where moonshot exhausts its quota would
+    # surface as ``provider: deepseek`` — the model the user never
+    # actually saw fail.
     last_exc = primary_exc
+    last_failing_request = request
+
     for model_name, provider in get_fallback_chain():
         _emit(
             f"  -> Falling back to {model_name} ({provider}) "
@@ -288,8 +301,9 @@ async def _try_fallbacks(
                     f"-- aborting fallback chain",
                     style="red",
                 )
-                raise
+                _raise_normalized(fb_request, fb_exc)
             last_exc = fb_exc
+            last_failing_request = fb_request
             _emit(
                 f"  x {model_name} also failed: {type(fb_exc).__name__}: {fb_exc}",
                 style="red",
@@ -303,7 +317,24 @@ async def _try_fallbacks(
             )
 
     _emit("  All fallbacks exhausted -- re-raising last error", style="red")
-    raise last_exc
+    _raise_normalized(last_failing_request, last_exc)
+
+
+def _raise_normalized(request: ModelRequest, exc: Exception) -> None:
+    """Wrap *exc* in a ``ProviderStreamError`` attributed to
+    ``request.model`` and raise, so the outer chain sees the failure
+    tagged with the model that actually raised.
+
+    Falls back to a plain ``raise`` when the model isn't from a
+    recognized provider (``_normalize`` returns None) — nothing useful
+    to add.
+    """
+    from .error_normalization import _normalize
+
+    normalized = _normalize(request, exc)
+    if normalized is not None:
+        raise normalized from exc
+    raise exc
 
 
 def _guard_and_fallback(
@@ -330,7 +361,7 @@ def _guard_and_fallback(
             f"Model error ({reason}) -- not eligible for fallback, re-raising",
             style="red",
         )
-        raise primary_exc
+        _raise_normalized(request, primary_exc)
     return _try_fallbacks(request, invoke, primary_exc)
 
 

EvoScientist/middleware/repetitive_tool_guard.py

@@ -0,0 +1,350 @@
+"""Detect deterministic tool loops and compact only provider-facing history."""
+
+from __future__ import annotations
+
+import json
+import logging
+import re
+from collections.abc import Awaitable, Callable, Mapping, Sequence
+from dataclasses import dataclass
+from typing import Any
+
+from langchain.agents.middleware.types import (
+    AgentMiddleware,
+    ModelRequest,
+    ModelResponse,
+)
+
+from ..llm.errors import AgentControlError
+
+logger = logging.getLogger(__name__)
+
+DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD = 2
+DEFAULT_MAX_CONSECUTIVE_TOOL_ERRORS = 3
+
+_TRANSIENT_PATTERNS = (
+    "timeout",
+    "timed out",
+    "cancelled",
+    "canceled",
+    "connection",
+    "rate limit",
+    "too many requests",
+    "temporarily unavailable",
+    "service unavailable",
+    "overloaded",
+    "bad gateway",
+    "gateway timeout",
+    "http 500",
+    "http 502",
+    "http 503",
+    "http 504",
+)
+_DETERMINISTIC_PATTERNS: tuple[tuple[str, tuple[str, ...]], ...] = (
+    (
+        "INVALID_ARGUMENTS",
+        ("invalid argument", "validation error", "schema", "bad input"),
+    ),
+    ("UNKNOWN_TOOL", ("not a valid tool", "unknown tool", "tool not found")),
+    ("UNSUPPORTED", ("not supported", "unsupported", "not implemented")),
+    (
+        "POLICY_DENIED",
+        ("permission denied", "forbidden", "policy denied", "not allowed"),
+    ),
+)
+_SAFE_CODE_RE = re.compile(r"^[A-Za-z][A-Za-z0-9_.:-]{0,95}$")
+_DETERMINISTIC_CODE_MARKERS = (
+    "INVALID",
+    "VALIDATION",
+    "SCHEMA",
+    "UNKNOWN_TOOL",
+    "NOT_FOUND",
+    "UNSUPPORTED",
+    "NOT_IMPLEMENTED",
+    "POLICY",
+    "PERMISSION",
+    "FORBIDDEN",
+    "DENIED",
+)
+
+
+@dataclass(frozen=True, slots=True)
+class RepetitiveToolHistoryRepair:
+    messages: list[Any]
+    blocked_tool_names: frozenset[str]
+    removed_rounds: int
+    tail_repetitions: int = 0
+    tail_consecutive_errors: int = 0
+
+
+@dataclass(frozen=True, slots=True)
+class _ToolRound:
+    messages: tuple[Any, ...]
+    signature: tuple[tuple[str, str, str], ...]
+    tool_names: frozenset[str]
+    deterministic_error: bool
+
+
+def _canonical_tool_args(value: Any) -> str:
+    if isinstance(value, str):
+        try:
+            value = json.loads(value)
+        except json.JSONDecodeError:
+            return value.strip()
+    try:
+        return json.dumps(
+            value,
+            ensure_ascii=False,
+            sort_keys=True,
+            separators=(",", ":"),
+            default=str,
+        )
+    except (TypeError, ValueError):
+        return repr(value)
+
+
+def _deterministic_result_code(message: Any) -> str | None:
+    additional = getattr(message, "additional_kwargs", None)
+    additional = additional if isinstance(additional, Mapping) else {}
+    raw_code = additional.get("error_code") or additional.get("code")
+    status = str(getattr(message, "status", "") or "").lower()
+    content = str(getattr(message, "content", "") or "")
+    lowered = content.lower()
+
+    if any(pattern in lowered for pattern in _TRANSIENT_PATTERNS):
+        return None
+    if isinstance(raw_code, str) and _SAFE_CODE_RE.fullmatch(raw_code.strip()):
+        normalized = raw_code.strip().upper()
+        if any(
+            pattern.replace(" ", "_") in normalized for pattern in _TRANSIENT_PATTERNS
+        ):
+            return None
+        if any(marker in normalized for marker in _DETERMINISTIC_CODE_MARKERS):
+            return normalized
+        return None
+    is_error = status == "error" or lowered.startswith("error:")
+    if not is_error:
+        return None
+    for code, patterns in _DETERMINISTIC_PATTERNS:
+        if any(pattern in lowered for pattern in patterns):
+            return code
+    return None
+
+
+def _parse_tool_round(
+    messages: Sequence[Any], start: int
+) -> tuple[_ToolRound, int] | None:
+    assistant = messages[start]
+    if getattr(assistant, "type", None) != "ai":
+        return None
+    raw_calls = list(getattr(assistant, "tool_calls", None) or [])
+    calls = [call for call in raw_calls if isinstance(call, Mapping)]
+    if not calls or len(calls) != len(raw_calls):
+        return None
+
+    end = start + 1
+    results: list[Any] = []
+    while end < len(messages) and getattr(messages[end], "type", None) == "tool":
+        results.append(messages[end])
+        end += 1
+    if not results:
+        return None
+    results_by_id = {
+        str(getattr(result, "tool_call_id", "") or "").strip(): result
+        for result in results
+        if str(getattr(result, "tool_call_id", "") or "").strip()
+    }
+
+    signature: list[tuple[str, str, str]] = []
+    tool_names: set[str] = set()
+    for index, call in enumerate(calls):
+        name = str(call.get("name") or "").strip()
+        call_id = str(call.get("id") or "").strip()
+        if not name or not call_id:
+            return None
+        result = results_by_id.get(call_id)
+        if result is None and index < len(results):
+            candidate = results[index]
+            if not str(getattr(candidate, "tool_call_id", "") or "").strip():
+                result = candidate
+        if result is None:
+            return None
+        result_code = _deterministic_result_code(result)
+        if result_code is None:
+            return _ToolRound(
+                messages=(assistant, *results),
+                signature=(),
+                tool_names=frozenset(),
+                deterministic_error=False,
+            ), end
+        signature.append((name, _canonical_tool_args(call.get("args")), result_code))
+        tool_names.add(name)
+
+    return (
+        _ToolRound(
+            messages=(assistant, *results),
+            signature=tuple(signature),
+            tool_names=frozenset(tool_names),
+            deterministic_error=True,
+        ),
+        end,
+    )
+
+
+def collapse_repetitive_tool_rounds(
+    messages: Sequence[Any],
+    *,
+    threshold: int = DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD,
+) -> RepetitiveToolHistoryRepair:
+    """Build a provider-only projection while preserving audit history.
+
+    Only the middle rounds of three-or-more identical deterministic error
+    groups are omitted. The first and last observations remain, and callers
+    must never persist this projection back to a checkpoint.
+    """
+    if not isinstance(threshold, int) or isinstance(threshold, bool) or threshold < 0:
+        raise ValueError("repetitive tool call threshold must be non-negative")
+    original = list(messages)
+    segments: list[Any | _ToolRound] = []
+    index = 0
+    while index < len(original):
+        parsed = _parse_tool_round(original, index)
+        if parsed is None:
+            segments.append(original[index])
+            index += 1
+            continue
+        tool_round, index = parsed
+        segments.append(tool_round)
+
+    tail_repetitions = 0
+    tail_consecutive_errors = 0
+    if segments and isinstance(segments[-1], _ToolRound):
+        tail = segments[-1]
+        if tail.deterministic_error:
+            cursor = len(segments) - 1
+            while cursor >= 0 and isinstance(segments[cursor], _ToolRound):
+                current = segments[cursor]
+                if not current.deterministic_error:
+                    break
+                tail_consecutive_errors += 1
+                cursor -= 1
+            cursor = len(segments) - 1
+            while cursor >= 0 and isinstance(segments[cursor], _ToolRound):
+                current = segments[cursor]
+                if (
+                    not current.deterministic_error
+                    or current.signature != tail.signature
+                ):
+                    break
+                tail_repetitions += 1
+                cursor -= 1
+
+    projected: list[Any] = []
+    removed_rounds = 0
+    index = 0
+    while index < len(segments):
+        segment = segments[index]
+        if not isinstance(segment, _ToolRound) or not segment.deterministic_error:
+            if isinstance(segment, _ToolRound):
+                projected.extend(segment.messages)
+            else:
+                projected.append(segment)
+            index += 1
+            continue
+        end = index + 1
+        while (
+            end < len(segments)
+            and isinstance(segments[end], _ToolRound)
+            and segments[end].deterministic_error
+            and segments[end].signature == segment.signature
+        ):
+            end += 1
+        group = segments[index:end]
+        should_compact = threshold > 0 and len(group) >= threshold and len(group) > 2
+        if should_compact:
+            projected.extend(group[0].messages)
+            projected.extend(group[-1].messages)
+            removed_rounds += len(group) - 2
+        else:
+            for item in group:
+                projected.extend(item.messages)
+        index = end
+
+    blocked = (
+        segments[-1].tool_names
+        if tail_repetitions and isinstance(segments[-1], _ToolRound)
+        else frozenset()
+    )
+    return RepetitiveToolHistoryRepair(
+        messages=projected,
+        blocked_tool_names=blocked,
+        removed_rounds=removed_rounds,
+        tail_repetitions=tail_repetitions,
+        tail_consecutive_errors=tail_consecutive_errors,
+    )
+
+
+class RepetitiveToolCallGuardMiddleware(AgentMiddleware):
+    """Stop deterministic loops before another model request is made."""
+
+    name = "repetitive_tool_call_guard"
+
+    def __init__(
+        self,
+        *,
+        threshold: int = DEFAULT_REPETITIVE_TOOL_CALL_THRESHOLD,
+        max_consecutive_errors: int = DEFAULT_MAX_CONSECUTIVE_TOOL_ERRORS,
+    ) -> None:
+        super().__init__()
+        for name, value in {
+            "threshold": threshold,
+            "max_consecutive_errors": max_consecutive_errors,
+        }.items():
+            if not isinstance(value, int) or isinstance(value, bool) or value < 0:
+                raise ValueError(f"{name} must be a non-negative integer")
+        self.threshold = threshold
+        self.max_consecutive_errors = max_consecutive_errors
+
+    def _prepare_request(self, request: ModelRequest) -> ModelRequest:
+        repair = collapse_repetitive_tool_rounds(
+            request.messages,
+            threshold=self.threshold,
+        )
+        if self.threshold and repair.tail_repetitions >= self.threshold:
+            raise AgentControlError(
+                "MODEL_TOOL_LOOP_DETECTED",
+                "A deterministic repeated tool-call loop was stopped.",
+                status_code=422,
+                retryable=False,
+            )
+        if (
+            self.max_consecutive_errors
+            and repair.tail_consecutive_errors >= self.max_consecutive_errors
+        ):
+            raise AgentControlError(
+                "MODEL_TOOL_ERROR_LIMIT",
+                "Too many consecutive deterministic tool errors were stopped.",
+                status_code=422,
+                retryable=False,
+            )
+        if repair.removed_rounds:
+            logger.info(
+                "Compacted deterministic tool errors for provider projection: removed_rounds=%d",
+                repair.removed_rounds,
+            )
+            return request.override(messages=repair.messages)
+        return request
+
+    def wrap_model_call(
+        self,
+        request: ModelRequest,
+        handler: Callable[[ModelRequest], ModelResponse],
+    ) -> ModelResponse:
+        return handler(self._prepare_request(request))
+
+    async def awrap_model_call(
+        self,
+        request: ModelRequest,
+        handler: Callable[[ModelRequest], Awaitable[ModelResponse]],
+    ) -> ModelResponse:
+        return await handler(self._prepare_request(request))

EvoScientist/middleware/tool_protocol_guard.py

@@ -0,0 +1,361 @@
+"""Validate completed model tool calls before they can reach ToolNode."""
+
+from __future__ import annotations
+
+import hashlib
+import json
+from collections.abc import Awaitable, Callable, Mapping, Sequence
+from typing import Any
+
+from langchain.agents.middleware.types import (
+    AgentMiddleware,
+    ExtendedModelResponse,
+    ModelRequest,
+    ModelResponse,
+)
+from langchain_core.messages import AIMessage
+from langchain_core.tools import BaseTool
+
+from ..llm.errors import ModelToolProtocolError, _provider_from_model
+
+_TOOL_BLOCK_TYPES = frozenset({"tool_call", "function_call"})
+_MAX_DIAGNOSTIC_KEYS = 16
+_MAX_DIAGNOSTIC_KEY_CHARS = 64
+
+
+def _tool_name(tool: BaseTool | Mapping[str, Any] | Any) -> str | None:
+    if isinstance(tool, BaseTool):
+        return tool.name.strip() or None
+    if isinstance(tool, Mapping):
+        value = tool.get("name")
+        if not value and isinstance(tool.get("function"), Mapping):
+            value = tool["function"].get("name")
+        if isinstance(value, str) and value.strip():
+            return value.strip()
+        return None
+    value = getattr(tool, "name", None)
+    return value.strip() if isinstance(value, str) and value.strip() else None
+
+
+def _ai_messages(response: Any) -> list[AIMessage]:
+    """Extract final AI messages from every LangChain middleware response shape."""
+    if isinstance(response, AIMessage):
+        return [response]
+    if isinstance(response, ExtendedModelResponse):
+        response = response.model_response
+    elif not isinstance(response, ModelResponse):
+        nested = getattr(response, "model_response", None)
+        if nested is not None:
+            response = nested
+    result = getattr(response, "result", None)
+    if not isinstance(result, Sequence) or isinstance(result, str | bytes):
+        return []
+    return [message for message in result if isinstance(message, AIMessage)]
+
+
+def _block_identity(block: Mapping[str, Any]) -> tuple[str, str]:
+    call_id = str(block.get("id") or block.get("call_id") or "").strip()
+    name = block.get("name") or block.get("tool_name")
+    function = block.get("function")
+    if not name and isinstance(function, Mapping):
+        name = function.get("name")
+    return call_id, str(name or "").strip()
+
+
+def _value_digest(value: Any) -> str:
+    try:
+        encoded = json.dumps(
+            value,
+            ensure_ascii=False,
+            sort_keys=True,
+            separators=(",", ":"),
+            default=lambda item: f"<{type(item).__name__}>",
+        ).encode("utf-8")
+    except (TypeError, ValueError):
+        encoded = f"<{type(value).__name__}:unserializable>".encode()
+    return "sha256:" + hashlib.sha256(encoded).hexdigest()[:16]
+
+
+def _argument_diagnostic(value: Any, *, present: bool) -> dict[str, Any]:
+    if not present:
+        return {"args_present": False, "args_type": "missing"}
+    if isinstance(value, Mapping):
+        keys = sorted(str(key)[:_MAX_DIAGNOSTIC_KEY_CHARS] for key in value)
+        return {
+            "args_present": True,
+            "args_type": "object",
+            "args_key_count": len(keys),
+            "args_keys": keys[:_MAX_DIAGNOSTIC_KEYS],
+            "args_keys_truncated": len(keys) > _MAX_DIAGNOSTIC_KEYS,
+            "args_digest": _value_digest(value),
+        }
+    if isinstance(value, Sequence) and not isinstance(value, str | bytes):
+        value_type = "array"
+    elif isinstance(value, str):
+        value_type = "string"
+    elif value is None:
+        value_type = "null"
+    else:
+        value_type = type(value).__name__
+    return {
+        "args_present": True,
+        "args_type": value_type,
+        "args_digest": _value_digest(value),
+    }
+
+
+def _summarize_call(call: Any) -> dict[str, Any]:
+    if not isinstance(call, Mapping):
+        return {"call_type": type(call).__name__}
+    function = call.get("function")
+    function = function if isinstance(function, Mapping) else {}
+    call_id = str(call.get("id") or call.get("call_id") or "").strip()
+    name = call.get("name") or call.get("tool_name") or function.get("name")
+    name = str(name or "").strip()
+    if "args" in call:
+        args = call.get("args")
+        args_present = True
+    elif "arguments" in call:
+        args = call.get("arguments")
+        args_present = True
+    elif "arguments" in function:
+        args = function.get("arguments")
+        args_present = True
+    else:
+        args = None
+        args_present = False
+    summary = {
+        "call_type": "object",
+        "name": name or "<missing>",
+        "id_present": bool(call_id),
+        **_argument_diagnostic(args, present=args_present),
+    }
+    if call_id:
+        summary["id_fingerprint"] = _value_digest(call_id)
+    return summary
+
+
+def _raw_openai_call(message: AIMessage, call_index: int) -> Any | None:
+    additional = getattr(message, "additional_kwargs", None)
+    additional = additional if isinstance(additional, Mapping) else {}
+    raw_calls = additional.get("tool_calls")
+    if (
+        isinstance(raw_calls, Sequence)
+        and not isinstance(raw_calls, str | bytes)
+        and call_index < len(raw_calls)
+    ):
+        return raw_calls[call_index]
+    return None
+
+
+def _call_diagnostic(
+    message: AIMessage,
+    call: Any,
+    *,
+    source: str,
+    call_index: int,
+    call_count: int,
+) -> dict[str, Any]:
+    diagnostic = {
+        "source": source,
+        "call_index": call_index,
+        "call_count": call_count,
+        **_summarize_call(call),
+    }
+    raw_call = _raw_openai_call(message, call_index)
+    diagnostic["raw_openai_call_available"] = raw_call is not None
+    if raw_call is not None:
+        diagnostic["raw_openai_call"] = _summarize_call(raw_call)
+    return diagnostic
+
+
+def _route_metadata(request: ModelRequest) -> dict[str, Any]:
+    model = request.model
+    metadata = getattr(model, "metadata", None)
+    metadata = metadata if isinstance(metadata, Mapping) else {}
+    provider = metadata.get("route_provider") or _provider_from_model(model)
+    model_id = metadata.get("route_model")
+    if not model_id:
+        model_id = (
+            getattr(model, "model_name", None)
+            or getattr(model, "model", None)
+            or getattr(model, "model_id", None)
+        )
+    generation = metadata.get("route_config_generation")
+    try:
+        config_generation = int(generation) if generation is not None else None
+    except (TypeError, ValueError):
+        config_generation = None
+    return {
+        "provider": str(provider) if provider else None,
+        "model": str(model_id) if model_id else None,
+        "route_key": str(metadata.get("route_key"))
+        if metadata.get("route_key")
+        else None,
+        "config_generation": config_generation,
+        "api_mode": str(metadata.get("route_api_mode"))
+        if metadata.get("route_api_mode")
+        else None,
+        "endpoint": str(metadata.get("route_endpoint"))
+        if metadata.get("route_endpoint")
+        else None,
+        "tool_call_transport": str(metadata.get("route_tool_call_transport"))
+        if metadata.get("route_tool_call_transport")
+        else None,
+    }
+
+
+def _raise_protocol_error(
+    request: ModelRequest,
+    reason: str,
+    *,
+    call_id: str | None = None,
+    call_diagnostic: dict[str, Any] | None = None,
+) -> None:
+    raise ModelToolProtocolError(
+        reason,
+        call_id=call_id or None,
+        call_diagnostic=call_diagnostic,
+        **_route_metadata(request),
+    )
+
+
+def _validate_message(
+    message: AIMessage,
+    request: ModelRequest,
+    allowed_names: frozenset[str],
+) -> None:
+    invalid_calls = list(getattr(message, "invalid_tool_calls", None) or [])
+    if invalid_calls:
+        invalid = invalid_calls[0]
+        call_id = str(invalid.get("id") or "") if isinstance(invalid, Mapping) else ""
+        _raise_protocol_error(
+            request,
+            "invalid_final_call",
+            call_id=call_id,
+            call_diagnostic=_call_diagnostic(
+                message,
+                invalid,
+                source="invalid_tool_calls",
+                call_index=0,
+                call_count=len(invalid_calls),
+            ),
+        )
+
+    parsed_by_id: dict[str, str] = {}
+    parsed_calls = list(getattr(message, "tool_calls", None) or [])
+    for call_index, raw_call in enumerate(parsed_calls):
+        diagnostic = _call_diagnostic(
+            message,
+            raw_call,
+            source="parsed_tool_calls",
+            call_index=call_index,
+            call_count=len(parsed_calls),
+        )
+        if not isinstance(raw_call, Mapping):
+            _raise_protocol_error(
+                request, "invalid_final_call", call_diagnostic=diagnostic
+            )
+        call_id = str(raw_call.get("id") or raw_call.get("call_id") or "").strip()
+        name = str(raw_call.get("name") or "").strip()
+        if not name:
+            _raise_protocol_error(
+                request,
+                "missing_name",
+                call_id=call_id,
+                call_diagnostic=diagnostic,
+            )
+        if name not in allowed_names:
+            _raise_protocol_error(
+                request,
+                "unknown_name",
+                call_id=call_id,
+                call_diagnostic=diagnostic,
+            )
+        if not call_id:
+            _raise_protocol_error(request, "missing_id", call_diagnostic=diagnostic)
+        if call_id in parsed_by_id:
+            _raise_protocol_error(
+                request,
+                "duplicate_id",
+                call_id=call_id,
+                call_diagnostic=diagnostic,
+            )
+        args = raw_call.get("args")
+        if not isinstance(args, Mapping):
+            _raise_protocol_error(
+                request,
+                "invalid_args",
+                call_id=call_id,
+                call_diagnostic=diagnostic,
+            )
+        parsed_by_id[call_id] = name
+
+    content = getattr(message, "content", None)
+    if not isinstance(content, list):
+        return
+    seen_block_ids: set[str] = set()
+    tool_blocks = [
+        block
+        for block in content
+        if isinstance(block, Mapping) and block.get("type") in _TOOL_BLOCK_TYPES
+    ]
+    for block_index, block in enumerate(tool_blocks):
+        diagnostic = _call_diagnostic(
+            message,
+            block,
+            source="content_blocks",
+            call_index=block_index,
+            call_count=len(tool_blocks),
+        )
+        call_id, name = _block_identity(block)
+        if not call_id:
+            _raise_protocol_error(request, "missing_id", call_diagnostic=diagnostic)
+        if call_id in seen_block_ids:
+            _raise_protocol_error(
+                request,
+                "duplicate_id",
+                call_id=call_id,
+                call_diagnostic=diagnostic,
+            )
+        seen_block_ids.add(call_id)
+        parsed_name = parsed_by_id.get(call_id)
+        if parsed_name is None or (name and name != parsed_name):
+            _raise_protocol_error(
+                request,
+                "inconsistent_block",
+                call_id=call_id,
+                call_diagnostic=diagnostic,
+            )
+
+
+class ToolProtocolGuardMiddleware(AgentMiddleware):
+    """Fail closed on malformed final tool calls using the actual request tools."""
+
+    name = "tool_protocol_guard"
+
+    @staticmethod
+    def _validate(response: Any, request: ModelRequest) -> None:
+        allowed_names = frozenset(
+            name for tool in request.tools if (name := _tool_name(tool)) is not None
+        )
+        for message in _ai_messages(response):
+            _validate_message(message, request, allowed_names)
+
+    def wrap_model_call(
+        self,
+        request: ModelRequest,
+        handler: Callable[[ModelRequest], ModelResponse],
+    ) -> ModelResponse:
+        response = handler(request)
+        self._validate(response, request)
+        return response
+
+    async def awrap_model_call(
+        self,
+        request: ModelRequest,
+        handler: Callable[[ModelRequest], Awaitable[ModelResponse]],
+    ) -> ModelResponse:
+        response = await handler(request)
+        self._validate(response, request)
+        return response

EvoScientist/middleware/tool_selector.py

@@ -48,6 +48,7 @@ DEFAULT_ALWAYS_INCLUDE_TOOLS: frozenset[str] = frozenset(
         "read_memory",
         "record_observation",
         "search_observations",
+        "write_todos",
     }
 )
 
@@ -132,10 +133,21 @@ class _ConditionalToolSelectorMiddleware(AgentMiddleware):
             return self._build_selector(request).wrap_model_call(
                 request, _handler_after_selection
             )
-        except Exception:
+        except Exception as exc:
             if _handler_called:
                 raise  # Error from downstream model — don't retry
-            # Selector itself failed (e.g., structured output not supported).
+            from ..llm.errors import ProviderStreamError
+            from .error_normalization import _is_provider_error
+
+            if isinstance(exc, ProviderStreamError) or _is_provider_error(exc):
+                # Auth / quota / connection failures on the selector's
+                # own model. Falling back to "use all tools" would hit
+                # the same provider anyway (same client, likely same
+                # credentials). Surface it instead so the user sees
+                # the real cause.
+                raise
+            # Structured-output shape / config failure — gracefully
+            # degrade to using all tools.
             logger.debug("Tool selector failed, using all tools", exc_info=True)
             if self._track_stream_selection:
                 _selector_active = False
@@ -171,9 +183,16 @@ class _ConditionalToolSelectorMiddleware(AgentMiddleware):
             return await self._build_selector(request).awrap_model_call(
                 request, _handler_after_selection
             )
-        except Exception:
+        except Exception as exc:
             if _handler_called:
                 raise
+            from ..llm.errors import ProviderStreamError
+            from .error_normalization import _is_provider_error
+
+            if isinstance(exc, ProviderStreamError) or _is_provider_error(exc):
+                # See sync path — surface provider errors, degrade only
+                # on shape / config failures.
+                raise
             logger.debug("Tool selector failed, using all tools", exc_info=True)
             if self._track_stream_selection:
                 _selector_active = False
@@ -258,6 +277,15 @@ def create_tool_selector_middleware(
 
         model = _ensure_chat_model()
     safe_model = disable_thinking(model)
+    safe_model = safe_model.model_copy(
+        update={
+            "tags": [*(safe_model.tags or []), "metering:tool_selector"],
+            "metadata": {
+                **(safe_model.metadata or {}),
+                "metering_scope": "tool_selector",
+            },
+        }
+    )
 
     system_prompt = (
         "You are selecting tools for a scientific research agent. "

EvoScientist/paths.py

@@ -5,6 +5,7 @@ from __future__ import annotations
 import logging
 import os
 import shutil
+from collections.abc import Iterator
 from datetime import datetime
 from pathlib import Path
 
@@ -215,3 +216,65 @@ def resolve_virtual_path(virtual_path: str) -> Path:
     """Resolve a virtual workspace path (e.g. /image.png) to a real filesystem path."""
     vpath = virtual_path if virtual_path.startswith("/") else "/" + virtual_path
     return (_active_workspace / vpath.lstrip("/")).resolve()
+
+
+def evoscientist_root() -> Path:
+    """Return the application root used by Gateway-managed runtime data."""
+    env_root = os.environ.get("EVOSCIENTIST_HOME")
+    if env_root:
+        return Path(env_root).expanduser().resolve()
+    return DATA_DIR.expanduser().resolve()
+
+
+_EVOSCIENTIST_DATA_ROOT: Path | None = None
+
+
+def _data_root() -> Path:
+    """Return the root directory for isolated Web user workspaces."""
+    global _EVOSCIENTIST_DATA_ROOT
+    if _EVOSCIENTIST_DATA_ROOT is not None:
+        return _EVOSCIENTIST_DATA_ROOT
+
+    env_root = os.environ.get("EVOSCIENTIST_DATA_ROOT")
+    if env_root:
+        root = Path(env_root).expanduser().resolve()
+    else:
+        root = evoscientist_root() / "data"
+    _EVOSCIENTIST_DATA_ROOT = root
+    return root
+
+
+def user_data_dir(user_id: str) -> Path:
+    """Return and create the isolated data directory for a Web user."""
+    path = _data_root() / user_id
+    path.mkdir(parents=True, exist_ok=True)
+    return path
+
+
+def iter_user_data_dirs() -> Iterator[Path]:
+    """Yield existing Web user directories without creating the data root."""
+    root = _data_root()
+    if not root.exists():
+        return
+    for path in root.iterdir():
+        if path.is_dir():
+            yield path
+
+
+def thread_data_dir(user_id: str, thread_id: str) -> Path:
+    """Return and create a user's isolated thread workspace."""
+    path = user_data_dir(user_id) / thread_id
+    path.mkdir(parents=True, exist_ok=True)
+    return path
+
+
+def global_data_dir(user_id: str) -> Path:
+    """Return and create a user's directory shared across all threads."""
+    path = user_data_dir(user_id) / "__global__"
+    path.mkdir(parents=True, exist_ok=True)
+    return path
+
+
+def uploads_dir() -> Path:
+    """Return the Gateway upload staging directory."""
+    return evoscientist_root() / "uploads"

EvoScientist/runtime_integrations.py

@@ -0,0 +1,116 @@
+"""Optional runtime services supplied by an application embedding EvoScientist.
+
+The CLI package must not import a concrete web gateway. Applications such as
+Ai4Sci-Web can register their database, storage, metering, and media services
+at process startup through this module.
+"""
+
+from __future__ import annotations
+
+from collections.abc import Awaitable, Callable
+from dataclasses import dataclass, replace
+from datetime import date
+from pathlib import Path
+from typing import Any
+
+AsyncProvider = Callable[[], Awaitable[Any]]
+AsyncFileHandler = Callable[[Path], Awaitable[Any]]
+AsyncUsageRecorder = Callable[[str, str], Awaitable[Any]]
+ModelResolver = Callable[[str | None, str | None], Any | None]
+
+
+class RuntimeIntegrationUnavailable(RuntimeError):
+    """Raised when an optional host-provided service is not configured."""
+
+
+@dataclass(frozen=True)
+class RuntimeIntegrations:
+    app_connection_provider: AsyncProvider | None = None
+    session_connection_provider: AsyncProvider | None = None
+    session_dsn_provider: Callable[[], str | None] | None = None
+    current_date_provider: Callable[[], date] | None = None
+    user_storage_root_provider: Callable[[str], Path] | None = None
+    knowledge_file_handler: AsyncFileHandler | None = None
+    usage_recorder: AsyncUsageRecorder | None = None
+    image_backend_factory: Callable[[], Any] | None = None
+    model_resolver: ModelResolver | None = None
+
+
+_integrations = RuntimeIntegrations()
+
+
+def configure_runtime_integrations(**services: Any) -> RuntimeIntegrations:
+    """Register host-provided services and return the resulting configuration."""
+    global _integrations
+    _integrations = replace(_integrations, **services)
+    return _integrations
+
+
+def reset_runtime_integrations() -> None:
+    """Clear all host-provided services, primarily for tests."""
+    global _integrations
+    _integrations = RuntimeIntegrations()
+
+
+def has_session_connection_provider() -> bool:
+    return _integrations.session_connection_provider is not None
+
+
+def get_session_dsn() -> str | None:
+    provider = _integrations.session_dsn_provider
+    return provider() if provider is not None else None
+
+
+async def get_session_connection() -> Any:
+    provider = _integrations.session_connection_provider
+    if provider is None:
+        raise RuntimeIntegrationUnavailable(
+            "No session connection provider is configured"
+        )
+    return await provider()
+
+
+async def get_app_connection() -> Any:
+    provider = _integrations.app_connection_provider
+    if provider is None:
+        raise RuntimeIntegrationUnavailable(
+            "No application connection provider is configured"
+        )
+    return await provider()
+
+
+def current_date() -> date:
+    provider = _integrations.current_date_provider
+    return provider() if provider is not None else date.today()
+
+
+def resolve_user_storage_root(user_id: str) -> Path | None:
+    provider = _integrations.user_storage_root_provider
+    return provider(user_id) if provider is not None else None
+
+
+def resolve_runtime_model(model: str | None, provider: str | None = None) -> Any | None:
+    """Resolve a host-managed model configuration when one is registered."""
+    resolver = _integrations.model_resolver
+    return resolver(model, provider) if resolver is not None else None
+
+
+async def handle_knowledge_file(path: Path) -> None:
+    handler = _integrations.knowledge_file_handler
+    if handler is not None:
+        await handler(path)
+
+
+async def record_service_usage(service: str, action: str) -> None:
+    recorder = _integrations.usage_recorder
+    if recorder is not None:
+        await recorder(service, action)
+
+
+def get_image_backend() -> Any:
+    factory = _integrations.image_backend_factory
+    if factory is None:
+        raise RuntimeIntegrationUnavailable(
+            "Image generation is unavailable in this runtime. Configure an image backend first."
+        )
+    return factory()

EvoScientist/stream/emitter.py

@@ -7,6 +7,15 @@ All events contain a type and associated data dict.
 from dataclasses import dataclass
 from typing import Any
 
+STREAM_PROTOCOL_CAPABILITIES = frozenset(
+    {
+        "task_snapshot_v1",
+        "complete_tool_call_v1",
+        "correlated_tool_call_id_v1",
+        "final_invalid_tool_call_v1",
+    }
+)
+
 
 @dataclass
 class StreamEvent:
@@ -158,6 +167,14 @@ class StreamEventEmitter:
             },
         )
 
+    @staticmethod
+    def task_snapshot(source: str, items: list[dict[str, Any]]) -> StreamEvent:
+        """Emit the complete root-agent task state without product-specific IDs."""
+        return StreamEvent(
+            "task_snapshot",
+            {"type": "task_snapshot", "source": source, "items": items},
+        )
+
     @staticmethod
     def interrupt(
         interrupt_id: str,
@@ -213,6 +230,19 @@ class StreamEventEmitter:
         )
 
     @staticmethod
-    def error(message: str) -> StreamEvent:
+    def error(
+        message: str,
+        *,
+        code: str | None = None,
+        recoverable: bool | None = None,
+        details: dict[str, Any] | None = None,
+    ) -> StreamEvent:
         """Error event."""
-        return StreamEvent("error", {"type": "error", "message": message})
+        data: dict[str, Any] = {"type": "error", "message": message}
+        if code is not None:
+            data["code"] = code
+        if recoverable is not None:
+            data["recoverable"] = recoverable
+        if details is not None:
+            data["details"] = details
+        return StreamEvent("error", data)

EvoScientist/stream/events.py

@@ -8,13 +8,15 @@ import base64
 import inspect
 import mimetypes
 import os
+import warnings
 from collections.abc import AsyncGenerator, AsyncIterator, Mapping
 from dataclasses import dataclass
 from typing import Any, TypeAlias
 
+from langchain_core._api import LangChainBetaWarning
 from langchain_core.messages import AIMessage, AIMessageChunk, BaseMessage, ToolMessage
 from langgraph.graph import END
-from langgraph.types import Command, Interrupt
+from langgraph.types import Command, Interrupt, Overwrite
 
 from ..memory.worker_activity import clear_completed_memory_activity_counts
 from .emitter import StreamEventEmitter
@@ -43,6 +45,12 @@ GraphRunInput: TypeAlias = str | Command
 LangGraphStreamInput: TypeAlias = dict[str, list[dict[str, object]]] | Command
 _ValueMessageKey: TypeAlias = tuple[str, ...]
 
+warnings.filterwarnings(
+    "ignore",
+    message=r"The v3 streaming protocol on Pregel is experimental\.",
+    category=LangChainBetaWarning,
+)
+
 
 @dataclass(frozen=True, slots=True)
 class _AssistantValueMessage:
@@ -87,9 +95,10 @@ async def _clear_interrupted_graph_state(
     no output and leaves the messages channel unchanged. From the user's side the
     conversation looks like it lost all history because the agent stops responding.
 
-    The fix: ``aupdate_state(config, None, as_node=END)`` clears all pending tasks
-    and writes a checkpoint whose ``next`` is the empty tuple, without touching
-    any channel values (message history is preserved).
+    Recovery first removes malformed/incomplete tool protocol from the messages
+    channel, then ``aupdate_state(config, None, as_node=END)`` clears pending
+    tasks and writes a checkpoint whose ``next`` is the empty tuple. Completed
+    tool call/result pairs and all non-tool history are preserved.
 
     Critically, this only runs when the stuck state is *not* a legitimate
     human-in-the-loop interrupt. The agent pauses via ``interrupt()`` /
@@ -106,10 +115,9 @@ async def _clear_interrupted_graph_state(
     _log = logging.getLogger(__name__)
     try:
         snapshot = await agent.aget_state(config)
-        # Only act when the graph is genuinely stuck (non-empty next tuple)...
-        if not snapshot or not getattr(snapshot, "next", None):
+        if not snapshot:
             return
-        # ...and not parked at a real human-in-the-loop interrupt.
+        # Never alter a real human-in-the-loop pause.
         if _snapshot_has_pending_interrupt(snapshot):
             _log.debug(
                 "Leaving interrupted graph state intact for thread %s: "
@@ -119,6 +127,13 @@ async def _clear_interrupted_graph_state(
             )
             return
 
+        await _repair_malformed_tool_history(agent, config, snapshot=snapshot)
+
+        # Only force END when the graph is genuinely stuck. Message repair also
+        # applies to failures that already left next empty.
+        if not getattr(snapshot, "next", None):
+            return
+
         stuck_at = snapshot.next
         await agent.aupdate_state(config, None, as_node=END)
         _log.debug(
@@ -134,6 +149,49 @@ async def _clear_interrupted_graph_state(
         )
 
 
+async def _repair_malformed_tool_history(
+    agent: Any,
+    config: dict[str, Any],
+    *,
+    snapshot: Any | None = None,
+) -> bool:
+    """Rewrite a checkpoint's messages to a replay-safe tool history.
+
+    Only structurally invalid protocol is removed. Completed tool call/result
+    pairs, including repeated successes and repeated errors, are audit and
+    billing facts and must remain in persistent history.
+    """
+
+    import logging
+
+    from ..llm.patches import _sanitize_openai_tool_history
+
+    _log = logging.getLogger(__name__)
+    if snapshot is None:
+        snapshot = await agent.aget_state(config)
+    if not snapshot or _snapshot_has_pending_interrupt(snapshot):
+        return False
+    values = getattr(snapshot, "values", None)
+    if not isinstance(values, Mapping):
+        return False
+    messages = values.get("messages")
+    if not isinstance(messages, list):
+        return False
+
+    repaired = _sanitize_openai_tool_history(messages)
+    if repaired == messages:
+        return False
+
+    await agent.aupdate_state(config, {"messages": Overwrite(repaired)})
+    _log.warning(
+        "Repaired structurally invalid tool history for thread %s: messages %d -> %d",
+        config.get("configurable", {}).get("thread_id", "?"),
+        len(messages),
+        len(repaired),
+    )
+    return True
+
+
 @dataclass(frozen=True)
 class _SubagentInfo:
     path: tuple[str, ...]
@@ -209,7 +267,12 @@ class _V3EventProcessor:
             tuple[tuple[str, ...], str], tuple[str, dict[str, Any]]
         ] = {}
         self._emitted_tool_calls: set[tuple[tuple[str, ...], str]] = set()
+        self._pending_tool_calls: dict[
+            tuple[tuple[str, ...], str], tuple[str, dict[str, Any]]
+        ] = {}
         self._emitted_interrupts: set[str] = set()
+        self._pending_invalid_tool_calls: dict[str, tuple[str, str]] = {}
+        self._last_task_snapshot: tuple[tuple[str, str], ...] | None = None
         self._selector = _ToolSelectionSuppressor(emitter)
 
     @staticmethod
@@ -237,13 +300,26 @@ class _V3EventProcessor:
         if method == "tools":
             return self._process_tool_event(namespace, _event_data(event), subagent)
         if method == "updates":
-            return self._process_update_event(_event_data(event))
+            return self._process_update_event(
+                _event_data(event), namespace=namespace, source="update"
+            )
         if method == "values":
             events: list[dict[str, Any]] = []
             params = event.get("params") or {}
             interrupts = params.get("interrupts") or ()
             if interrupts:
-                events.extend(self._process_update_event({"__interrupt__": interrupts}))
+                events.extend(
+                    self._process_update_event(
+                        {"__interrupt__": interrupts},
+                        namespace=namespace,
+                        source="values",
+                    )
+                )
+            events.extend(
+                self._process_update_event(
+                    _event_data(event), namespace=namespace, source="values"
+                )
+            )
             if self._process_value_message_snapshots and not namespace:
                 events.extend(self._process_value_messages(_event_data(event)))
             return events
@@ -382,6 +458,7 @@ class _V3EventProcessor:
                 inp, out = _usage_counts(usage) if usage is not None else (0, 0)
                 if inp or out:
                     events.append(self.emitter.usage_stats(inp, out).data)
+            events.extend(self._flush_invalid_tool_calls())
             return events
         return []
 
@@ -407,14 +484,12 @@ class _V3EventProcessor:
         if tool_call is None:
             return events
         tool_name, args, tool_call_id = tool_call
-        events.extend(
-            self._emit_tool_call_once(
-                namespace=namespace,
-                subagent=subagent,
-                name=tool_name,
-                args=args,
-                tool_call_id=tool_call_id,
-            )
+        self._pending_invalid_tool_calls.pop(tool_call_id, None)
+        self._pending_tool_calls[
+            (self._tool_scope(namespace, subagent), tool_call_id)
+        ] = (
+            tool_name,
+            args,
         )
         return events
 
@@ -457,6 +532,20 @@ class _V3EventProcessor:
             ]
         return [self.emitter.tool_call(name, args, tool_call_id).data]
 
+    def _pending_call_id(
+        self,
+        *,
+        scope: tuple[str, ...],
+        name: str,
+        args: dict[str, Any],
+    ) -> str:
+        matches = [
+            call_id
+            for (candidate_scope, call_id), candidate in self._pending_tool_calls.items()
+            if candidate_scope == scope and candidate == (name, args)
+        ]
+        return matches[0] if len(matches) == 1 else ""
+
     def _process_whole_message(
         self,
         msg: AIMessage | AIMessageChunk,
@@ -464,6 +553,16 @@ class _V3EventProcessor:
         namespace: tuple[str, ...],
     ) -> list[dict[str, Any]]:
         events: list[dict[str, Any]] = []
+        for invalid in getattr(msg, "invalid_tool_calls", ()) or ():
+            invalid_map = _as_raw_map(invalid)
+            if invalid_map is None:
+                continue
+            call_id = str(
+                invalid_map.get("id") or invalid_map.get("tool_call_id") or ""
+            )
+            name = str(invalid_map.get("name") or invalid_map.get("tool_name") or "")
+            key = call_id or f"chunk_{len(self._pending_invalid_tool_calls)}"
+            self._pending_invalid_tool_calls[key] = (call_id, name)
         additional = msg.additional_kwargs
         reasoning = additional.get("reasoning_content")
         emitted_reasoning = False
@@ -486,14 +585,12 @@ class _V3EventProcessor:
             if tool_call is None:
                 continue
             tool_name, args, tool_call_id = tool_call
-            events.extend(
-                self._emit_tool_call_once(
-                    namespace=namespace,
-                    subagent=subagent,
-                    name=tool_name,
-                    args=args,
-                    tool_call_id=tool_call_id,
-                )
+            self._pending_invalid_tool_calls.pop(tool_call_id, None)
+            self._pending_tool_calls[
+                (self._tool_scope(namespace, subagent), tool_call_id)
+            ] = (
+                tool_name,
+                args,
             )
 
         if subagent is None:
@@ -526,6 +623,9 @@ class _V3EventProcessor:
                 name,
                 args,
             )
+            self._pending_tool_calls.pop(
+                (self._tool_scope(namespace, subagent), tool_call_id), None
+            )
             events.extend(
                 self._emit_tool_call_once(
                     namespace=namespace,
@@ -572,6 +672,7 @@ class _V3EventProcessor:
                     content += "\n... (truncated)"
                 success = is_success(content)
 
+            lifecycle_key = (self._tool_scope(namespace, subagent), tool_call_id)
             if subagent is not None:
                 events.append(
                     self.emitter.subagent_tool_result(
@@ -583,22 +684,78 @@ class _V3EventProcessor:
                         instance_id=subagent.instance_id,
                     ).data
                 )
-                return events
-            events.append(
-                self.emitter.tool_result(
-                    name, content, success, tool_call_id=tool_call_id
-                ).data
-            )
+            else:
+                events.append(
+                    self.emitter.tool_result(
+                        name, content, success, tool_call_id=tool_call_id
+                    ).data
+                )
+            self._emitted_tool_calls.discard(lifecycle_key)
+            self._pending_tool_calls.pop(lifecycle_key, None)
             return events
 
         return []
 
-    def _process_update_event(self, data: object) -> list[dict[str, Any]]:
+    @staticmethod
+    def _normalize_task_items(value: object) -> list[dict[str, str]] | None:
+        if not isinstance(value, list):
+            return None
+        aliases = {
+            "todo": "pending",
+            "pending": "pending",
+            "active": "in_progress",
+            "in-progress": "in_progress",
+            "in_progress": "in_progress",
+            "done": "completed",
+            "completed": "completed",
+        }
+        items: list[dict[str, str]] = []
+        for raw in value:
+            raw_map = _as_raw_map(raw)
+            if raw_map is None:
+                continue
+            content = str(raw_map.get("content") or raw_map.get("task") or "").strip()
+            if not content:
+                continue
+            status = aliases.get(str(raw_map.get("status") or "pending").lower())
+            if status is None:
+                continue
+            items.append({"content": content, "status": status})
+        return items
+
+    @classmethod
+    def _find_task_items(cls, data: object) -> list[dict[str, str]] | None:
+        data_map = _as_raw_map(data)
+        if data_map is None:
+            return None
+        if "todos" in data_map:
+            return cls._normalize_task_items(data_map["todos"])
+        for value in data_map.values():
+            nested = _as_raw_map(value)
+            if nested is not None and "todos" in nested:
+                return cls._normalize_task_items(nested["todos"])
+        return None
+
+    def _process_update_event(
+        self,
+        data: object,
+        *,
+        namespace: tuple[str, ...] = (),
+        source: str = "update",
+    ) -> list[dict[str, Any]]:
         events: list[dict[str, Any]] = []
         data_map = _as_raw_map(data)
         if data_map is not None and "__interrupt__" in data_map:
             events.extend(self._process_interrupts(data_map["__interrupt__"]))
 
+        if not namespace:
+            items = self._find_task_items(data)
+            if items is not None:
+                signature = tuple((item["content"], item["status"]) for item in items)
+                if signature != self._last_task_snapshot:
+                    self._last_task_snapshot = signature
+                    events.append(self.emitter.task_snapshot(source, items).data)
+
         summarization_event = _find_summarization_event_payload(data)
         if summarization_event and not self._summarization_in_progress:
             signature = _summarization_event_signature(summarization_event)
@@ -614,6 +771,10 @@ class _V3EventProcessor:
             events.extend(self._emit_summarization_text(summary_text))
         return events
 
+    def _flush_invalid_tool_calls(self) -> list[dict[str, Any]]:
+        self._pending_invalid_tool_calls.clear()
+        return []
+
     def _process_interrupts(self, interrupts: object) -> list[dict[str, Any]]:
         events: list[dict[str, Any]] = []
         if not isinstance(interrupts, list | tuple):
@@ -652,26 +813,74 @@ class _V3EventProcessor:
             raw_questions = interrupt_map.get("questions")
             questions = raw_questions if isinstance(raw_questions, list) else []
             tc_id = str(interrupt_map.get("tool_call_id", ""))
-            return self._dedupe_interrupt_event(
-                self.emitter.ask_user_interrupt(
-                    interrupt_id,
-                    questions,
-                    tc_id,
-                ).data
+            events: list[dict[str, Any]] = []
+            candidate = self._pending_tool_calls.get(((), tc_id)) if tc_id else None
+            if candidate is not None:
+                events.extend(
+                    self._emit_tool_call_once(
+                        namespace=(),
+                        subagent=None,
+                        name=candidate[0],
+                        args=candidate[1],
+                        tool_call_id=tc_id,
+                    )
+                )
+            events.extend(
+                self._dedupe_interrupt_event(
+                    self.emitter.ask_user_interrupt(
+                        interrupt_id,
+                        questions,
+                        tc_id,
+                    ).data
+                )
             )
+            return events
 
         raw_action_reqs = interrupt_map.get("action_requests")
         action_reqs = raw_action_reqs if isinstance(raw_action_reqs, list) else []
         raw_review_cfgs = interrupt_map.get("review_configs")
         review_cfgs = raw_review_cfgs if isinstance(raw_review_cfgs, list) else None
         if action_reqs:
-            return self._dedupe_interrupt_event(
-                self.emitter.interrupt(
-                    interrupt_id,
-                    action_reqs,
-                    review_cfgs,
-                ).data
+            events: list[dict[str, Any]] = []
+            for raw_request in action_reqs:
+                request_map = _as_raw_map(raw_request)
+                if request_map is None:
+                    continue
+                call_id = str(
+                    request_map.get("id") or request_map.get("tool_call_id") or ""
+                )
+                name = str(request_map.get("name") or request_map.get("tool_name") or "")
+                args_map = _as_raw_map(
+                    request_map.get("args")
+                    if "args" in request_map
+                    else request_map.get("input")
+                )
+                if not call_id and name and args_map is not None:
+                    call_id = self._pending_call_id(
+                        scope=(),
+                        name=name,
+                        args=dict(args_map),
+                    )
+                if call_id and name and args_map is not None:
+                    events.extend(
+                        self._emit_tool_call_once(
+                            namespace=(),
+                            subagent=None,
+                            name=name,
+                            args=dict(args_map),
+                            tool_call_id=call_id,
+                        )
+                    )
+            events.extend(
+                self._dedupe_interrupt_event(
+                    self.emitter.interrupt(
+                        interrupt_id,
+                        action_reqs,
+                        review_cfgs,
+                    ).data
+                )
             )
+            return events
         return []
 
     def _dedupe_interrupt_event(self, event: dict[str, Any]) -> list[dict[str, Any]]:
@@ -797,6 +1006,8 @@ async def stream_agent_events(
     thread_id: str,
     metadata: dict[str, Any] | None = None,
     media: list[str] | None = None,
+    callbacks: list[Any] | None = None,
+    error_mode: str = "emit",
 ) -> AsyncGenerator[dict[str, Any], None]:
     """Stream events from a DeepAgents/LangGraph v3 run.
 
@@ -812,6 +1023,9 @@ async def stream_agent_events(
         metadata: Optional metadata dict merged into the LangGraph config
             (e.g. agent_name, updated_at for checkpoint persistence).
         media: Optional list of local file paths for attachments.
+        callbacks: Optional Runnable callbacks propagated to all nested model calls.
+        error_mode: ``emit`` preserves the generic error event; ``raise`` lets an
+            embedding host produce the single terminal error envelope.
 
     Yields:
         Event dicts: thinking, text, tool_call, tool_result,
@@ -821,6 +1035,8 @@ async def stream_agent_events(
     config: dict[str, Any] = {"configurable": {"thread_id": thread_id}}
     if metadata:
         config["metadata"] = metadata
+    if callbacks:
+        config["callbacks"] = callbacks
     emitter = StreamEventEmitter()
     existing_summarization_event: Mapping[str, object] | None = None
     try:
@@ -949,7 +1165,30 @@ async def stream_agent_events(
             yield item
     except Exception as e:
         _run_raised = True
-        yield emitter.error(str(e)).data
+        if error_mode == "emit":
+            payload = e.model_dump() if hasattr(e, "model_dump") else {}
+            if not isinstance(payload, Mapping):
+                payload = {}
+            code = str(payload.get("code") or "") or None
+            details = {
+                key: payload[key]
+                for key in (
+                    "reason",
+                    "provider",
+                    "model",
+                    "route_key",
+                    "config_generation",
+                    "api_mode",
+                    "call_id",
+                )
+                if payload.get(key) is not None
+            }
+            yield emitter.error(
+                str(payload.get("message") or e),
+                code=code,
+                recoverable=bool(payload.get("recoverable", True)) if code else None,
+                details=details or None,
+            ).data
         raise
     finally:
         if stream is not None:

README.md

@@ -10,7 +10,7 @@
 <a href="https://pypi.org/project/EvoScientist/"><picture>
   <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/EvoScientist/EvoScientist/main/.github/assets/badge-pypi-light.svg">
   <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/EvoScientist/EvoScientist/main/.github/assets/badge-pypi-dark.svg">
-  <img alt="PyPI v0.2.1" src="https://raw.githubusercontent.com/EvoScientist/EvoScientist/main/.github/assets/badge-pypi-light.svg" height="28">
+  <img alt="PyPI v0.2.2" src="https://raw.githubusercontent.com/EvoScientist/EvoScientist/main/.github/assets/badge-pypi-light.svg" height="28">
 </picture></a><a href="https://EvoScientist.github.io/"><picture>
   <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/EvoScientist/EvoScientist/main/.github/assets/badge-website-light.svg">
   <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/EvoScientist/EvoScientist/main/.github/assets/badge-website-dark.svg">
@@ -151,6 +151,7 @@ Moving beyond traditional human-in-the-loop systems, EvoScientist adopts a human
 <details>
 <summary>📦 Release Highlights — version changelog</summary>
 
+- **[11 Jul 2026]** **[v0.2.2](https://github.com/EvoScientist/EvoScientist/releases/tag/v0.2.2)** — New models selectable in onboarding and `/model`: GPT-5.6 (sol, terra, luna) for OpenAI and OpenRouter, plus Grok 4.5 and Tencent Hunyuan HY3 on OpenRouter; tighter config-file permissions and a reworked onboarding OAuth flow for auxiliary models.
 - **[05 Jul 2026]** **[v0.2.1](https://github.com/EvoScientist/EvoScientist/releases/tag/v0.2.1)** — AutoSkills: EvoMemory drafts reusable skills from its own observation clusters for you to review via `/autoskills`; a new `--output-format stream-json` for headless / SDK clients; richer slash-command completions; Windows UTF-8 config reads; a TUI welcome-banner fix; langchain-openrouter 0.2.5.
 - **[26 Jun 2026]** **[v0.2.0](https://github.com/EvoScientist/EvoScientist/releases/tag/v0.2.0)** — Scheduled tasks: cron-style recurring runs via `/schedule` or natural language, run unattended with shell-access gating; self-linking memory that connects observations into a knowledge graph (complements / contradicts / supersedes); a read-only `GET /api/models` endpoint for the WebUI model picker.
 - **[23 Jun 2026]** **[v0.1.9](https://github.com/EvoScientist/EvoScientist/releases/tag/v0.1.9)** — Hotfix for fresh installs: the first message crashed with `The subagent `task` tool cannot be exposed via `ptc`` after deepagents 0.6.11 / langchain-quickjs 0.3 reserved `task` as the REPL global. Removed `task` from the code-interpreter PTC allowlist (`task()` stays available as the REPL global; async dispatch stays in PTC) and pinned `deepagents[quickjs]~=0.6.11`.
@@ -433,10 +434,7 @@ EvoSci deploy                     # standalone LangGraph server for external UIs
 EvoSci -p "query" --output-format stream-json --auto-mode  # JSONL event stream on stdout (for programmatic clients)
 ```
 
-`--output-format stream-json` makes a single-shot (`-p`) run emit its native
-events as line-delimited JSON on stdout (one object per line), with all human
-output on stderr — the integration surface for headless clients (e.g. an agent
-runtime). See [docs/stream-json.md](docs/stream-json.md) for the event schema.
+`--output-format stream-json` makes a single-shot (`-p`) run emit its native events as line-delimited JSON on stdout (one object per line), with all human output on stderr — the integration surface for headless clients (e.g. an agent runtime). See [docs/guides/stream-json.md](docs/guides/stream-json.md) for the event schema.
 
 </details>
 

docs/README.md

@@ -11,6 +11,11 @@
 |------------------------------------------------------------|---------------------------------------------------------------------------------|
 | [macOS 24/7 Deployment](https://github.com/EvoScientist/EvoScientist/blob/main/docs/recipes/deployment-macos-24h.md#running-evoscientist-247-on-macos-telegram-bot--stt--ccproxy)   | Run EvoScientist as an always-on service on macOS with OAuth + Telegram + STT   |
 
+
+| Guide                                                      | Description                                                                     |
+|------------------------------------------------------------|---------------------------------------------------------------------------------|
+| [`stream-json` output protocol](https://github.com/EvoScientist/EvoScientist/blob/main/docs/guides/stream-json.md#stream-json-output-protocol)   | Line-delimited JSON event stream (`--output-format stream-json`) for driving EvoScientist headlessly from SDK / programmatic clients |
+
 ## Contributing a Recipe
 
 See the [Contributing Guide](../CONTRIBUTING.md) for general guidelines. When adding a new recipe:

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "EvoScientist"
-version = "0.2.1"
+version = "0.2.2"
 description = "EvoScientist: Towards Self-Evolving AI Scientists for End-to-End Scientific Discovery"
 readme = "README.md"
 requires-python = ">=3.11"
@@ -48,6 +48,7 @@ dependencies = [
 [dependency-groups]
 dev = [
     "pytest>=8.0",
+    "pytest-asyncio>=1.0",
     "pytest-cov>=5.0",
     "pytest-timeout>=2.4",
     "ruff>=0.5",
@@ -58,6 +59,7 @@ dev = [
 [project.optional-dependencies]
 dev = [
     "pytest>=8.0",
+    "pytest-asyncio>=1.0",
     "pytest-cov>=5.0",
     "pytest-timeout>=2.4",
     "ruff>=0.5",
@@ -117,6 +119,8 @@ EvoScientist = [
 
 [tool.pytest.ini_options]
 testpaths = ["tests"]
+asyncio_mode = "auto"
+asyncio_default_fixture_loop_scope = "function"
 filterwarnings = [
     "ignore::UserWarning:langchain_nvidia_ai_endpoints",
 ]

tests/conftest.py

@@ -1,34 +1,10 @@
 """Shared fixtures for EvoScientist tests."""
 
-import asyncio
+from pathlib import Path
 
 import pytest
 
-
-def run_async(coro):
-    """Run an async coroutine safely, cancelling pending tasks before closing.
-
-    This prevents 'Event loop is closed' errors from asyncio.Queue cleanup
-    when tasks are still waiting on Queue.get() at teardown time.
-    """
-    loop = asyncio.new_event_loop()
-    try:
-        return loop.run_until_complete(coro)
-    finally:
-        # Cancel all pending tasks so Queue getters don't raise on close
-        pending = asyncio.all_tasks(loop)
-        for task in pending:
-            task.cancel()
-        if pending:
-            loop.run_until_complete(asyncio.gather(*pending, return_exceptions=True))
-        loop.run_until_complete(loop.shutdown_asyncgens())
-        loop.close()
-
-
-@pytest.fixture(name="run_async")
-def run_async_fixture():
-    """Pytest fixture that exposes run_async as a callable for test functions."""
-    return run_async
+_NONEXISTENT_DOTENV = str(Path(__file__).with_name(".pytest-dotenv-does-not-exist"))
 
 
 @pytest.fixture(autouse=True)
@@ -192,3 +168,24 @@ def restore_model_passthrough_patch():
         yield
     finally:
         _reset()
+
+
+@pytest.fixture(autouse=True)
+def _isolate_dotenv(monkeypatch):
+    """Keep the developer's real .env out of the test environment.
+
+    ``get_effective_config`` runs ``load_dotenv(find_dotenv(usecwd=True),
+    override=True)``, so any test that loads config injects the repo's
+    real .env into ``os.environ`` for the rest of the pytest process.
+    An empty-valued line like ``MINIMAX_BASE_URL=`` then makes
+    ``os.environ.get(key, default)`` return "" instead of the default,
+    breaking unrelated tests later in the run (see issue #322).
+
+    Pointing ``find_dotenv`` at a fixed path that does not exist makes
+    ``load_dotenv`` a no-op without creating a temporary directory for
+    every test.
+    """
+    monkeypatch.setattr(
+        "EvoScientist.config.settings.find_dotenv",
+        lambda *args, **kwargs: _NONEXISTENT_DOTENV,
+    )

tests/stream_v3_fakes.py

@@ -9,7 +9,6 @@ from typing import Any
 from unittest.mock import MagicMock
 
 from EvoScientist.stream.events import stream_agent_events
-from tests.conftest import run_async
 
 
 async def async_iter(items: Iterable[Any]) -> AsyncIterator[Any]:
@@ -17,24 +16,20 @@ async def async_iter(items: Iterable[Any]) -> AsyncIterator[Any]:
         yield item
 
 
-def collect_events(
+async def collect_events(
     agent,
     message: str = "hi",
     thread_id: str = "t1",
 ):
-    """Collect stream_agent_events output for synchronous tests."""
-
-    async def _run():
-        events = []
-        async for ev in stream_agent_events(
-            agent,
-            message,
-            thread_id,
-        ):
-            events.append(ev)
-        return events
-
-    return run_async(_run())
+    """Collect stream_agent_events output for tests."""
+    events = []
+    async for ev in stream_agent_events(
+        agent,
+        message,
+        thread_id,
+    ):
+        events.append(ev)
+    return events
 
 
 def protocol_event(

tests/test_additional_channel_smoke.py

@@ -10,18 +10,17 @@ from EvoScientist.channels.imessage.channel_rpc import (
 )
 from EvoScientist.channels.qq.channel import QQChannel, QQConfig
 from EvoScientist.channels.signal.channel import SignalChannel, SignalConfig
-from tests.conftest import run_async as _run
 
 
 class TestEmailChannelSmoke:
-    def test_start_raises_without_required_imap_settings(self):
+    async def test_start_raises_without_required_imap_settings(self):
         channel = EmailChannel(EmailConfig())
         with pytest.raises(
             ChannelError, match="imap_host and imap_username are required"
         ):
-            _run(channel.start())
+            await channel.start()
 
-    def test_send_returns_false_when_smtp_not_ready(self):
+    async def test_send_returns_false_when_smtp_not_ready(self):
         channel = EmailChannel(EmailConfig())
         msg = OutboundMessage(
             channel="email",
@@ -29,16 +28,16 @@ class TestEmailChannelSmoke:
             content="hello",
             metadata={"chat_id": "user@example.com"},
         )
-        assert _run(channel.send(msg)) is False
+        assert await channel.send(msg) is False
 
 
 class TestSignalChannelSmoke:
-    def test_start_raises_without_phone_number(self):
+    async def test_start_raises_without_phone_number(self):
         channel = SignalChannel(SignalConfig())
         with pytest.raises(ChannelError, match="phone_number is required"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_send_returns_false_when_not_connected(self):
+    async def test_send_returns_false_when_not_connected(self):
         channel = SignalChannel(SignalConfig(phone_number="+123456789"))
         msg = OutboundMessage(
             channel="signal",
@@ -46,27 +45,29 @@ class TestSignalChannelSmoke:
             content="hello",
             metadata={"chat_id": "+123456789"},
         )
-        assert _run(channel.send(msg)) is False
+        assert await channel.send(msg) is False
 
 
 class TestQQChannelSmoke:
-    def test_start_raises_when_sdk_missing(self, monkeypatch):
+    async def test_start_raises_when_sdk_missing(self, monkeypatch):
         from EvoScientist.channels.qq import channel as qq_module
 
         monkeypatch.setattr(qq_module, "QQ_AVAILABLE", False)
         channel = QQChannel(QQConfig(app_id="id", app_secret="secret"))
         with pytest.raises(ChannelError, match="SDK not installed"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_start_raises_without_credentials_when_sdk_available(self, monkeypatch):
+    async def test_start_raises_without_credentials_when_sdk_available(
+        self, monkeypatch
+    ):
         from EvoScientist.channels.qq import channel as qq_module
 
         monkeypatch.setattr(qq_module, "QQ_AVAILABLE", True)
         channel = QQChannel(QQConfig(app_id="", app_secret=""))
         with pytest.raises(ChannelError, match="app_id and app_secret are required"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_send_returns_false_without_client(self):
+    async def test_send_returns_false_without_client(self):
         channel = QQChannel(QQConfig(app_id="id", app_secret="secret"))
         msg = OutboundMessage(
             channel="qq",
@@ -74,11 +75,11 @@ class TestQQChannelSmoke:
             content="hello",
             metadata={"chat_id": "openid"},
         )
-        assert _run(channel.send(msg)) is False
+        assert await channel.send(msg) is False
 
 
 class TestIMessageChannelSmoke:
-    def test_start_wraps_rpc_bootstrap_error(self, monkeypatch):
+    async def test_start_wraps_rpc_bootstrap_error(self, monkeypatch):
         async def _broken_start(self):
             raise RuntimeError("imsg not found")
 
@@ -87,9 +88,9 @@ class TestIMessageChannelSmoke:
         monkeypatch.setattr(imessage_module.ImsgRpcClient, "start", _broken_start)
         channel = IMessageChannelRpc(IMessageConfig())
         with pytest.raises(ChannelError, match="Failed to start imsg"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_send_returns_false_without_rpc_client(self):
+    async def test_send_returns_false_without_rpc_client(self):
         channel = IMessageChannelRpc(IMessageConfig())
         msg = OutboundMessage(
             channel="imessage",
@@ -97,4 +98,4 @@ class TestIMessageChannelSmoke:
             content="hello",
             metadata={"chat_id": "+123456789"},
         )
-        assert _run(channel.send(msg)) is False
+        assert await channel.send(msg) is False

tests/test_agent_factory_extensions.py

@@ -0,0 +1,137 @@
+def test_create_cli_agent_accepts_host_backend_and_memory_options(
+    monkeypatch, tmp_path
+):
+    import EvoScientist.EvoScientist as agent_module
+    from EvoScientist.config.settings import EvoScientistConfig
+
+    calls = {}
+    workspace_backend = object()
+    chat_model = object()
+
+    class _CompositeBackend:
+        def __init__(self, *, default, routes):
+            calls["default_backend"] = default
+            calls["routes"] = routes
+
+    class _MemoryBackend:
+        def __init__(self, **kwargs):
+            calls["memory_backend_kwargs"] = kwargs
+
+    class _SkillsBackend:
+        def __init__(self, **kwargs):
+            calls["skills_backend_kwargs"] = kwargs
+
+    class _Agent:
+        def with_config(self, config):
+            calls["agent_config"] = config
+            return self
+
+    cfg = EvoScientistConfig(auto_approve=True, recursion_limit=321)
+
+    monkeypatch.setattr("deepagents.backends.CompositeBackend", _CompositeBackend)
+    monkeypatch.setattr("deepagents.create_deep_agent", lambda **kwargs: _Agent())
+    monkeypatch.setattr("EvoScientist.backends.MemoryFilesystemBackend", _MemoryBackend)
+    monkeypatch.setattr("EvoScientist.backends.MergedSkillsBackend", _SkillsBackend)
+    monkeypatch.setattr(agent_module, "set_active_workspace", lambda path: None)
+    monkeypatch.setattr(
+        agent_module,
+        "_get_default_middleware",
+        lambda **kwargs: calls.setdefault("middleware_kwargs", kwargs) or [],
+    )
+    monkeypatch.setattr(
+        agent_module,
+        "load_mcp_and_build_kwargs",
+        lambda *args, **kwargs: {"subagents": [{"name": "research"}]},
+    )
+
+    memory_dir = tmp_path / "memory"
+    result = agent_module.create_cli_agent(
+        workspace_dir=str(tmp_path / "workspace"),
+        checkpointer=object(),
+        config=cfg,
+        chat_model=chat_model,
+        workspace_backend=workspace_backend,
+        memory_dir=memory_dir,
+        tool_selector_threshold=8,
+        memory_max_inline_profile_chars=1000,
+        enable_subagents=False,
+        enable_background_execution=False,
+    )
+
+    assert isinstance(result, _Agent)
+    assert calls["default_backend"] is workspace_backend
+    assert calls["memory_backend_kwargs"] == {
+        "root_dir": str(memory_dir),
+        "virtual_mode": True,
+    }
+    assert calls["middleware_kwargs"]["memory_dir"] == str(memory_dir)
+    assert calls["middleware_kwargs"]["tool_selector_threshold"] == 8
+    assert calls["middleware_kwargs"]["memory_max_inline_profile_chars"] == 1000
+    assert calls["middleware_kwargs"]["enable_background_execution"] is False
+    assert calls["agent_config"] == {"recursion_limit": 321}
+
+
+def test_create_cli_agent_installs_route_middleware_in_fixed_slot(monkeypatch, tmp_path):
+    import EvoScientist.EvoScientist as agent_module
+    from EvoScientist.config.settings import EvoScientistConfig
+
+    calls = {}
+
+    class _Middleware:
+        def __init__(self, name):
+            self.name = name
+
+    class _Backend:
+        def __init__(self, **_kwargs):
+            pass
+
+    class _CompositeBackend:
+        def __init__(self, **_kwargs):
+            pass
+
+    class _Agent:
+        def with_config(self, _config):
+            return self
+
+    default_chain = [
+        _Middleware("error_normalization"),
+        _Middleware("configurable_model"),
+        _Middleware("context_editing"),
+        _Middleware("tool_protocol_guard"),
+    ]
+    route = _Middleware("gateway_route_fallback")
+
+    monkeypatch.setattr("deepagents.backends.CompositeBackend", _CompositeBackend)
+    monkeypatch.setattr("deepagents.create_deep_agent", lambda **_kwargs: _Agent())
+    monkeypatch.setattr("EvoScientist.backends.MemoryFilesystemBackend", _Backend)
+    monkeypatch.setattr("EvoScientist.backends.MergedSkillsBackend", _Backend)
+    monkeypatch.setattr(agent_module, "set_active_workspace", lambda _path: None)
+    def fake_default_middleware(**kwargs):
+        calls["middleware_kwargs"] = kwargs
+        return list(default_chain)
+
+    monkeypatch.setattr(agent_module, "_get_default_middleware", fake_default_middleware)
+
+    def fake_load(_backend, middleware, **_kwargs):
+        calls["middleware"] = middleware
+        return {"subagents": []}
+
+    monkeypatch.setattr(agent_module, "load_mcp_and_build_kwargs", fake_load)
+
+    agent_module.create_cli_agent(
+        workspace_dir=str(tmp_path),
+        checkpointer=object(),
+        config=EvoScientistConfig(auto_approve=True),
+        chat_model=object(),
+        workspace_backend=object(),
+        main_agent_route_middleware=route,
+    )
+
+    assert calls["middleware_kwargs"]["enable_legacy_model_fallback"] is False
+    assert [middleware.name for middleware in calls["middleware"][:5]] == [
+        "error_normalization",
+        "configurable_model",
+        "gateway_route_fallback",
+        "context_editing",
+        "tool_protocol_guard",
+    ]

tests/test_agent_loader.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import asyncio
+import threading
 
 import pytest
 
@@ -93,88 +94,95 @@ def _make_loader_fn(agent_value="AGENT", fail_with=None, capture=None):
     return _loader
 
 
-def _run(coro):
-    return asyncio.run(coro)
+class _GatedThreadLoader:
+    """Callable loader that blocks until tests explicitly release it."""
+
+    def __init__(self, agent_value="AGENT", progress_events=()):
+        self.agent_value = agent_value
+        self.progress_events = tuple(progress_events)
+        self.started = threading.Event()
+        self.release = threading.Event()
+        self.finished = threading.Event()
+
+    def __call__(self, *, on_mcp_progress=None):
+        self.started.set()
+        self.release.wait(timeout=1)
+        try:
+            if on_mcp_progress is not None:
+                for event in self.progress_events:
+                    on_mcp_progress(*event)
+            return self.agent_value
+        finally:
+            self.finished.set()
+
+
+async def _wait_for_event(event, timeout=1):
+    return await asyncio.to_thread(event.wait, timeout)
 
 
 class TestBackgroundAgentLoaderStart:
-    def test_start_creates_task_and_forwards_kwargs(self):
+    async def test_start_creates_task_and_forwards_kwargs(self):
         captured: dict = {}
         loader = BackgroundAgentLoader(_make_loader_fn(capture=captured))
 
-        async def _go():
-            loader.start(workspace_dir="/ws", checkpointer="CK")
-            assert loader.task is not None
-            assert loader.is_pending
-            await loader.await_ready()
+        loader.start(workspace_dir="/ws", checkpointer="CK")
+        assert loader.task is not None
+        assert loader.is_pending
+        await loader.await_ready()
 
-        _run(_go())
         assert captured["kwargs"][0] == {"workspace_dir": "/ws", "checkpointer": "CK"}
 
-    def test_start_bumps_load_id(self):
+    async def test_start_bumps_load_id(self):
         loader = BackgroundAgentLoader(_make_loader_fn())
 
-        async def _go():
-            assert loader._load_id == 0
-            loader.start()
-            assert loader._load_id == 1
-            loader.start()
-            assert loader._load_id == 2
-            await loader.await_ready()
+        assert loader._load_id == 0
+        loader.start()
+        assert loader._load_id == 1
+        loader.start()
+        assert loader._load_id == 2
+        await loader.await_ready()
 
-        _run(_go())
+    async def test_start_cancels_in_flight_prior_task(self):
+        blocking = _GatedThreadLoader("LATE")
 
-    def test_start_cancels_in_flight_prior_task(self):
-        import time
-
-        def _blocking(*, on_mcp_progress=None):
-            time.sleep(0.05)
-            return "LATE"
-
-        async def _go():
-            loader = BackgroundAgentLoader(_blocking)
-            loader.start()
-            first_task = loader.task
-            # Supersede immediately; asyncio.to_thread wrapper gets cancelled.
-            loader._loader_fn = _make_loader_fn("FRESH")
-            loader.start()
-            agent = await loader.await_ready()
-            assert agent == "FRESH"
-            # Let the first thread drain so its done callback (gated) fires.
-            await asyncio.sleep(0.1)
-            assert first_task.cancelled() or first_task.done()
-
-        _run(_go())
+        loader = BackgroundAgentLoader(blocking)
+        loader.start()
+        first_task = loader.task
+        assert first_task is not None
+        assert await _wait_for_event(blocking.started)
+        # Supersede immediately; asyncio.to_thread wrapper gets cancelled.
+        loader._loader_fn = _make_loader_fn("FRESH")
+        loader.start()
+        agent = await loader.await_ready()
+        assert agent == "FRESH"
+        blocking.release.set()
+        try:
+            await first_task
+        except asyncio.CancelledError:
+            pass
+        assert first_task.cancelled() or first_task.done()
 
 
 class TestBackgroundAgentLoaderCallbacks:
-    def test_progress_hook_sees_events_in_order(self):
+    async def test_progress_hook_sees_events_in_order(self):
         events: list[tuple[str, str, str]] = []
         loader = BackgroundAgentLoader(
             _make_loader_fn(capture={}),
             on_progress=lambda e, s, d: events.append((e, s, d)),
         )
 
-        async def _go():
-            loader.start()
-            await loader.await_ready()
+        loader.start()
+        await loader.await_ready()
 
-        _run(_go())
         assert events == [("start", "srv", ""), ("success", "srv", "1")]
 
-    def test_stale_progress_events_are_dropped(self):
+    async def test_stale_progress_events_are_dropped(self):
         """A progress event fired after a newer `start` must not reach the hook."""
-        import time
-
+        slow_loader = _GatedThreadLoader(
+            "slow-agent", progress_events=[("success", "from-slow", "1")]
+        )
         seen: list[str] = []
 
-        # Loader 1 sleeps so its progress event fires AFTER load 2 starts.
-        def slow_loader(*, on_mcp_progress=None):
-            time.sleep(0.08)
-            if on_mcp_progress is not None:
-                on_mcp_progress("success", "from-slow", "1")
-            return "slow-agent"
-
         def fast_loader(*, on_mcp_progress=None):
             if on_mcp_progress is not None:
                 on_mcp_progress("success", "from-fast", "1")
@@ -184,36 +192,32 @@ class TestBackgroundAgentLoaderCallbacks:
             slow_loader, on_progress=lambda e, s, d: seen.append(s)
         )
 
-        async def _go():
-            loader.start()
-            # Supersede before the slow thread's event fires.
-            await asyncio.sleep(0.01)
-            loader._loader_fn = fast_loader
-            loader.start()
-            await loader.await_ready()
-            # Let the superseded thread finish (its event is gated out).
-            await asyncio.sleep(0.1)
+        loader.start()
+        assert await _wait_for_event(slow_loader.started)
+        # Loader 1 waits so its progress event fires AFTER load 2 starts.
+        loader._loader_fn = fast_loader
+        loader.start()
+        await loader.await_ready()
+        slow_loader.release.set()
+        assert await _wait_for_event(slow_loader.finished)
 
-        _run(_go())
         assert "from-fast" in seen
         assert "from-slow" not in seen
 
-    def test_success_callback_fires_on_completion(self):
+    async def test_success_callback_fires_on_completion(self):
         got = []
         loader = BackgroundAgentLoader(
             _make_loader_fn("MY_AGENT"),
             on_success=lambda a: got.append(a),
         )
 
-        async def _go():
-            loader.start()
-            await loader.await_ready()
-            await asyncio.sleep(0)  # let done-callback run
+        loader.start()
+        await loader.await_ready()
+        await asyncio.sleep(0)  # let done-callback run
 
-        _run(_go())
         assert got == ["MY_AGENT"]
 
-    def test_failure_callback_fires_on_error(self):
+    async def test_failure_callback_fires_on_error(self):
         err = RuntimeError("load failed")
         got_failures = []
         got_successes = []
@@ -223,40 +227,33 @@ class TestBackgroundAgentLoaderCallbacks:
             on_failure=lambda e: got_failures.append(e),
         )
 
-        async def _go():
-            loader.start()
-            with pytest.raises(RuntimeError, match="load failed"):
-                await loader.await_ready()
-            await asyncio.sleep(0)
+        loader.start()
+        with pytest.raises(RuntimeError, match="load failed"):
+            await loader.await_ready()
+        await asyncio.sleep(0)
 
-        _run(_go())
         assert got_failures == [err]
         assert got_successes == []
 
 
 class TestBackgroundAgentLoaderAwaitReady:
-    def test_returns_cached_agent_without_reawaiting(self):
+    async def test_returns_cached_agent_without_reawaiting(self):
         captured: dict = {}
         loader = BackgroundAgentLoader(_make_loader_fn("A", capture=captured))
 
-        async def _go():
-            loader.start()
-            assert await loader.await_ready() == "A"
-            assert await loader.await_ready() == "A"
+        loader.start()
+        assert await loader.await_ready() == "A"
+        assert await loader.await_ready() == "A"
 
-        _run(_go())
         assert len(captured["kwargs"]) == 1
 
-    def test_raises_if_started_not_called(self):
+    async def test_raises_if_started_not_called(self):
         loader = BackgroundAgentLoader(_make_loader_fn())
 
-        async def _go():
-            with pytest.raises(RuntimeError, match="before start"):
-                await loader.await_ready()
+        with pytest.raises(RuntimeError, match="before start"):
+            await loader.await_ready()
 
-        _run(_go())
-
-    def test_reraises_real_error_on_subsequent_awaits(self):
+    async def test_reraises_real_error_on_subsequent_awaits(self):
         """After a failure, ``await_ready`` must keep raising the real exception —
         not the "before start()" sentinel — until ``start`` is called again."""
 
@@ -265,16 +262,13 @@ class TestBackgroundAgentLoaderAwaitReady:
 
         loader = BackgroundAgentLoader(_fail)
 
-        async def _go():
-            loader.start()
-            with pytest.raises(RuntimeError, match="bad MCP config"):
-                await loader.await_ready()
-            with pytest.raises(RuntimeError, match="bad MCP config"):
-                await loader.await_ready()
+        loader.start()
+        with pytest.raises(RuntimeError, match="bad MCP config"):
+            await loader.await_ready()
+        with pytest.raises(RuntimeError, match="bad MCP config"):
+            await loader.await_ready()
 
-        _run(_go())
-
-    def test_needs_restart_flags_failed_load_for_retry(self):
+    async def test_needs_restart_flags_failed_load_for_retry(self):
         calls = {"n": 0}
 
         def flaky(*, on_mcp_progress=None):
@@ -285,17 +279,14 @@ class TestBackgroundAgentLoaderAwaitReady:
 
         loader = BackgroundAgentLoader(flaky)
 
-        async def _go():
-            assert loader.needs_restart  # never started
-            loader.start()
-            with pytest.raises(RuntimeError):
-                await loader.await_ready()
-            assert loader.needs_restart  # failed, caller may retry
-            loader.start()
-            assert await loader.await_ready() == "SECOND"
-            assert not loader.needs_restart  # success → no retry
-
-        _run(_go())
+        assert loader.needs_restart  # never started
+        loader.start()
+        with pytest.raises(RuntimeError):
+            await loader.await_ready()
+        assert loader.needs_restart  # failed, caller may retry
+        loader.start()
+        assert await loader.await_ready() == "SECOND"
+        assert not loader.needs_restart  # success → no retry
 
 
 class TestBackgroundAgentLoaderAdopt:
@@ -305,26 +296,19 @@ class TestBackgroundAgentLoaderAdopt:
         assert loader.agent == "EXTERNAL"
         assert not loader.is_pending
 
-    def test_adopt_supersedes_in_flight_load(self):
+    async def test_adopt_supersedes_in_flight_load(self):
         """A late background completion must not overwrite an adopted agent."""
-        import time
+        slow_loader = _GatedThreadLoader("FROM_BACKGROUND")
 
-        def _slow(*, on_mcp_progress=None):
-            time.sleep(0.08)
-            return "FROM_BACKGROUND"
+        loader = BackgroundAgentLoader(slow_loader)
 
-        loader = BackgroundAgentLoader(_slow)
-
-        async def _go():
-            loader.start()
-            await asyncio.sleep(0.01)
-            loader.adopt("FROM_MODEL")
-            # Give the background thread time to finish and fire its
-            # done-callback; the generation token should make it a no-op.
-            await asyncio.sleep(0.1)
-            assert loader.agent == "FROM_MODEL"
-
-        _run(_go())
+        loader.start()
+        assert await _wait_for_event(slow_loader.started)
+        loader.adopt("FROM_MODEL")
+        slow_loader.release.set()
+        assert await _wait_for_event(slow_loader.finished)
+        await asyncio.sleep(0)
+        assert loader.agent == "FROM_MODEL"
 
 
 class TestBackgroundAgentLoaderIsPending:
@@ -332,29 +316,22 @@ class TestBackgroundAgentLoaderIsPending:
         loader = BackgroundAgentLoader(_make_loader_fn())
         assert not loader.is_pending
 
-    def test_false_after_completion(self):
+    async def test_false_after_completion(self):
         loader = BackgroundAgentLoader(_make_loader_fn())
 
-        async def _go():
-            loader.start()
-            await loader.await_ready()
+        loader.start()
+        await loader.await_ready()
 
-        _run(_go())
         assert not loader.is_pending
 
-    def test_true_between_start_and_completion(self):
-        import time
+    async def test_true_between_start_and_completion(self):
+        wait_loader = _GatedThreadLoader("ok")
 
-        def _wait_loader(*, on_mcp_progress=None):
-            time.sleep(0.05)
-            return "ok"
+        loader = BackgroundAgentLoader(wait_loader)
 
-        loader = BackgroundAgentLoader(_wait_loader)
-
-        async def _go():
-            loader.start()
-            assert loader.is_pending
-            await loader.await_ready()
-            assert not loader.is_pending
-
-        _run(_go())
+        loader.start()
+        assert await _wait_for_event(wait_loader.started)
+        assert loader.is_pending
+        wait_loader.release.set()
+        await loader.await_ready()
+        assert not loader.is_pending

tests/test_async_notifier.py

@@ -5,6 +5,8 @@ import queue
 from types import SimpleNamespace
 from unittest.mock import AsyncMock, MagicMock
 
+import pytest
+
 from EvoScientist.cli import async_notifier
 from EvoScientist.cli.async_notifier import (
     dedup_notifications,
@@ -28,12 +30,6 @@ def test_notification_dataclass_fields():
 
 
 def test_notification_queue_is_module_level_fifo():
-    # Drain anything left over from other tests
-    while True:
-        try:
-            async_notifier._notification_queue.get_nowait()
-        except queue.Empty:
-            break
     n1 = async_notifier.AsyncTaskNotification("a", "x", "success", "")
     n2 = async_notifier.AsyncTaskNotification("b", "x", "success", "")
     async_notifier._notification_queue.put(n1)
@@ -51,7 +47,7 @@ def _drain_queue(q):
             return items
 
 
-def test_read_async_tasks_from_gateway_reads_state_values(run_async):
+async def test_read_async_tasks_from_gateway_reads_state_values():
     gateway = FakeGraphGateway(
         state_values={
             "async_tasks": {
@@ -60,18 +56,16 @@ def test_read_async_tasks_from_gateway_reads_state_values(run_async):
         }
     )
 
-    tasks = run_async(
-        async_notifier.read_async_tasks_from_gateway(
-            gateway,
-            GraphTarget(local_graph=MagicMock()),
-            "tid",
-        )
+    tasks = await async_notifier.read_async_tasks_from_gateway(
+        gateway,
+        GraphTarget(local_graph=MagicMock()),
+        "tid",
     )
 
     assert tasks == {"task-1": {"status": "success"}}
 
 
-def test_watcher_pushes_notification_on_stream_end(run_async):
+async def test_watcher_pushes_notification_on_stream_end():
     # Stream yields one "values" chunk with the final state, then closes
     final_state = {
         "messages": [{"type": "ai", "content": "Quantum superposition is..."}]
@@ -87,10 +81,7 @@ def test_watcher_pushes_notification_on_stream_end(run_async):
     # runs.get is used to fetch terminal status when stream ends
     client.runs.get = AsyncMock(return_value={"status": "success"})
 
-    _drain_all(async_notifier)
-    run_async(
-        async_notifier.watch_run_and_notify(client, "thr-1", "run-1", "writing-agent")
-    )
+    await async_notifier.watch_run_and_notify(client, "thr-1", "run-1", "writing-agent")
 
     notifs = _drain_queue(async_notifier._notification_queue)
     assert len(notifs) == 1
@@ -99,7 +90,7 @@ def test_watcher_pushes_notification_on_stream_end(run_async):
     assert notifs[0].status == "success"
 
 
-def test_watcher_pushes_error_status_on_stream_exception(run_async):
+async def test_watcher_pushes_error_status_on_stream_exception():
     async def fake_stream(*a, **kw):
         raise RuntimeError("network broken")
         yield  # unreachable; makes this an async generator
@@ -111,14 +102,13 @@ def test_watcher_pushes_error_status_on_stream_exception(run_async):
         return_value={"status": "error", "error": "network broken"}
     )
 
-    _drain_all(async_notifier)
-    run_async(async_notifier.watch_run_and_notify(client, "thr-4", "run-4", "agentZ"))
+    await async_notifier.watch_run_and_notify(client, "thr-4", "run-4", "agentZ")
 
     notif = async_notifier._notification_queue.get_nowait()
     assert notif.status == "error"
 
 
-def test_spawn_watcher_replaces_existing_for_same_thread(run_async):
+async def test_spawn_watcher_replaces_existing_for_same_thread():
     """A second spawn_watcher with the same thread_id cancels the old watcher
     and registers the new one — supports update_async_task creating a new
     run_id on the same thread_id."""
@@ -138,43 +128,34 @@ def test_spawn_watcher_replaces_existing_for_same_thread(run_async):
     client.runs.join_stream = fake_stream_long
     client.runs.get = AsyncMock(return_value={"status": "success"})
 
-    async def scenario():
-        # Clear all queues and the watcher registries
-        async_notifier._active_watchers.clear()
-        async_notifier._watcher_by_thread.clear()
-        _drain_all(async_notifier)
+    # First spawn for thread X, run R1
+    t1 = async_notifier.spawn_watcher(client, "thr-X", "R1", "agent")
+    assert t1 is not None
+    assert async_notifier._watcher_by_thread["thr-X"] is t1
+    await asyncio.sleep(0.02)  # let it start streaming
 
-        # First spawn for thread X, run R1
-        t1 = async_notifier.spawn_watcher(client, "thr-X", "R1", "agent")
-        assert t1 is not None
-        assert async_notifier._watcher_by_thread["thr-X"] is t1
-        await asyncio.sleep(0.02)  # let it start streaming
+    # Second spawn for SAME thread X, NEW run R2
+    t2 = async_notifier.spawn_watcher(client, "thr-X", "R2", "agent")
+    assert t2 is not None
+    assert t2 is not t1
+    assert async_notifier._watcher_by_thread["thr-X"] is t2
 
-        # Second spawn for SAME thread X, NEW run R2
-        t2 = async_notifier.spawn_watcher(client, "thr-X", "R2", "agent")
-        assert t2 is not None
-        assert t2 is not t1
-        assert async_notifier._watcher_by_thread["thr-X"] is t2
+    # Old watcher should be cancelled
+    await asyncio.sleep(0.02)
+    assert t1.cancelled() or t1.done()
 
-        # Old watcher should be cancelled
-        await asyncio.sleep(0.02)
-        assert t1.cancelled() or t1.done()
+    # Cleanup the new task too
+    t2.cancel()
+    try:
+        await t2
+    except asyncio.CancelledError:
+        pass
 
-        # Cleanup the new task too
-        t2.cancel()
-        try:
-            await t2
-        except asyncio.CancelledError:
-            pass
-
-        # Cancelled watchers don't push notifications
-        assert _drain_one_queue_helper(async_notifier._notification_queue) == []
-        assert _drain_one_queue_helper(async_notifier._unrouted_queue) == []
-        if hasattr(async_notifier, "_notifications_by_thread"):
-            for q in async_notifier._notifications_by_thread.values():
-                assert _drain_one_queue_helper(q) == []
-
-    run_async(scenario())
+    # Cancelled watchers don't push notifications
+    assert _drain_one_queue_helper(async_notifier._notification_queue) == []
+    assert _drain_one_queue_helper(async_notifier._unrouted_queue) == []
+    for q in async_notifier._notifications_by_thread.values():
+        assert _drain_one_queue_helper(q) == []
 
 
 # ============================================================================
@@ -324,13 +305,6 @@ def test_format_notification_lines_timeout_uses_warning_icon():
 
 def test_drain_returns_all_pending_and_empties_queue():
     """drain_notifications pulls every pending notification and empties queue."""
-    # Clear the queue first
-    while True:
-        try:
-            async_notifier._notification_queue.get_nowait()
-        except queue.Empty:
-            break
-
     # Add three notifications
     for tid in ("a", "b", "c"):
         async_notifier._notification_queue.put(
@@ -463,17 +437,12 @@ def test_dedup_preserves_order():
 # ============================================================================
 
 
-def test_consume_notifications_calls_runner_with_batched_message(run_async):
+async def test_consume_notifications_calls_runner_with_batched_message():
     """When notifications arrive and agent is idle, consume_notifications fires
     the supplied async runner once with the formatted batch message and notifs list."""
     from EvoScientist.cli import async_notifier as an
 
     # Set up two pending notifications, no dedup match
-    while True:
-        try:
-            an._notification_queue.get_nowait()
-        except queue.Empty:
-            break
     an._notification_queue.put(an.AsyncTaskNotification("t1", "wA", "success", "", ""))
     an._notification_queue.put(an.AsyncTaskNotification("t2", "wB", "success", "", ""))
 
@@ -486,21 +455,15 @@ def test_consume_notifications_calls_runner_with_batched_message(run_async):
     async def fake_state_reader() -> dict:
         return {}  # no dedup info
 
-    run_async(an.consume_notifications(fake_runner, fake_state_reader))
+    await an.consume_notifications(fake_runner, fake_state_reader)
     assert "wA" in captured["text"]
     assert "wB" in captured["text"]
     assert len(captured["notifs"]) == 2
 
 
-def test_consume_notifications_no_op_when_queue_empty(run_async):
+async def test_consume_notifications_no_op_when_queue_empty():
     from EvoScientist.cli import async_notifier as an
 
-    while True:
-        try:
-            an._notification_queue.get_nowait()
-        except queue.Empty:
-            break
-
     called = False
 
     async def fake_runner(text: str, notifs: list):
@@ -510,7 +473,7 @@ def test_consume_notifications_no_op_when_queue_empty(run_async):
     async def fake_state_reader():
         return {}
 
-    run_async(an.consume_notifications(fake_runner, fake_state_reader))
+    await an.consume_notifications(fake_runner, fake_state_reader)
     assert called is False
 
 
@@ -521,7 +484,7 @@ def test_consume_notifications_no_op_when_queue_empty(run_async):
 # ============================================================================
 
 
-def test_notification_consuming_flag_prevents_reentry(run_async):
+async def test_notification_consuming_flag_prevents_reentry():
     """The _notification_consuming guard prevents two overlapping consumers.
 
     Verifies the flag contract used by _consume_notifications_tui:
@@ -536,13 +499,6 @@ def test_notification_consuming_flag_prevents_reentry(run_async):
     """
     from EvoScientist.cli import async_notifier as an
 
-    # Clear the queue
-    while True:
-        try:
-            an._notification_queue.get_nowait()
-        except queue.Empty:
-            break
-
     state = {"inject_count": 0, "consuming": False}
 
     async def counting_runner(text: str, notifs: list) -> None:
@@ -565,45 +521,42 @@ def test_notification_consuming_flag_prevents_reentry(run_async):
     n1 = an.AsyncTaskNotification("g1", "writing-agent", "success", "", "")
     n2 = an.AsyncTaskNotification("g2", "data-agent", "success", "", "")
 
-    async def scenario():
-        # Scenario 1: normal flow — flag cleared, second consumer runs fine.
-        await guarded_consume(n1)
-        assert state["inject_count"] == 1
-        assert state["consuming"] is False  # finally ran
+    # Scenario 1: normal flow — flag cleared, second consumer runs fine.
+    await guarded_consume(n1)
+    assert state["inject_count"] == 1
+    assert state["consuming"] is False  # finally ran
 
-        state["inject_count"] = 0
-        await guarded_consume(n2)
-        assert state["inject_count"] == 1
-        assert state["consuming"] is False
+    state["inject_count"] = 0
+    await guarded_consume(n2)
+    assert state["inject_count"] == 1
+    assert state["consuming"] is False
 
-        # Scenario 2: flag pre-set (first consumer in-flight) → second bails.
-        state["inject_count"] = 0
-        state["consuming"] = True  # simulate first consumer running
-        an._notification_queue.put(n1)
-        await guarded_consume(n1)  # should be blocked immediately
-        assert state["inject_count"] == 0  # runner never called
-        state["consuming"] = False  # cleanup
+    # Scenario 2: flag pre-set (first consumer in-flight) → second bails.
+    state["inject_count"] = 0
+    state["consuming"] = True  # simulate first consumer running
+    an._notification_queue.put(n1)
+    await guarded_consume(n1)  # should be blocked immediately
+    assert state["inject_count"] == 0  # runner never called
+    state["consuming"] = False  # cleanup
 
-        # Scenario 3: exception in runner → flag still cleared by finally.
-        async def raising_runner(text: str, notifs: list) -> None:
-            raise RuntimeError("boom")
+    # Scenario 3: exception in runner → flag still cleared by finally.
+    async def raising_runner(text: str, notifs: list) -> None:
+        raise RuntimeError("boom")
 
-        async def guarded_consume_raising(notif):
-            if state["consuming"]:
-                return
-            state["consuming"] = True
-            try:
-                an._notification_queue.put(notif)
-                await an.consume_notifications(raising_runner, fake_state_reader)
-            except RuntimeError:
-                pass
-            finally:
-                state["consuming"] = False
+    async def guarded_consume_raising(notif):
+        if state["consuming"]:
+            return
+        state["consuming"] = True
+        try:
+            an._notification_queue.put(notif)
+            await an.consume_notifications(raising_runner, fake_state_reader)
+        except RuntimeError:
+            pass
+        finally:
+            state["consuming"] = False
 
-        await guarded_consume_raising(n2)
-        assert state["consuming"] is False  # cleared despite exception
-
-    run_async(scenario())
+    await guarded_consume_raising(n2)
+    assert state["consuming"] is False  # cleared despite exception
 
 
 # ============================================================================
@@ -613,33 +566,42 @@ def test_notification_consuming_flag_prevents_reentry(run_async):
 
 def _drain_all(an_mod):
     """Drain every queue (per-thread + unrouted) so tests start clean."""
-    if hasattr(an_mod, "_notification_queue"):
+    while True:
+        try:
+            an_mod._notification_queue.get_nowait()
+        except queue.Empty:
+            break
+    for q in list(an_mod._notifications_by_thread.values()):
         while True:
             try:
-                an_mod._notification_queue.get_nowait()
-            except queue.Empty:
-                break
-    if hasattr(an_mod, "_notifications_by_thread"):
-        for q in list(an_mod._notifications_by_thread.values()):
-            while True:
-                try:
-                    q.get_nowait()
-                except queue.Empty:
-                    break
-    if hasattr(an_mod, "_unrouted_queue"):
-        while True:
-            try:
-                an_mod._unrouted_queue.get_nowait()
+                q.get_nowait()
             except queue.Empty:
                 break
+    while True:
+        try:
+            an_mod._unrouted_queue.get_nowait()
+        except queue.Empty:
+            break
 
 
-def test_consume_only_drains_matching_thread(run_async):
+def _reset_notifier_state(an_mod):
+    _drain_all(an_mod)
+    an_mod._active_watchers.clear()
+    an_mod._watcher_by_thread.clear()
+
+
+@pytest.fixture(autouse=True)
+def _clean_async_notifier_state():
+    _reset_notifier_state(async_notifier)
+    yield
+    _reset_notifier_state(async_notifier)
+
+
+async def test_consume_only_drains_matching_thread():
     """Notifications tagged with origin_cli_thread_id only drain when the
     consumer is invoked with the matching current_thread_id."""
     from EvoScientist.cli import async_notifier as an
 
-    _drain_all(an)
     n_a = an.AsyncTaskNotification(
         "tA", "writing-agent", "success", "", "", origin_cli_thread_id="threadA"
     )
@@ -657,21 +619,17 @@ def test_consume_only_drains_matching_thread(run_async):
     async def state_reader() -> dict:
         return {}
 
-    run_async(
-        an.consume_notifications(runner, state_reader, current_thread_id="threadA")
-    )
+    await an.consume_notifications(runner, state_reader, current_thread_id="threadA")
     assert captured["runs"] == [["tA"]]
     # B's notification should still be queued
     assert an.has_pending_notifications("threadB")
-    _drain_all(an)
 
 
-def test_unrouted_notifications_drain_on_any_thread(run_async):
+async def test_unrouted_notifications_drain_on_any_thread():
     """Notifications without origin_cli_thread_id (legacy / direct put) drain
     regardless of the current_thread_id arg."""
     from EvoScientist.cli import async_notifier as an
 
-    _drain_all(an)
     an._notification_queue.put(
         an.AsyncTaskNotification("tU", "writing-agent", "success", "", "")
     )
@@ -684,19 +642,15 @@ def test_unrouted_notifications_drain_on_any_thread(run_async):
     async def state_reader() -> dict:
         return {}
 
-    run_async(
-        an.consume_notifications(runner, state_reader, current_thread_id="anything")
-    )
+    await an.consume_notifications(runner, state_reader, current_thread_id="anything")
     assert [n.task_id for n in captured["notifs"]] == ["tU"]
-    _drain_all(an)
 
 
-def test_thread_switch_drains_pending(run_async):
+async def test_thread_switch_drains_pending():
     """Pending notifications for thread B are not delivered while consumer
     asks for thread A; once consumer runs with thread B they drain."""
     from EvoScientist.cli import async_notifier as an
 
-    _drain_all(an)
     an._enqueue(
         an.AsyncTaskNotification(
             "tB", "writing-agent", "success", "", "", origin_cli_thread_id="threadB"
@@ -712,25 +666,19 @@ def test_thread_switch_drains_pending(run_async):
         return {}
 
     # First consume in thread A → no drain, B's notif still queued
-    run_async(
-        an.consume_notifications(runner, state_reader, current_thread_id="threadA")
-    )
+    await an.consume_notifications(runner, state_reader, current_thread_id="threadA")
     assert captured["runs"] == []
     assert an.has_pending_notifications("threadB")
 
     # Now switch to thread B → drains
-    run_async(
-        an.consume_notifications(runner, state_reader, current_thread_id="threadB")
-    )
+    await an.consume_notifications(runner, state_reader, current_thread_id="threadB")
     assert captured["runs"] == [["tB"]]
-    _drain_all(an)
 
 
 def test_has_pending_notifications_respects_routing():
     """has_pending_notifications returns true only for matching or unrouted."""
     from EvoScientist.cli import async_notifier as an
 
-    _drain_all(an)
     # Unrouted always counts
     an._notification_queue.put(
         an.AsyncTaskNotification("tU", "writing-agent", "success", "", "")
@@ -748,7 +696,6 @@ def test_has_pending_notifications_respects_routing():
     assert an.has_pending_notifications("threadA") is True
     assert an.has_pending_notifications("threadB") is False
     assert an.has_pending_notifications() is False  # no unrouted, no current_thread
-    _drain_all(an)
 
 
 # ============================================================================
@@ -762,7 +709,7 @@ def test_has_pending_notifications_respects_routing():
 # ============================================================================
 
 
-def test_watcher_reports_error_on_in_band_error_event(run_async):
+async def test_watcher_reports_error_on_in_band_error_event():
     """SSE error event in the stream → notification.status == 'error'."""
 
     async def fake_stream(*a, **kw):
@@ -777,8 +724,7 @@ def test_watcher_reports_error_on_in_band_error_event(run_async):
         return_value={"status": "success"}
     )  # would mislead — should NOT be consulted
 
-    _drain_all(async_notifier)
-    run_async(async_notifier.watch_run_and_notify(client, "thrE", "rE", "agentE"))
+    await async_notifier.watch_run_and_notify(client, "thrE", "rE", "agentE")
 
     notif = async_notifier._notification_queue.get_nowait()
     assert notif.status == "error"
@@ -786,7 +732,7 @@ def test_watcher_reports_error_on_in_band_error_event(run_async):
     client.runs.get.assert_not_awaited()
 
 
-def test_watcher_clean_exit_with_runs_get_success_is_success(run_async):
+async def test_watcher_clean_exit_with_runs_get_success_is_success():
     """Clean stream exit + runs.get reports success → status=success."""
 
     async def fake_stream(*a, **kw):
@@ -798,15 +744,14 @@ def test_watcher_clean_exit_with_runs_get_success_is_success(run_async):
     client.runs.join_stream = fake_stream
     client.runs.get = AsyncMock(return_value={"status": "success"})
 
-    _drain_all(async_notifier)
-    run_async(async_notifier.watch_run_and_notify(client, "thrS", "rS", "agentS"))
+    await async_notifier.watch_run_and_notify(client, "thrS", "rS", "agentS")
 
     notif = async_notifier._notification_queue.get_nowait()
     assert notif.status == "success"
     client.runs.get.assert_awaited_once()
 
 
-def test_watcher_clean_exit_with_runs_get_error_is_race_safe(run_async):
+async def test_watcher_clean_exit_with_runs_get_error_is_race_safe():
     """Clean stream exit + no in-band error event + runs.get returns 'error'
     → status=success (race-safe).
 
@@ -827,14 +772,13 @@ def test_watcher_clean_exit_with_runs_get_error_is_race_safe(run_async):
     client.runs.join_stream = fake_stream
     client.runs.get = AsyncMock(return_value={"status": "error"})
 
-    _drain_all(async_notifier)
-    run_async(async_notifier.watch_run_and_notify(client, "thrS", "rS", "agentS"))
+    await async_notifier.watch_run_and_notify(client, "thrS", "rS", "agentS")
 
     notif = async_notifier._notification_queue.get_nowait()
     assert notif.status == "success"
 
 
-def test_watcher_clean_exit_with_runs_get_running_drops_notification(run_async):
+async def test_watcher_clean_exit_with_runs_get_running_drops_notification():
     """Reproduces the production bug: clean SSE close while run is still
     actually running (HTTP keep-alive timeout under concurrency).
 
@@ -856,11 +800,8 @@ def test_watcher_clean_exit_with_runs_get_running_drops_notification(run_async):
     client.runs.join_stream = fake_stream
     client.runs.get = AsyncMock(return_value={"status": "running"})
 
-    _drain_all(async_notifier)
-    run_async(
-        async_notifier.watch_run_and_notify(
-            client, "thr-bug", "rB", "data-analysis-agent"
-        )
+    await async_notifier.watch_run_and_notify(
+        client, "thr-bug", "rB", "data-analysis-agent"
     )
 
     # No notification should have been enqueued anywhere.
@@ -872,7 +813,7 @@ def test_watcher_clean_exit_with_runs_get_running_drops_notification(run_async):
     assert client.runs.get.await_count >= 1
 
 
-def test_watcher_unknown_status_treated_as_non_terminal(run_async):
+async def test_watcher_unknown_status_treated_as_non_terminal():
     """Future / unrecognized status values should trigger a re-join, not a
     false-positive notification.
 
@@ -893,8 +834,7 @@ def test_watcher_unknown_status_treated_as_non_terminal(run_async):
         side_effect=[{"status": "queued"}, {"status": "success"}]
     )
 
-    _drain_all(async_notifier)
-    run_async(async_notifier.watch_run_and_notify(client, "thrU", "rU", "agentU"))
+    await async_notifier.watch_run_and_notify(client, "thrU", "rU", "agentU")
 
     notif = async_notifier._notification_queue.get_nowait()
     assert notif.status == "success"
@@ -902,7 +842,7 @@ def test_watcher_unknown_status_treated_as_non_terminal(run_async):
     assert client.runs.get.await_count == 2
 
 
-def test_watcher_runs_get_persistent_failure_drops_notification(run_async, monkeypatch):
+async def test_watcher_runs_get_persistent_failure_drops_notification(monkeypatch):
     """If ``runs.get`` keeps raising, the watcher cannot verify terminal
     state and MUST drop the notification rather than default to
     ``"success"`` — otherwise a transient server outage reintroduces the
@@ -921,22 +861,20 @@ def test_watcher_runs_get_persistent_failure_drops_notification(run_async, monke
 
     monkeypatch.setattr(async_notifier.asyncio, "sleep", _no_sleep)
 
-    _drain_all(async_notifier)
-    run_async(async_notifier.watch_run_and_notify(client, "thrG", "rG", "agentG"))
+    await async_notifier.watch_run_and_notify(client, "thrG", "rG", "agentG")
 
     # No notification — watcher exhausted the reconnect budget. Check every
     # queue routing could send to so a future routing change can't make this
     # test silently false-pass.
     assert _drain_one_queue_helper(async_notifier._unrouted_queue) == []
     assert _drain_one_queue_helper(async_notifier._notification_queue) == []
-    if hasattr(async_notifier, "_notifications_by_thread"):
-        for q in async_notifier._notifications_by_thread.values():
-            assert _drain_one_queue_helper(q) == []
+    for q in async_notifier._notifications_by_thread.values():
+        assert _drain_one_queue_helper(q) == []
     # 1 initial + _MAX_RECONNECT_ATTEMPTS retries = 11 calls total.
     assert client.runs.get.await_count == async_notifier._MAX_RECONNECT_ATTEMPTS + 1
 
 
-def test_watcher_runs_get_transient_failure_recovers(run_async, monkeypatch):
+async def test_watcher_runs_get_transient_failure_recovers(monkeypatch):
     """A single ``runs.get`` failure followed by a successful response on
     retry must produce a correct notification — verifies the bounded
     retry path actually recovers from transient outages instead of just
@@ -957,15 +895,14 @@ def test_watcher_runs_get_transient_failure_recovers(run_async, monkeypatch):
 
     monkeypatch.setattr(async_notifier.asyncio, "sleep", _no_sleep)
 
-    _drain_all(async_notifier)
-    run_async(async_notifier.watch_run_and_notify(client, "thrT", "rT", "agentT"))
+    await async_notifier.watch_run_and_notify(client, "thrT", "rT", "agentT")
 
     notif = async_notifier._notification_queue.get_nowait()
     assert notif.status == "success"
     assert client.runs.get.await_count == 2
 
 
-def test_watcher_re_joins_stream_until_terminal_status(run_async):
+async def test_watcher_re_joins_stream_until_terminal_status():
     """When runs.get returns 'running' on attempt N but a terminal status
     on attempt N+1, the watcher re-joins, observes the terminal status,
     and enqueues the notification correctly."""
@@ -980,8 +917,7 @@ def test_watcher_re_joins_stream_until_terminal_status(run_async):
         side_effect=[{"status": "running"}, {"status": "success"}]
     )
 
-    _drain_all(async_notifier)
-    run_async(async_notifier.watch_run_and_notify(client, "thrR", "rR", "agentR"))
+    await async_notifier.watch_run_and_notify(client, "thrR", "rR", "agentR")
 
     notif = async_notifier._notification_queue.get_nowait()
     assert notif.status == "success"
@@ -995,15 +931,12 @@ def test_watcher_re_joins_stream_until_terminal_status(run_async):
 # ============================================================================
 
 
-def test_consume_notifications_propagates_inject_exception(run_async):
+async def test_consume_notifications_propagates_inject_exception():
     """If the run_message callback raises, consume_notifications propagates
     the exception to the caller — pollers wrap it in try/except so the
     poller task does not die."""
-    import pytest
-
     from EvoScientist.cli import async_notifier as an
 
-    _drain_all(an)
     an._notification_queue.put(
         an.AsyncTaskNotification("tX", "writing-agent", "success", "", "")
     )
@@ -1015,11 +948,10 @@ def test_consume_notifications_propagates_inject_exception(run_async):
         return {}
 
     with pytest.raises(RuntimeError, match="kaboom"):
-        run_async(an.consume_notifications(boom_runner, state_reader))
-    _drain_all(an)
+        await an.consume_notifications(boom_runner, state_reader)
 
 
-def test_watcher_skips_notification_on_stream_fail_with_nonterminal_status(run_async):
+async def test_watcher_skips_notification_on_stream_fail_with_nonterminal_status():
     """When the SSE stream errors AND runs.get returns a non-terminal status
     (e.g. ``pending`` because the run is still alive), the watcher must
     NOT enqueue a notification — otherwise the user sees a confusing
@@ -1035,15 +967,13 @@ def test_watcher_skips_notification_on_stream_fail_with_nonterminal_status(run_a
     client.runs.join_stream = fake_stream
     client.runs.get = AsyncMock(return_value={"status": "pending"})
 
-    _drain_all(async_notifier)
-    run_async(async_notifier.watch_run_and_notify(client, "thrP", "rP", "agentP"))
+    await async_notifier.watch_run_and_notify(client, "thrP", "rP", "agentP")
 
     # No notification should have been enqueued in any queue.
     assert _drain_one_queue_helper(async_notifier._unrouted_queue) == []
     assert _drain_one_queue_helper(async_notifier._notification_queue) == []
-    if hasattr(async_notifier, "_notifications_by_thread"):
-        for q in async_notifier._notifications_by_thread.values():
-            assert _drain_one_queue_helper(q) == []
+    for q in async_notifier._notifications_by_thread.values():
+        assert _drain_one_queue_helper(q) == []
 
 
 def _drain_one_queue_helper(q):
@@ -1060,8 +990,6 @@ def test_active_watchers_grace_filters_by_thread():
     (otherwise consume_notifications grace period would block thread A by up
     to 3s waiting for thread B's unrelated watchers to finish)."""
 
-    async_notifier._active_watchers.clear()
-
     # Sentinel handles — only their identity matters here, not their type
     handle_a = object()
     handle_b = object()

tests/test_async_watcher_middleware.py

@@ -7,7 +7,6 @@ deepagents internals. It hooks into ``awrap_tool_call`` and only fires on
 
 from __future__ import annotations
 
-import asyncio
 from types import SimpleNamespace
 from unittest.mock import MagicMock, patch
 
@@ -105,7 +104,7 @@ def _make_middleware():
     return mw, fake_client
 
 
-def test_middleware_spawns_watcher_on_start_async_task():
+async def test_middleware_spawns_watcher_on_start_async_task():
     """A successful start_async_task tool call must spawn one watcher per task."""
     from langgraph.types import Command
 
@@ -142,7 +141,7 @@ def test_middleware_spawns_watcher_on_start_async_task():
         )
 
     with patch.object(async_notifier, "spawn_watcher", side_effect=fake_spawn):
-        result = asyncio.run(mw.awrap_tool_call(request, fake_handler))
+        result = await mw.awrap_tool_call(request, fake_handler)
 
     assert isinstance(result, Command)
     assert spawn_calls == [
@@ -150,7 +149,7 @@ def test_middleware_spawns_watcher_on_start_async_task():
     ]
 
 
-def test_middleware_spawns_watcher_on_update_async_task():
+async def test_middleware_spawns_watcher_on_update_async_task():
     """A successful update_async_task call must also spawn a (replacement) watcher."""
     from langgraph.types import Command
 
@@ -183,7 +182,7 @@ def test_middleware_spawns_watcher_on_update_async_task():
         )
 
     with patch.object(async_notifier, "spawn_watcher", side_effect=fake_spawn):
-        asyncio.run(mw.awrap_tool_call(request, fake_handler))
+        await mw.awrap_tool_call(request, fake_handler)
 
     assert len(spawn_calls) == 1
     args, kwargs = spawn_calls[0]
@@ -195,7 +194,7 @@ def test_middleware_spawns_watcher_on_update_async_task():
     assert kwargs["origin_cli_thread_id"] == "cli-thread-A"
 
 
-def test_middleware_pre_cancels_old_watcher_on_update():
+async def test_middleware_pre_cancels_old_watcher_on_update():
     """update_async_task must cancel the existing watcher BEFORE invoking the handler.
 
     Otherwise the new run interrupts the old run's stream, which closes
@@ -221,14 +220,14 @@ def test_middleware_pre_cancels_old_watcher_on_update():
 
     try:
         with patch.object(async_notifier, "spawn_watcher"):
-            asyncio.run(mw.awrap_tool_call(request, fake_handler))
+            await mw.awrap_tool_call(request, fake_handler)
     finally:
         async_notifier._watcher_by_thread.pop("task-1", None)
 
     assert cancel_observed_before_handler["value"] is True
 
 
-def test_middleware_passes_through_unrelated_tools():
+async def test_middleware_passes_through_unrelated_tools():
     """A non-launch tool call must not spawn any watcher and must return result unchanged."""
     mw, _ = _make_middleware()
 
@@ -240,13 +239,13 @@ def test_middleware_passes_through_unrelated_tools():
     request = _build_request("ls", {"path": "/"}, thread_id="t")
 
     with patch.object(async_notifier, "spawn_watcher") as mock_spawn:
-        result = asyncio.run(mw.awrap_tool_call(request, fake_handler))
+        result = await mw.awrap_tool_call(request, fake_handler)
 
     assert result is sentinel
     assert mock_spawn.call_count == 0
 
 
-def test_middleware_handles_non_command_results_gracefully():
+async def test_middleware_handles_non_command_results_gracefully():
     """If the launch tool returns a string (validation error), no watcher is spawned."""
     mw, _ = _make_middleware()
 
@@ -260,13 +259,13 @@ def test_middleware_handles_non_command_results_gracefully():
     )
 
     with patch.object(async_notifier, "spawn_watcher") as mock_spawn:
-        result = asyncio.run(mw.awrap_tool_call(request, fake_handler))
+        result = await mw.awrap_tool_call(request, fake_handler)
 
     assert result == "Unknown async subagent type `bogus`"
     assert mock_spawn.call_count == 0
 
 
-def test_middleware_origin_thread_id_is_none_when_runtime_config_missing():
+async def test_middleware_origin_thread_id_is_none_when_runtime_config_missing():
     """When runtime.config is empty, origin_cli_thread_id must be None (not crash)."""
     from langgraph.types import Command
 
@@ -299,12 +298,12 @@ def test_middleware_origin_thread_id_is_none_when_runtime_config_missing():
         )
 
     with patch.object(async_notifier, "spawn_watcher", side_effect=fake_spawn):
-        asyncio.run(mw.awrap_tool_call(request, fake_handler))
+        await mw.awrap_tool_call(request, fake_handler)
 
     assert captured.get("origin_cli_thread_id") is None
 
 
-def test_middleware_swallows_spawn_exceptions():
+async def test_middleware_swallows_spawn_exceptions():
     """spawn_watcher errors must not propagate up — middleware logs and continues."""
     from langgraph.types import Command
 
@@ -336,7 +335,7 @@ def test_middleware_swallows_spawn_exceptions():
 
     with patch.object(async_notifier, "spawn_watcher", side_effect=boom):
         # Should not raise.
-        result = asyncio.run(mw.awrap_tool_call(request, fake_handler))
+        result = await mw.awrap_tool_call(request, fake_handler)
 
     assert isinstance(result, Command)
 
@@ -356,7 +355,9 @@ def test_middleware_swallows_spawn_exceptions():
         ),
     ],
 )
-def test_middleware_picks_correct_prompt_field_per_tool(tool_name, args, prompt_field):
+async def test_middleware_picks_correct_prompt_field_per_tool(
+    tool_name, args, prompt_field
+):
     """start_async_task uses 'description'; update_async_task uses 'message'."""
     from langgraph.types import Command
 
@@ -385,12 +386,12 @@ def test_middleware_picks_correct_prompt_field_per_tool(tool_name, args, prompt_
     request = _build_request(tool_name, args, thread_id="t")
 
     with patch.object(async_notifier, "spawn_watcher", side_effect=fake_spawn):
-        asyncio.run(mw.awrap_tool_call(request, fake_handler))
+        await mw.awrap_tool_call(request, fake_handler)
 
     assert captured_prompt["value"] == prompt_field
 
 
-def test_middleware_prompt_field_is_tool_name_gated_not_fallback_chained():
+async def test_middleware_prompt_field_is_tool_name_gated_not_fallback_chained():
     """update_async_task with extra `description` arg must still use `message`.
 
     Guards against the previous `args.get('description') or args.get('message')`
@@ -432,12 +433,12 @@ def test_middleware_prompt_field_is_tool_name_gated_not_fallback_chained():
     )
 
     with patch.object(async_notifier, "spawn_watcher", side_effect=fake_spawn):
-        asyncio.run(mw.awrap_tool_call(request, fake_handler))
+        await mw.awrap_tool_call(request, fake_handler)
 
     assert captured_prompt["value"] == "use this"
 
 
-def test_middleware_pre_cancel_swallows_unexpected_errors():
+async def test_middleware_pre_cancel_swallows_unexpected_errors():
     """A faulty old-watcher handle must not block the handler from running."""
     from langgraph.types import Command
 
@@ -460,7 +461,7 @@ def test_middleware_pre_cancel_swallows_unexpected_errors():
     try:
         with patch.object(async_notifier, "spawn_watcher"):
             # Should not raise.
-            asyncio.run(mw.awrap_tool_call(request, fake_handler))
+            await mw.awrap_tool_call(request, fake_handler)
     finally:
         async_notifier._watcher_by_thread.pop("t1", None)
 

tests/test_autoskills.py

@@ -1,6 +1,5 @@
 from __future__ import annotations
 
-import asyncio
 import json
 from types import SimpleNamespace
 
@@ -917,16 +916,16 @@ class _AsyncFakeCrons:
         return [{"cron_id": "cron-async"}]
 
 
-def test_alist_autoskill_schedules_uses_async_client_and_explicit_limit(monkeypatch):
+async def test_alist_autoskill_schedules_uses_async_client_and_explicit_limit(
+    monkeypatch,
+):
     crons = _AsyncFakeCrons()
     client = SimpleNamespace(crons=crons)
     monkeypatch.setattr("langgraph_sdk.get_client", lambda **_kwargs: client)
 
-    rows = asyncio.run(
-        alist_autoskill_schedules(
-            EvoScientistConfig(),
-            limit=3,
-        )
+    rows = await alist_autoskill_schedules(
+        EvoScientistConfig(),
+        limit=3,
     )
 
     assert rows == [{"cron_id": "cron-async"}]

tests/test_background.py

@@ -91,7 +91,7 @@ def test_stop_already_finished_is_graceful(tmp_path):
     assert "already finished" in bg.stop(pid)
 
 
-def test_exited_elapsed_is_frozen(tmp_path):
+def test_exited_elapsed_is_frozen(tmp_path, monkeypatch):
     """Elapsed for an exited process freezes at its runtime, it must not keep growing."""
     pid = bg.launch(_true_cmd(), str(tmp_path))
     assert _wait_until(lambda: bg._PROCESSES[pid].finished_ts is not None)
@@ -99,7 +99,7 @@ def test_exited_elapsed_is_frozen(tmp_path):
     proc = bg._PROCESSES[pid]
     assert proc.finished_ts is not None
     first = bg._elapsed(proc)
-    time.sleep(1.1)  # intentional: prove elapsed stays frozen, not ticking up
+    monkeypatch.setattr(bg.time, "time", lambda: proc.finished_ts + 100.0)
     assert bg._elapsed(proc) == first
 
 

tests/test_bus_integration.py

@@ -12,7 +12,6 @@ import pytest
 from EvoScientist.channels.bus.events import InboundMessage
 from EvoScientist.channels.bus.message_bus import MessageBus
 from EvoScientist.channels.channel_manager import ChannelManager
-from tests.conftest import run_async as _run
 from tests.fakes import QueueFakeChannel as FakeChannel
 
 
@@ -58,7 +57,7 @@ def clean_channel_state():
 class TestBusInboundConsumer:
     """Test the _bus_inbound_consumer queue bridge."""
 
-    def test_processes_inbound_and_publishes_outbound(self):
+    async def test_processes_inbound_and_publishes_outbound(self):
         """InboundMessage -> queue -> response -> OutboundMessage flow."""
         from EvoScientist.cli.channel import (
             _bus_inbound_consumer,
@@ -68,54 +67,51 @@ class TestBusInboundConsumer:
 
         _drain_queue(_message_queue)
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
+        consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
 
-            await bus.publish_inbound(
-                InboundMessage(
-                    channel="fake",
-                    sender_id="user1",
-                    chat_id="chat1",
-                    content="hello agent",
-                )
+        await bus.publish_inbound(
+            InboundMessage(
+                channel="fake",
+                sender_id="user1",
+                chat_id="chat1",
+                content="hello agent",
             )
+        )
 
-            # Wait for consumer to enqueue the message
-            for _ in range(20):
-                if not _message_queue.empty():
-                    break
-                await asyncio.sleep(0.05)
+        # Wait for consumer to enqueue the message
+        for _ in range(20):
+            if not _message_queue.empty():
+                break
+            await asyncio.sleep(0.05)
 
-            msg = _message_queue.get_nowait()
-            assert msg.content == "hello agent"
-            assert msg.sender == "user1"
-            assert msg.channel_type == "fake"
+        msg = _message_queue.get_nowait()
+        assert msg.content == "hello agent"
+        assert msg.sender == "user1"
+        assert msg.channel_type == "fake"
 
-            # Simulate main-thread response
-            _set_channel_response(msg.msg_id, "Reply to: hello agent")
+        # Simulate main-thread response
+        _set_channel_response(msg.msg_id, "Reply to: hello agent")
 
-            outbound = await asyncio.wait_for(
-                bus.consume_outbound(),
-                timeout=2.0,
-            )
-            assert outbound.channel == "fake"
-            assert outbound.chat_id == "chat1"
-            assert "Reply to: hello agent" in outbound.content
+        outbound = await asyncio.wait_for(
+            bus.consume_outbound(),
+            timeout=2.0,
+        )
+        assert outbound.channel == "fake"
+        assert outbound.chat_id == "chat1"
+        assert "Reply to: hello agent" in outbound.content
 
-            consumer.cancel()
-            try:
-                await consumer
-            except asyncio.CancelledError:
-                pass
+        consumer.cancel()
+        try:
+            await consumer
+        except asyncio.CancelledError:
+            pass
 
-        _run(_test())
-
-    def test_no_response_fallback(self):
+    async def test_no_response_fallback(self):
         """Empty response is replaced with 'No response' fallback."""
         from EvoScientist.cli.channel import (
             _bus_inbound_consumer,
@@ -125,47 +121,44 @@ class TestBusInboundConsumer:
 
         _drain_queue(_message_queue)
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
+        consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
 
-            await bus.publish_inbound(
-                InboundMessage(
-                    channel="fake",
-                    sender_id="user1",
-                    chat_id="chat1",
-                    content="test",
-                )
+        await bus.publish_inbound(
+            InboundMessage(
+                channel="fake",
+                sender_id="user1",
+                chat_id="chat1",
+                content="test",
             )
+        )
 
-            for _ in range(20):
-                if not _message_queue.empty():
-                    break
-                await asyncio.sleep(0.05)
+        for _ in range(20):
+            if not _message_queue.empty():
+                break
+            await asyncio.sleep(0.05)
 
-            msg = _message_queue.get_nowait()
-            # Set empty response — falsy, so consumer falls back to "No response"
-            _set_channel_response(msg.msg_id, "")
+        msg = _message_queue.get_nowait()
+        # Set empty response — falsy, so consumer falls back to "No response"
+        _set_channel_response(msg.msg_id, "")
 
-            outbound = await asyncio.wait_for(
-                bus.consume_outbound(),
-                timeout=2.0,
-            )
-            assert outbound.content == "No response"
+        outbound = await asyncio.wait_for(
+            bus.consume_outbound(),
+            timeout=2.0,
+        )
+        assert outbound.content == "No response"
 
-            consumer.cancel()
-            try:
-                await consumer
-            except asyncio.CancelledError:
-                pass
+        consumer.cancel()
+        try:
+            await consumer
+        except asyncio.CancelledError:
+            pass
 
-        _run(_test())
-
-    def test_late_response_after_timeout_still_publishes(self, monkeypatch):
+    async def test_late_response_after_timeout_still_publishes(self, monkeypatch):
         """A response that arrives after the bridge timeout is still forwarded."""
         from EvoScientist.cli import channel as channel_mod
         from EvoScientist.cli.channel import (
@@ -179,56 +172,53 @@ class TestBusInboundConsumer:
 
         _drain_queue(_message_queue)
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
+        consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
 
-            await bus.publish_inbound(
-                InboundMessage(
-                    channel="fake",
-                    sender_id="user1",
-                    chat_id="chat1",
-                    content="slow request",
-                    message_id="msg-123",
-                )
+        await bus.publish_inbound(
+            InboundMessage(
+                channel="fake",
+                sender_id="user1",
+                chat_id="chat1",
+                content="slow request",
+                message_id="msg-123",
             )
+        )
 
-            for _ in range(20):
-                if not _message_queue.empty():
-                    break
-                await asyncio.sleep(0.05)
+        for _ in range(20):
+            if not _message_queue.empty():
+                break
+            await asyncio.sleep(0.05)
 
-            msg = _message_queue.get_nowait()
+        msg = _message_queue.get_nowait()
 
-            notice = await asyncio.wait_for(
-                bus.consume_outbound(),
-                timeout=1.0,
-            )
-            assert "Still working on it" in notice.content
-            assert notice.reply_to == "msg-123"
+        notice = await asyncio.wait_for(
+            bus.consume_outbound(),
+            timeout=1.0,
+        )
+        assert "Still working on it" in notice.content
+        assert notice.reply_to == "msg-123"
 
-            _set_channel_response(msg.msg_id, "final answer")
+        _set_channel_response(msg.msg_id, "final answer")
 
-            outbound = await asyncio.wait_for(
-                bus.consume_outbound(),
-                timeout=1.0,
-            )
-            assert outbound.content == "final answer"
-            assert outbound.reply_to == "msg-123"
+        outbound = await asyncio.wait_for(
+            bus.consume_outbound(),
+            timeout=1.0,
+        )
+        assert outbound.content == "final answer"
+        assert outbound.reply_to == "msg-123"
 
-            consumer.cancel()
-            try:
-                await consumer
-            except asyncio.CancelledError:
-                pass
+        consumer.cancel()
+        try:
+            await consumer
+        except asyncio.CancelledError:
+            pass
 
-        _run(_test())
-
-    def test_late_timeout_keeps_active_request_cancellable(self, monkeypatch):
+    async def test_late_timeout_keeps_active_request_cancellable(self, monkeypatch):
         """Late timeout must not discard an active request's cancel scope."""
         from EvoScientist.cli import channel as channel_mod
         from EvoScientist.cli.channel import (
@@ -243,191 +233,179 @@ class TestBusInboundConsumer:
         monkeypatch.setattr(channel_mod, "_RESPONSE_TIMEOUT", 0.05)
         monkeypatch.setattr(channel_mod, "_LATE_RESPONSE_TIMEOUT", 0.05)
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            task = asyncio.create_task(
-                _handle_bus_message(
-                    bus,
-                    manager,
-                    InboundMessage(
-                        channel="fake",
-                        sender_id="user1",
-                        chat_id="chat1",
-                        content="still running",
-                        message_id="msg-active",
-                    ),
-                )
-            )
-
-            queued = None
-            for _ in range(20):
-                with _message_queue.mutex:
-                    queued = _message_queue.queue[0] if _message_queue.queue else None
-                if queued is not None:
-                    break
-                await asyncio.sleep(0.05)
-
-            assert queued is not None
-            assert _claim_channel_request(queued) is True
-
-            notice = await asyncio.wait_for(bus.consume_outbound(), timeout=1.0)
-            assert "Still working on it" in notice.content
-
-            await task
-
-            assert _channel_request_state(queued.msg_id) == "active"
-            cancel_scope = _channel_message_cancel_scope(queued)
-            assert not display_mod.is_stream_cancel_requested(cancel_scope)
-
-            await _handle_bus_message(
+        task = asyncio.create_task(
+            _handle_bus_message(
                 bus,
                 manager,
                 InboundMessage(
                     channel="fake",
                     sender_id="user1",
                     chat_id="chat1",
-                    content="/stop",
-                    message_id="msg-stop-active",
+                    content="still running",
+                    message_id="msg-active",
                 ),
             )
+        )
 
-            ack = await asyncio.wait_for(bus.consume_outbound(), timeout=1.0)
-            assert ack.content == "Stopped."
-            assert ack.reply_to == "msg-stop-active"
-            assert display_mod.is_stream_cancel_requested(cancel_scope)
+        queued = None
+        for _ in range(20):
+            with _message_queue.mutex:
+                queued = _message_queue.queue[0] if _message_queue.queue else None
+            if queued is not None:
+                break
+            await asyncio.sleep(0.05)
 
-        _run(_test())
+        assert queued is not None
+        assert _claim_channel_request(queued) is True
 
-    def test_cancelled_wait_cleans_pending_response(self):
+        notice = await asyncio.wait_for(bus.consume_outbound(), timeout=1.0)
+        assert "Still working on it" in notice.content
+
+        await task
+
+        assert _channel_request_state(queued.msg_id) == "active"
+        cancel_scope = _channel_message_cancel_scope(queued)
+        assert not display_mod.is_stream_cancel_requested(cancel_scope)
+
+        await _handle_bus_message(
+            bus,
+            manager,
+            InboundMessage(
+                channel="fake",
+                sender_id="user1",
+                chat_id="chat1",
+                content="/stop",
+                message_id="msg-stop-active",
+            ),
+        )
+
+        ack = await asyncio.wait_for(bus.consume_outbound(), timeout=1.0)
+        assert ack.content == "Stopped."
+        assert ack.reply_to == "msg-stop-active"
+        assert display_mod.is_stream_cancel_requested(cancel_scope)
+
+    async def test_cancelled_wait_cleans_pending_response(self):
         """Cancelling a pending bus message should not leak its response slot."""
         from EvoScientist.cli import channel as channel_mod
         from EvoScientist.cli.channel import _handle_bus_message, _message_queue
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            task = asyncio.create_task(
-                _handle_bus_message(
-                    bus,
-                    manager,
-                    InboundMessage(
-                        channel="fake",
-                        sender_id="user1",
-                        chat_id="chat1",
-                        content="cancel me",
-                    ),
-                )
+        task = asyncio.create_task(
+            _handle_bus_message(
+                bus,
+                manager,
+                InboundMessage(
+                    channel="fake",
+                    sender_id="user1",
+                    chat_id="chat1",
+                    content="cancel me",
+                ),
             )
+        )
 
-            for _ in range(20):
-                if not _message_queue.empty():
-                    break
-                await asyncio.sleep(0.05)
+        for _ in range(20):
+            if not _message_queue.empty():
+                break
+            await asyncio.sleep(0.05)
 
-            queued = _message_queue.get_nowait()
-            with channel_mod._response_lock:
-                assert queued.msg_id in channel_mod._pending_responses
+        queued = _message_queue.get_nowait()
+        with channel_mod._response_lock:
+            assert queued.msg_id in channel_mod._pending_responses
 
-            task.cancel()
-            with pytest.raises(asyncio.CancelledError):
-                await task
+        task.cancel()
+        with pytest.raises(asyncio.CancelledError):
+            await task
 
-            with channel_mod._response_lock:
-                assert queued.msg_id not in channel_mod._pending_responses
+        with channel_mod._response_lock:
+            assert queued.msg_id not in channel_mod._pending_responses
 
-        _run(_test())
-
-    def test_consumer_shutdown_cleans_pending_response(self):
+    async def test_consumer_shutdown_cleans_pending_response(self):
         """Stopping the consumer should cancel late waits and clear state."""
         from EvoScientist.cli import channel as channel_mod
         from EvoScientist.cli.channel import _bus_inbound_consumer, _message_queue
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
+        consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
 
-            await bus.publish_inbound(
-                InboundMessage(
-                    channel="fake",
-                    sender_id="user1",
-                    chat_id="chat1",
-                    content="slow shutdown",
-                )
+        await bus.publish_inbound(
+            InboundMessage(
+                channel="fake",
+                sender_id="user1",
+                chat_id="chat1",
+                content="slow shutdown",
             )
+        )
 
-            for _ in range(20):
-                if not _message_queue.empty():
-                    break
-                await asyncio.sleep(0.05)
+        for _ in range(20):
+            if not _message_queue.empty():
+                break
+            await asyncio.sleep(0.05)
 
-            queued = _message_queue.get_nowait()
-            with channel_mod._response_lock:
-                assert queued.msg_id in channel_mod._pending_responses
+        queued = _message_queue.get_nowait()
+        with channel_mod._response_lock:
+            assert queued.msg_id in channel_mod._pending_responses
 
-            consumer.cancel()
-            await consumer
+        consumer.cancel()
+        await consumer
 
-            with channel_mod._response_lock:
-                assert queued.msg_id not in channel_mod._pending_responses
+        with channel_mod._response_lock:
+            assert queued.msg_id not in channel_mod._pending_responses
 
-        _run(_test())
-
-    def test_stop_during_hitl_wait_releases_wait_and_acks(self):
+    async def test_stop_during_hitl_wait_releases_wait_and_acks(self):
         """`/stop` should wake pending HITL wait and publish immediate ack."""
         from EvoScientist.cli import channel as channel_mod
         from EvoScientist.cli.channel import _bus_inbound_consumer, _message_queue
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            hitl_event = channel_mod._register_hitl_wait("fake", "chat1")
-            consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
+        hitl_event = channel_mod._register_hitl_wait("fake", "chat1")
+        consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
 
-            await bus.publish_inbound(
-                InboundMessage(
-                    channel="fake",
-                    sender_id="user1",
-                    chat_id="chat1",
-                    content="/stop",
-                    message_id="m-stop-1",
-                )
+        await bus.publish_inbound(
+            InboundMessage(
+                channel="fake",
+                sender_id="user1",
+                chat_id="chat1",
+                content="/stop",
+                message_id="m-stop-1",
             )
+        )
 
-            for _ in range(20):
-                if hitl_event.is_set():
-                    break
-                await asyncio.sleep(0.05)
-            assert hitl_event.is_set()
-            assert channel_mod._pop_hitl_reply("fake", "chat1") == "/stop"
+        for _ in range(20):
+            if hitl_event.is_set():
+                break
+            await asyncio.sleep(0.05)
+        assert hitl_event.is_set()
+        assert channel_mod._pop_hitl_reply("fake", "chat1") == "/stop"
 
-            outbound = await asyncio.wait_for(bus.consume_outbound(), timeout=2.0)
-            assert outbound.content == "Stopped."
-            assert outbound.reply_to == "m-stop-1"
-            assert _message_queue.empty()
+        outbound = await asyncio.wait_for(bus.consume_outbound(), timeout=2.0)
+        assert outbound.content == "Stopped."
+        assert outbound.reply_to == "m-stop-1"
+        assert _message_queue.empty()
 
-            consumer.cancel()
-            try:
-                await consumer
-            except asyncio.CancelledError:
-                pass
+        consumer.cancel()
+        try:
+            await consumer
+        except asyncio.CancelledError:
+            pass
 
-        _run(_test())
-
-    def test_stop_cancels_queued_request_before_main_thread_processes_it(self):
+    async def test_stop_cancels_queued_request_before_main_thread_processes_it(self):
         """`/stop` should cancel a queued request instead of only acking."""
         from EvoScientist.cli import channel as channel_mod
         from EvoScientist.cli.channel import (
@@ -436,73 +414,70 @@ class TestBusInboundConsumer:
             _message_queue,
         )
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            task = asyncio.create_task(
-                _handle_bus_message(
-                    bus,
-                    manager,
-                    InboundMessage(
-                        channel="fake",
-                        sender_id="user1",
-                        chat_id="chat1",
-                        content="please work",
-                        message_id="m-work-1",
-                    ),
-                )
-            )
-
-            queued = None
-            for _ in range(20):
-                with _message_queue.mutex:
-                    queued = _message_queue.queue[0] if _message_queue.queue else None
-                if queued is not None:
-                    break
-                await asyncio.sleep(0.05)
-
-            assert queued is not None
-            with channel_mod._response_lock:
-                assert queued.msg_id in channel_mod._pending_responses
-
-            await _handle_bus_message(
+        task = asyncio.create_task(
+            _handle_bus_message(
                 bus,
                 manager,
                 InboundMessage(
                     channel="fake",
                     sender_id="user1",
                     chat_id="chat1",
-                    content="/stop",
-                    message_id="m-stop-2",
+                    content="please work",
+                    message_id="m-work-1",
                 ),
             )
+        )
 
-            with pytest.raises(asyncio.CancelledError):
-                await task
+        queued = None
+        for _ in range(20):
+            with _message_queue.mutex:
+                queued = _message_queue.queue[0] if _message_queue.queue else None
+            if queued is not None:
+                break
+            await asyncio.sleep(0.05)
 
-            skipped = _message_queue.get_nowait()
-            assert skipped.msg_id == queued.msg_id
-            assert _claim_or_complete_channel_request(skipped) is False
+        assert queued is not None
+        with channel_mod._response_lock:
+            assert queued.msg_id in channel_mod._pending_responses
 
-            with channel_mod._response_lock:
-                assert queued.msg_id not in channel_mod._pending_responses
-            with channel_mod._channel_request_lock:
-                assert queued.msg_id not in channel_mod._channel_requests
-                assert queued.msg_id not in channel_mod._cancelled_channel_messages
-                assert "fake:chat1" not in channel_mod._session_requests
+        await _handle_bus_message(
+            bus,
+            manager,
+            InboundMessage(
+                channel="fake",
+                sender_id="user1",
+                chat_id="chat1",
+                content="/stop",
+                message_id="m-stop-2",
+            ),
+        )
 
-            outbound = await asyncio.wait_for(bus.consume_outbound(), timeout=2.0)
-            assert outbound.content == "Stopped."
-            assert outbound.reply_to == "m-stop-2"
-            with pytest.raises(asyncio.TimeoutError):
-                await asyncio.wait_for(bus.consume_outbound(), timeout=0.2)
+        with pytest.raises(asyncio.CancelledError):
+            await task
 
-        _run(_test())
+        skipped = _message_queue.get_nowait()
+        assert skipped.msg_id == queued.msg_id
+        assert _claim_or_complete_channel_request(skipped) is False
 
-    def test_stop_leaves_resolved_response_available_for_delivery(self):
+        with channel_mod._response_lock:
+            assert queued.msg_id not in channel_mod._pending_responses
+        with channel_mod._channel_request_lock:
+            assert queued.msg_id not in channel_mod._channel_requests
+            assert queued.msg_id not in channel_mod._cancelled_channel_messages
+            assert "fake:chat1" not in channel_mod._session_requests
+
+        outbound = await asyncio.wait_for(bus.consume_outbound(), timeout=2.0)
+        assert outbound.content == "Stopped."
+        assert outbound.reply_to == "m-stop-2"
+        with pytest.raises(asyncio.TimeoutError):
+            await asyncio.wait_for(bus.consume_outbound(), timeout=0.2)
+
+    async def test_stop_leaves_resolved_response_available_for_delivery(self):
         """`/stop` must not steal a response whose waiter already resolved."""
         from EvoScientist.cli import channel as channel_mod
         from EvoScientist.cli.channel import (
@@ -515,42 +490,39 @@ class TestBusInboundConsumer:
             _set_channel_response,
         )
 
-        async def _test():
-            msg = ChannelMessage(
-                msg_id="msg-resolved",
-                content="already answered",
-                sender="user1",
-                channel_type="fake",
-                metadata={},
-                channel_ref=None,
-                bus_ref=None,
-                chat_id="chat1",
-                message_id="m-resolved",
-            )
+        msg = ChannelMessage(
+            msg_id="msg-resolved",
+            content="already answered",
+            sender="user1",
+            channel_type="fake",
+            metadata={},
+            channel_ref=None,
+            bus_ref=None,
+            chat_id="chat1",
+            message_id="m-resolved",
+        )
 
-            waiter = _enqueue_channel_message(msg)
-            assert _claim_channel_request(msg) is True
+        waiter = _enqueue_channel_message(msg)
+        assert _claim_channel_request(msg) is True
 
-            _set_channel_response(msg.msg_id, "final answer")
-            assert await asyncio.wait_for(asyncio.shield(waiter), timeout=1.0) == (
-                "final answer"
-            )
+        _set_channel_response(msg.msg_id, "final answer")
+        assert await asyncio.wait_for(asyncio.shield(waiter), timeout=1.0) == (
+            "final answer"
+        )
 
-            cancelled_count, active_count = _cancel_channel_session("fake", "chat1")
-            assert cancelled_count == 0
-            assert active_count == 0
+        cancelled_count, active_count = _cancel_channel_session("fake", "chat1")
+        assert cancelled_count == 0
+        assert active_count == 0
 
-            with channel_mod._response_lock:
-                assert msg.msg_id in channel_mod._pending_responses
-            with channel_mod._channel_request_lock:
-                assert msg.msg_id not in channel_mod._cancelled_channel_messages
+        with channel_mod._response_lock:
+            assert msg.msg_id in channel_mod._pending_responses
+        with channel_mod._channel_request_lock:
+            assert msg.msg_id not in channel_mod._cancelled_channel_messages
 
-            assert _pop_channel_response(msg.msg_id) == "final answer"
-            _complete_channel_request(msg.msg_id)
+        assert _pop_channel_response(msg.msg_id) == "final answer"
+        _complete_channel_request(msg.msg_id)
 
-        _run(_test())
-
-    def test_message_counting(self):
+    async def test_message_counting(self):
         """Messages are counted via record_message."""
         from EvoScientist.cli.channel import (
             _bus_inbound_consumer,
@@ -560,45 +532,42 @@ class TestBusInboundConsumer:
 
         _drain_queue(_message_queue)
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
+        consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
 
-            await bus.publish_inbound(
-                InboundMessage(
-                    channel="fake",
-                    sender_id="u1",
-                    chat_id="c1",
-                    content="test",
-                )
+        await bus.publish_inbound(
+            InboundMessage(
+                channel="fake",
+                sender_id="u1",
+                chat_id="c1",
+                content="test",
             )
+        )
 
-            for _ in range(20):
-                if not _message_queue.empty():
-                    break
-                await asyncio.sleep(0.05)
+        for _ in range(20):
+            if not _message_queue.empty():
+                break
+            await asyncio.sleep(0.05)
 
-            msg = _message_queue.get_nowait()
-            _set_channel_response(msg.msg_id, "ok")
+        msg = _message_queue.get_nowait()
+        _set_channel_response(msg.msg_id, "ok")
 
-            await asyncio.wait_for(bus.consume_outbound(), timeout=2.0)
+        await asyncio.wait_for(bus.consume_outbound(), timeout=2.0)
 
-            assert manager._message_counts["fake"]["received"] == 1
-            assert manager._message_counts["fake"]["sent"] == 1
+        assert manager._message_counts["fake"]["received"] == 1
+        assert manager._message_counts["fake"]["sent"] == 1
 
-            consumer.cancel()
-            try:
-                await consumer
-            except asyncio.CancelledError:
-                pass
+        consumer.cancel()
+        try:
+            await consumer
+        except asyncio.CancelledError:
+            pass
 
-        _run(_test())
-
-    def test_channel_message_carries_metadata(self):
+    async def test_channel_message_carries_metadata(self):
         """ChannelMessage carries metadata, chat_id, and message_id."""
         from EvoScientist.cli.channel import (
             _bus_inbound_consumer,
@@ -608,49 +577,46 @@ class TestBusInboundConsumer:
 
         _drain_queue(_message_queue)
 
-        async def _test():
-            bus = MessageBus()
-            manager = ChannelManager(bus)
-            ch = FakeChannel()
-            manager.register(ch)
+        bus = MessageBus()
+        manager = ChannelManager(bus)
+        ch = FakeChannel()
+        manager.register(ch)
 
-            consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
+        consumer = asyncio.create_task(_bus_inbound_consumer(bus, manager))
 
-            await bus.publish_inbound(
-                InboundMessage(
-                    channel="fake",
-                    sender_id="user1",
-                    chat_id="chat1",
-                    content="with metadata",
-                    metadata={"key": "value"},
-                    message_id="msg-123",
-                )
+        await bus.publish_inbound(
+            InboundMessage(
+                channel="fake",
+                sender_id="user1",
+                chat_id="chat1",
+                content="with metadata",
+                metadata={"key": "value"},
+                message_id="msg-123",
             )
+        )
 
-            for _ in range(20):
-                if not _message_queue.empty():
-                    break
-                await asyncio.sleep(0.05)
+        for _ in range(20):
+            if not _message_queue.empty():
+                break
+            await asyncio.sleep(0.05)
 
-            msg = _message_queue.get_nowait()
-            assert msg.content == "with metadata"
-            assert msg.metadata == {"key": "value"}
-            assert msg.chat_id == "chat1"
-            assert msg.message_id == "msg-123"
-            assert msg.channel_ref is ch
+        msg = _message_queue.get_nowait()
+        assert msg.content == "with metadata"
+        assert msg.metadata == {"key": "value"}
+        assert msg.chat_id == "chat1"
+        assert msg.message_id == "msg-123"
+        assert msg.channel_ref is ch
 
-            _set_channel_response(msg.msg_id, "done")
+        _set_channel_response(msg.msg_id, "done")
 
-            outbound = await asyncio.wait_for(
-                bus.consume_outbound(),
-                timeout=2.0,
-            )
-            assert outbound.reply_to == "msg-123"
+        outbound = await asyncio.wait_for(
+            bus.consume_outbound(),
+            timeout=2.0,
+        )
+        assert outbound.reply_to == "msg-123"
 
-            consumer.cancel()
-            try:
-                await consumer
-            except asyncio.CancelledError:
-                pass
-
-        _run(_test())
+        consumer.cancel()
+        try:
+            await consumer
+        except asyncio.CancelledError:
+            pass

tests/test_ccproxy_manager.py

@@ -6,6 +6,8 @@ from unittest.mock import MagicMock, patch
 import pytest
 
 from EvoScientist.ccproxy_manager import (
+    _CCPROXY_AUTH_TIMEOUT_SECONDS,
+    _CCPROXY_HEALTH_TIMEOUT_SECONDS,
     check_ccproxy_auth,
     ensure_ccproxy,
     is_ccproxy_available,
@@ -15,6 +17,7 @@ from EvoScientist.ccproxy_manager import (
     setup_codex_env,
     start_ccproxy,
     stop_ccproxy,
+    write_ccproxy_config,
 )
 
 # =============================================================================
@@ -52,6 +55,8 @@ class TestCheckCcproxyAuth:
         mock_run.assert_called_once()
         cmd = mock_run.call_args[0][0]
         assert cmd[1:] == ["auth", "status", "claude_api"]
+        # ccproxy CLI cold start takes ~10s; timeout must leave headroom
+        assert mock_run.call_args[1]["timeout"] == _CCPROXY_AUTH_TIMEOUT_SECONDS
 
     @patch("subprocess.run")
     def test_valid_auth_codex(self, mock_run):
@@ -123,9 +128,10 @@ class TestIsCcproxyRunning:
 
 
 class TestStartCcproxy:
+    @patch("EvoScientist.ccproxy_manager.logger.warning")
     @patch("EvoScientist.ccproxy_manager.is_ccproxy_running")
     @patch("subprocess.Popen")
-    def test_success(self, mock_popen, mock_running):
+    def test_success(self, mock_popen, mock_running, mock_warning):
         proc = MagicMock()
         proc.poll.return_value = None
         mock_popen.return_value = proc
@@ -134,6 +140,11 @@ class TestStartCcproxy:
 
         result = start_ccproxy(8000)
         assert result is proc
+        mock_warning.assert_called_once_with(
+            "Starting ccproxy on port %d; first startup may take up to %d seconds",
+            8000,
+            _CCPROXY_HEALTH_TIMEOUT_SECONDS,
+        )
 
     @patch("EvoScientist.ccproxy_manager.is_ccproxy_running", return_value=False)
     @patch("EvoScientist.ccproxy_manager.time")
@@ -143,7 +154,11 @@ class TestStartCcproxy:
         proc.poll.return_value = None
         mock_popen.return_value = proc
         # Simulate time passing beyond deadline
-        mock_time.monotonic.side_effect = [0, 0, 31]
+        mock_time.monotonic.side_effect = [
+            0,
+            0,
+            _CCPROXY_HEALTH_TIMEOUT_SECONDS + 1,
+        ]
         mock_time.sleep = MagicMock()
 
         with pytest.raises(RuntimeError, match="did not become healthy"):
@@ -154,6 +169,54 @@ class TestStartCcproxy:
         with pytest.raises(FileNotFoundError):
             start_ccproxy(8000)
 
+    @patch("EvoScientist.ccproxy_manager.is_ccproxy_running")
+    @patch("subprocess.Popen")
+    def test_passes_generated_config(self, mock_popen, mock_running, tmp_path):
+        proc = MagicMock()
+        proc.poll.return_value = None
+        mock_popen.return_value = proc
+        mock_running.side_effect = [True]
+
+        with patch("EvoScientist.config.get_config_dir", return_value=tmp_path):
+            start_ccproxy(8000)
+
+        cmd = mock_popen.call_args[0][0]
+        assert "--config" in cmd
+        assert cmd[cmd.index("--config") + 1] == str(tmp_path / "ccproxy.toml")
+
+    @patch("EvoScientist.ccproxy_manager.is_ccproxy_running")
+    @patch("EvoScientist.ccproxy_manager.write_ccproxy_config", side_effect=OSError)
+    @patch("subprocess.Popen")
+    def test_config_write_failure_starts_without_config(
+        self, mock_popen, mock_write, mock_running
+    ):
+        proc = MagicMock()
+        proc.poll.return_value = None
+        mock_popen.return_value = proc
+        mock_running.side_effect = [True]
+
+        start_ccproxy(8000)
+
+        cmd = mock_popen.call_args[0][0]
+        assert "--config" not in cmd
+
+
+# =============================================================================
+# write_ccproxy_config
+# =============================================================================
+
+
+class TestWriteCcproxyConfig:
+    def test_writes_codex_mapping_override(self, tmp_path):
+        config_dir = tmp_path / "missing" / "config"
+        with patch("EvoScientist.config.get_config_dir", return_value=config_dir):
+            path = write_ccproxy_config()
+
+        assert path == str(config_dir / "ccproxy.toml")
+        content = (config_dir / "ccproxy.toml").read_text(encoding="utf-8")
+        assert "[plugins.codex]" in content
+        assert "model_mappings = []" in content
+
 
 # =============================================================================
 # ensure_ccproxy

tests/test_channel_command.py

@@ -3,8 +3,6 @@
 from types import SimpleNamespace
 from unittest.mock import MagicMock, patch
 
-from tests.conftest import run_async as _run
-
 
 def _ctx():
     from EvoScientist.commands.base import ChannelRuntime, CommandContext
@@ -55,7 +53,7 @@ class TestNeedsAgent:
 class TestStartPath:
     """Start flow must propagate agent/thread_id globals."""
 
-    def test_start_binds_channel_runtime(self):
+    async def test_start_binds_channel_runtime(self):
         from EvoScientist.commands.implementation.channel import ChannelCommand
 
         ctx, _ui = _ctx()
@@ -77,11 +75,11 @@ class TestStartPath:
                 return_value=config,
             ),
         ):
-            _run(ChannelCommand().execute(ctx, ["telegram"]))
+            await ChannelCommand().execute(ctx, ["telegram"])
         assert ctx.channel_runtime.agent is ctx.agent
         assert ctx.channel_runtime.thread_id == "tid-42"
 
-    def test_start_propagates_send_thinking(self):
+    async def test_start_propagates_send_thinking(self):
         """send_thinking flag must reach _start_channels_bus_mode."""
         from EvoScientist.commands.implementation.channel import ChannelCommand
 
@@ -111,14 +109,14 @@ class TestStartPath:
                 return_value=config,
             ),
         ):
-            _run(ChannelCommand().execute(ctx, ["telegram"]))
+            await ChannelCommand().execute(ctx, ["telegram"])
         assert captured["agent"] is ctx.agent
         assert captured["thread_id"] == "tid-42"
         assert captured["send_thinking"] is False
 
 
 class TestAddToRunningPath:
-    def test_add_to_running_binds_channel_runtime(self):
+    async def test_add_to_running_binds_channel_runtime(self):
         from EvoScientist.commands.implementation.channel import ChannelCommand
 
         ctx, _ui = _ctx()
@@ -140,11 +138,11 @@ class TestAddToRunningPath:
                 return_value=config,
             ),
         ):
-            _run(ChannelCommand().execute(ctx, ["discord"]))
+            await ChannelCommand().execute(ctx, ["discord"])
         assert ctx.channel_runtime.agent is ctx.agent
         assert ctx.channel_runtime.thread_id == "tid-42"
 
-    def test_add_to_running_propagates_send_thinking(self):
+    async def test_add_to_running_propagates_send_thinking(self):
         """Adding to a running bus must honor config.channel_send_thinking."""
         from EvoScientist.commands.implementation.channel import ChannelCommand
 
@@ -173,13 +171,13 @@ class TestAddToRunningPath:
                 return_value=config,
             ),
         ):
-            _run(ChannelCommand().execute(ctx, ["discord"]))
+            await ChannelCommand().execute(ctx, ["discord"])
         assert captured["channel_type"] == "discord"
         assert captured["send_thinking"] is True
 
 
 class TestStatusPath:
-    def test_status_without_running_channels(self):
+    async def test_status_without_running_channels(self):
         from EvoScientist.commands.implementation.channel import ChannelCommand
 
         ctx, ui = _ctx()
@@ -198,6 +196,6 @@ class TestStatusPath:
                 return_value=config,
             ),
         ):
-            _run(ChannelCommand().execute(ctx, ["status"]))
+            await ChannelCommand().execute(ctx, ["status"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("No messaging channels" in m for m in msgs)

tests/test_channel_command_ui.py

@@ -8,7 +8,6 @@ import pytest
 
 from EvoScientist.commands.channel_ui import ChannelCommandUI
 from EvoScientist.gateway import ThreadStore
-from tests.conftest import run_async as _run
 from tests.fakes import FakeGraphGateway, FakeThreadStore
 
 
@@ -57,7 +56,7 @@ def _sent_text(bus_ref) -> str:
     )
 
 
-def test_handle_session_resume_sends_history_back_to_channel_without_local_duplicate():
+async def test_handle_session_resume_sends_history_back_to_channel_without_local_duplicate():
     callback = AsyncMock()
     bus_ref = SimpleNamespace(publish_outbound=AsyncMock())
 
@@ -72,7 +71,7 @@ def test_handle_session_resume_sends_history_back_to_channel_without_local_dupli
         thread_store=thread_store,
     )
 
-    _run(_run_resume(ui, "thread-42", "/workspace"))
+    await _run_resume(ui, "thread-42", "/workspace")
 
     callback.assert_awaited_once_with("thread-42", "/workspace")
     assert thread_store.calls == [("get_thread_messages", "thread-42")]
@@ -84,7 +83,7 @@ def test_handle_session_resume_sends_history_back_to_channel_without_local_dupli
     assert "EvoScientist: Here is the saved answer." in text
 
 
-def test_handle_session_resume_propagates_callback_abort_without_history():
+async def test_handle_session_resume_propagates_callback_abort_without_history():
     callback = AsyncMock(side_effect=RuntimeError("workspace conflict"))
     bus_ref = SimpleNamespace(publish_outbound=AsyncMock())
     thread_store = FakeThreadStore()
@@ -95,7 +94,7 @@ def test_handle_session_resume_propagates_callback_abort_without_history():
     )
 
     with pytest.raises(RuntimeError, match="workspace conflict"):
-        _run(_run_resume(ui, "thread-42", "/workspace"))
+        await _run_resume(ui, "thread-42", "/workspace")
 
     callback.assert_awaited_once_with("thread-42", "/workspace")
     assert thread_store.calls == []
@@ -103,7 +102,7 @@ def test_handle_session_resume_propagates_callback_abort_without_history():
     assert captured == []
 
 
-def test_handle_session_resume_reports_history_load_error():
+async def test_handle_session_resume_reports_history_load_error():
     callback = AsyncMock()
     bus_ref = SimpleNamespace(publish_outbound=AsyncMock())
     ui, captured = _make_ui(
@@ -114,7 +113,7 @@ def test_handle_session_resume_reports_history_load_error():
         ),
     )
 
-    _run(_run_resume(ui, "thread-42", "/workspace"))
+    await _run_resume(ui, "thread-42", "/workspace")
 
     callback.assert_awaited_once_with("thread-42", "/workspace")
     assert captured == []
@@ -123,7 +122,7 @@ def test_handle_session_resume_reports_history_load_error():
     assert "history unavailable: db locked" in text
 
 
-def test_handle_session_resume_distinguishes_non_displayable_messages():
+async def test_handle_session_resume_distinguishes_non_displayable_messages():
     bus_ref = SimpleNamespace(publish_outbound=AsyncMock())
     ui, captured = _make_ui(
         bus_ref=bus_ref,
@@ -132,7 +131,7 @@ def test_handle_session_resume_distinguishes_non_displayable_messages():
         ),
     )
 
-    _run(_run_resume(ui, "thread-42", "/workspace"))
+    await _run_resume(ui, "thread-42", "/workspace")
 
     assert captured == [
         "Resumed session: thread-42\nNo displayable messages in this session."

tests/test_channel_debug.py

@@ -11,8 +11,6 @@ from EvoScientist.channels.debug import (
     emit_debug_event_if,
 )
 
-from .conftest import run_async
-
 
 def test_debug_trace_enabled_from_bool():
     assert debug_trace_enabled(True) is True
@@ -75,10 +73,10 @@ def _make_channel_context(*, debug_trace=True, name="test_channel"):
     return {"channel": channel}
 
 
-def test_middleware_dedup_emits_structured_event(caplog):
+async def test_middleware_dedup_emits_structured_event(caplog):
     from EvoScientist.channels.middleware import DedupMiddleware
 
-    async def _run():
+    with caplog.at_level(logging.DEBUG):
         mw = DedupMiddleware()
         ctx = _make_channel_context()
         raw = _make_raw(message_id="dup1")
@@ -91,49 +89,40 @@ def test_middleware_dedup_emits_structured_event(caplog):
         caplog.clear()
         result = await mw.process_inbound(raw, ctx)
         assert result is None
-
-    with caplog.at_level(logging.DEBUG):
-        run_async(_run())
     assert "middleware_dedup_drop" in caplog.text
     assert "message_id=dup1" in caplog.text
 
 
-def test_middleware_allowlist_emits_structured_event(caplog):
+async def test_middleware_allowlist_emits_structured_event(caplog):
     from EvoScientist.channels.middleware import AllowListMiddleware
 
-    async def _run():
+    with caplog.at_level(logging.DEBUG):
         mw = AllowListMiddleware(allowed_senders={"allowed_user"})
         ctx = _make_channel_context()
         raw = _make_raw(sender_id="blocked_user")
         result = await mw.process_inbound(raw, ctx)
         assert result is None
-
-    with caplog.at_level(logging.DEBUG):
-        run_async(_run())
     assert "middleware_allowlist_drop" in caplog.text
     assert "reason=sender_not_allowed" in caplog.text
 
 
-def test_middleware_mention_gating_emits_structured_event(caplog):
+async def test_middleware_mention_gating_emits_structured_event(caplog):
     from EvoScientist.channels.middleware import MentionGatingMiddleware
 
-    async def _run():
+    with caplog.at_level(logging.DEBUG):
         mw = MentionGatingMiddleware(require_mention="group")
         ctx = _make_channel_context()
         raw = _make_raw(is_group=True, was_mentioned=False)
         result = await mw.process_inbound(raw, ctx)
         assert result is None
-
-    with caplog.at_level(logging.DEBUG):
-        run_async(_run())
     assert "middleware_mention_drop" in caplog.text
     assert "policy=group" in caplog.text
 
 
-def test_typing_manager_emits_trace_events(caplog):
+async def test_typing_manager_emits_trace_events(caplog):
     from EvoScientist.channels.middleware import TypingManager
 
-    async def _run():
+    with caplog.at_level(logging.DEBUG):
         send_action = AsyncMock(side_effect=RuntimeError("typing api down"))
         mgr = TypingManager(
             send_action,
@@ -144,17 +133,14 @@ def test_typing_manager_emits_trace_events(caplog):
         await mgr.start("chat1")
         await asyncio.sleep(0.01)
         await mgr.stop("chat1")
-
-    with caplog.at_level(logging.DEBUG):
-        run_async(_run())
     assert "typing_error" in caplog.text
     assert "chat_id=chat1" in caplog.text
 
 
-def test_ack_reaction_emits_error_traces(caplog):
+async def test_ack_reaction_emits_error_traces(caplog):
     from EvoScientist.channels.middleware import AckReactionMiddleware
 
-    async def _run():
+    with caplog.at_level(logging.DEBUG):
         send_fn = AsyncMock()
         remove_fn = AsyncMock(side_effect=RuntimeError("remove failed"))
         ack = AckReactionMiddleware(
@@ -178,16 +164,13 @@ def test_ack_reaction_emits_error_traces(caplog):
         send_fn.reset_mock()
         send_fn.side_effect = RuntimeError("api down")
         await ack.send_ack("chat2", "msg2")
-
-    with caplog.at_level(logging.DEBUG):
-        run_async(_run())
     assert "ack_send_error" in caplog.text
     assert "ack_remove_error" in caplog.text
     assert "api down" in caplog.text
     assert "remove failed" in caplog.text
 
 
-def test_inbound_raw_event_emitted(caplog):
+async def test_inbound_raw_event_emitted(caplog):
     """Integration-style: _enqueue_raw emits inbound_raw at the top."""
     from EvoScientist.channels.base import Channel, RawIncoming
 
@@ -218,19 +201,16 @@ def test_inbound_raw_event_emitted(caplog):
     config.ack_scope = "off"
     config.dedup_ttl = 3600
 
-    async def _run():
+    with caplog.at_level(logging.DEBUG):
         with patch.object(Channel, "__abstractmethods__", set()):
             ch = _TestChannel(config)
         raw = RawIncoming(sender_id="u1", chat_id="c1", text="hi", message_id="m1")
         await ch._enqueue_raw(raw)
-
-    with caplog.at_level(logging.DEBUG):
-        run_async(_run())
     assert "inbound_raw" in caplog.text
     assert "sender_id=u1" in caplog.text
 
 
-def test_format_fallback_emits_event(caplog):
+async def test_format_fallback_emits_event(caplog):
     """_send_with_format_fallback emits outbound_format_fallback on fallback."""
     from EvoScientist.channels.base import Channel
 
@@ -268,13 +248,10 @@ def test_format_fallback_emits_event(caplog):
         if call_count == 1:
             raise ValueError("parse error in formatted text")
 
-    async def _run():
+    with caplog.at_level(logging.DEBUG):
         with patch.object(Channel, "__abstractmethods__", set()):
             ch = _TestChannel(config)
         await ch._send_with_format_fallback(_failing_send, "<b>hi</b>", "hi")
-
-    with caplog.at_level(logging.DEBUG):
-        run_async(_run())
     assert "outbound_format_fallback" in caplog.text
     assert call_count == 2
 
@@ -304,7 +281,7 @@ def test_trace_mixin_trace_event(caplog):
     assert "key=val" in caplog.text
 
 
-def test_standalone_dispatcher_treats_false_send_as_error(caplog):
+async def test_standalone_dispatcher_treats_false_send_as_error(caplog):
     from EvoScientist.channels.bus import MessageBus
     from EvoScientist.channels.bus.events import OutboundMessage
     from EvoScientist.channels.standalone import standalone_outbound_dispatcher
@@ -315,7 +292,7 @@ def test_standalone_dispatcher_treats_false_send_as_error(caplog):
     channel.send = AsyncMock(return_value=False)
     bus = MessageBus()
 
-    async def _run():
+    with caplog.at_level(logging.DEBUG):
         task = asyncio.create_task(standalone_outbound_dispatcher(bus, channel))
         await bus.publish_outbound(
             OutboundMessage(channel="test", chat_id="c1", content="hi")
@@ -326,14 +303,11 @@ def test_standalone_dispatcher_treats_false_send_as_error(caplog):
             await task
         except asyncio.CancelledError:
             pass
-
-    with caplog.at_level(logging.DEBUG):
-        run_async(_run())
     assert "standalone_dispatch_error" in caplog.text
     assert "send() returned False" in caplog.text
 
 
-def test_standalone_dispatcher_sends_media():
+async def test_standalone_dispatcher_sends_media():
     from EvoScientist.channels.bus import MessageBus
     from EvoScientist.channels.bus.events import OutboundMessage
     from EvoScientist.channels.standalone import standalone_outbound_dispatcher
@@ -345,21 +319,16 @@ def test_standalone_dispatcher_sends_media():
     channel.send_media = AsyncMock(return_value=True)
     bus = MessageBus()
 
-    async def _run():
-        task = asyncio.create_task(standalone_outbound_dispatcher(bus, channel))
-        await bus.publish_outbound(
-            OutboundMessage(
-                channel="test", chat_id="c1", content="", media=["/tmp/a.png"]
-            )
-        )
-        await asyncio.sleep(0.05)
-        task.cancel()
-        try:
-            await task
-        except asyncio.CancelledError:
-            pass
-
-    run_async(_run())
+    task = asyncio.create_task(standalone_outbound_dispatcher(bus, channel))
+    await bus.publish_outbound(
+        OutboundMessage(channel="test", chat_id="c1", content="", media=["/tmp/a.png"])
+    )
+    await asyncio.sleep(0.05)
+    task.cancel()
+    try:
+        await task
+    except asyncio.CancelledError:
+        pass
     channel.send_media.assert_awaited_once_with(
         recipient="c1",
         file_path="/tmp/a.png",

tests/test_cli_channel_slash.py

@@ -14,7 +14,6 @@ from EvoScientist.cli.channel import (
 from EvoScientist.cli.channel import (
     dispatch_channel_slash_command as _dispatch_channel_slash_command,
 )
-from tests.conftest import run_async as _run
 from tests.fakes import FakeGraphGateway, FakeThreadStore
 
 
@@ -43,25 +42,23 @@ def _make_msg(
     )
 
 
-def test_non_slash_returns_false():
+async def test_non_slash_returns_false():
     """Plain text messages must fall through to the agent."""
     msg = _make_msg(content="hello agent")
     append = MagicMock()
-    handled = _run(
-        dispatch_channel_slash_command(
-            msg,
-            agent=None,
-            thread_id="t1",
-            workspace_dir=None,
-            checkpointer=None,
-            append_system=append,
-        )
+    handled = await dispatch_channel_slash_command(
+        msg,
+        agent=None,
+        thread_id="t1",
+        workspace_dir=None,
+        checkpointer=None,
+        append_system=append,
     )
     assert handled is False
     append.assert_not_called()
 
 
-def test_unresolved_slash_returns_false():
+async def test_unresolved_slash_returns_false():
     """Unknown slash commands must fall through (matches TUI behavior)."""
     msg = _make_msg(content="/unknown-cmd")
     append = MagicMock()
@@ -69,20 +66,18 @@ def test_unresolved_slash_returns_false():
         "EvoScientist.commands.manager.manager.resolve",
         return_value=None,
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent=None,
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=append,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent=None,
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=append,
         )
     assert handled is False
 
 
-def test_successful_slash_execution_sets_response_and_breadcrumb():
+async def test_successful_slash_execution_sets_response_and_breadcrumb():
     """Known slash command: cmd_manager.execute ran, helper returns True,
     sends a confirmation to the channel user, and appends a local log line."""
     msg = _make_msg()
@@ -100,15 +95,13 @@ def test_successful_slash_execution_sets_response_and_breadcrumb():
         ) as mock_execute,
         patch("EvoScientist.cli.channel._set_channel_response") as mock_set_resp,
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent="fake-agent",
-                thread_id="t1",
-                workspace_dir="/tmp",
-                checkpointer=None,
-                append_system=append,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent="fake-agent",
+            thread_id="t1",
+            workspace_dir="/tmp",
+            checkpointer=None,
+            append_system=append,
         )
     assert handled is True
     mock_execute.assert_awaited_once()
@@ -119,7 +112,7 @@ def test_successful_slash_execution_sets_response_and_breadcrumb():
     assert any("Executed command from" in t for t in breadcrumbs)
 
 
-def test_slash_dispatch_passes_graph_gateway_to_command_context():
+async def test_slash_dispatch_passes_graph_gateway_to_command_context():
     msg = _make_msg()
     fake_cmd = MagicMock()
     fake_cmd.needs_agent.return_value = False
@@ -142,23 +135,21 @@ def test_slash_dispatch_passes_graph_gateway_to_command_context():
         ),
         patch("EvoScientist.cli.channel._set_channel_response"),
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent="fake-agent",
-                thread_id="t1",
-                workspace_dir="/tmp",
-                checkpointer=None,
-                append_system=append,
-                graph_gateway=graph_gateway,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent="fake-agent",
+            thread_id="t1",
+            workspace_dir="/tmp",
+            checkpointer=None,
+            append_system=append,
+            graph_gateway=graph_gateway,
         )
 
     assert handled is True
     assert captured["graph_gateway"] is graph_gateway
 
 
-def test_needs_agent_awaits_loader_and_passes_result():
+async def test_needs_agent_awaits_loader_and_passes_result():
     """Commands with needs_agent=True must await the loader and the
     resulting agent must flow through the CommandContext."""
     msg = _make_msg()
@@ -182,16 +173,14 @@ def test_needs_agent_awaits_loader_and_passes_result():
         ) as mock_execute,
         patch("EvoScientist.cli.channel._set_channel_response"),
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent=None,
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=append,
-                await_agent_ready=_await_ready,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent=None,
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=append,
+            await_agent_ready=_await_ready,
         )
     assert handled is True
     await_called.assert_called_once()
@@ -201,7 +190,7 @@ def test_needs_agent_awaits_loader_and_passes_result():
     assert ctx_arg.agent == "ready-agent"
 
 
-def test_await_agent_ready_failure_sets_error_response():
+async def test_await_agent_ready_failure_sets_error_response():
     msg = _make_msg()
     fake_cmd = MagicMock()
     fake_cmd.needs_agent.return_value = True
@@ -217,16 +206,14 @@ def test_await_agent_ready_failure_sets_error_response():
         ),
         patch("EvoScientist.cli.channel._set_channel_response") as mock_set_resp,
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent=None,
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=append,
-                await_agent_ready=_await_ready,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent=None,
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=append,
+            await_agent_ready=_await_ready,
         )
     assert handled is True
     mock_set_resp.assert_called_once()
@@ -235,7 +222,7 @@ def test_await_agent_ready_failure_sets_error_response():
     assert "agent blew up" in resp_text
 
 
-def test_cmd_manager_raises_returns_true_with_error():
+async def test_cmd_manager_raises_returns_true_with_error():
     """If cmd_manager.execute raises past its own try/except, the helper
     must absorb it, return True, and report via _set_channel_response."""
     msg = _make_msg()
@@ -253,15 +240,13 @@ def test_cmd_manager_raises_returns_true_with_error():
         ),
         patch("EvoScientist.cli.channel._set_channel_response") as mock_set_resp,
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent=None,
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=append,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent=None,
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=append,
         )
     assert handled is True
     mock_set_resp.assert_called_once()
@@ -270,7 +255,7 @@ def test_cmd_manager_raises_returns_true_with_error():
     assert "boom" in resp_text
 
 
-def test_on_cmd_completed_awaited_with_ctx_original_agent_and_cmd():
+async def test_on_cmd_completed_awaited_with_ctx_original_agent_and_cmd():
     """After a successful slash execute, the on_cmd_completed hook must
     be awaited with (ctx, original_agent, cmd) so Rich CLI can adopt an
     ``/model`` agent swap and refresh status for state-mutating commands."""
@@ -302,16 +287,14 @@ def test_on_cmd_completed_awaited_with_ctx_original_agent_and_cmd():
         ),
         patch("EvoScientist.cli.channel._set_channel_response"),
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent="original-agent",
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=append,
-                on_cmd_completed=_on_completed,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent="original-agent",
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=append,
+            on_cmd_completed=_on_completed,
         )
     assert handled is True
     assert captured["ctx_agent"] == "swapped-agent"
@@ -319,7 +302,7 @@ def test_on_cmd_completed_awaited_with_ctx_original_agent_and_cmd():
     assert captured["cmd_name"] == "/model"
 
 
-def test_on_cmd_completed_receives_cmd_for_new_and_compact():
+async def test_on_cmd_completed_receives_cmd_for_new_and_compact():
     """``/new`` / ``/compact`` invoked via channel must flow the cmd into
     the hook so the callback can still refresh status when the agent
     didn't swap — mirrors REPL ``interactive.py:1027-1030``."""
@@ -343,21 +326,19 @@ def test_on_cmd_completed_receives_cmd_for_new_and_compact():
             ),
             patch("EvoScientist.cli.channel._set_channel_response"),
         ):
-            _run(
-                dispatch_channel_slash_command(
-                    _make_msg(content=cmd_name),
-                    agent="same-agent",
-                    thread_id="t1",
-                    workspace_dir=None,
-                    checkpointer=None,
-                    append_system=MagicMock(),
-                    on_cmd_completed=_on_completed,
-                )
+            await dispatch_channel_slash_command(
+                _make_msg(content=cmd_name),
+                agent="same-agent",
+                thread_id="t1",
+                workspace_dir=None,
+                checkpointer=None,
+                append_system=MagicMock(),
+                on_cmd_completed=_on_completed,
             )
         assert captured["cmd_name"] == cmd_name, cmd_name
 
 
-def test_on_cmd_completed_skipped_on_fall_through_and_error():
+async def test_on_cmd_completed_skipped_on_fall_through_and_error():
     """The hook must NOT fire for unresolved slash, non-slash text, or
     when cmd_manager.execute raised."""
     fake_cmd = MagicMock()
@@ -369,16 +350,14 @@ def test_on_cmd_completed_skipped_on_fall_through_and_error():
 
     # Non-slash
     with patch("EvoScientist.cli.channel._set_channel_response"):
-        _run(
-            dispatch_channel_slash_command(
-                _make_msg(content="hi"),
-                agent=None,
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=MagicMock(),
-                on_cmd_completed=_noop,
-            )
+        await dispatch_channel_slash_command(
+            _make_msg(content="hi"),
+            agent=None,
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=MagicMock(),
+            on_cmd_completed=_noop,
         )
     # Unresolved slash
     with (
@@ -388,16 +367,14 @@ def test_on_cmd_completed_skipped_on_fall_through_and_error():
         ),
         patch("EvoScientist.cli.channel._set_channel_response"),
     ):
-        _run(
-            dispatch_channel_slash_command(
-                _make_msg(content="/nope"),
-                agent=None,
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=MagicMock(),
-                on_cmd_completed=_noop,
-            )
+        await dispatch_channel_slash_command(
+            _make_msg(content="/nope"),
+            agent=None,
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=MagicMock(),
+            on_cmd_completed=_noop,
         )
     # Execute raises
     with (
@@ -411,22 +388,20 @@ def test_on_cmd_completed_skipped_on_fall_through_and_error():
         ),
         patch("EvoScientist.cli.channel._set_channel_response"),
     ):
-        _run(
-            dispatch_channel_slash_command(
-                _make_msg(),
-                agent=None,
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=MagicMock(),
-                on_cmd_completed=_noop,
-            )
+        await dispatch_channel_slash_command(
+            _make_msg(),
+            agent=None,
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=MagicMock(),
+            on_cmd_completed=_noop,
         )
 
     completed.assert_not_called()
 
 
-def test_command_error_skips_completion_hook_and_reports_error():
+async def test_command_error_skips_completion_hook_and_reports_error():
     """A command caught as failed by CommandManager must not look successful."""
     msg = _make_msg(content="/resume abc")
     fake_cmd = MagicMock()
@@ -450,16 +425,14 @@ def test_command_error_skips_completion_hook_and_reports_error():
         ),
         patch("EvoScientist.cli.channel._set_channel_response") as mock_set_resp,
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent=None,
-                thread_id="old-thread",
-                workspace_dir="/old-workspace",
-                checkpointer=None,
-                append_system=MagicMock(),
-                on_cmd_completed=completed,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent=None,
+            thread_id="old-thread",
+            workspace_dir="/old-workspace",
+            checkpointer=None,
+            append_system=MagicMock(),
+            on_cmd_completed=completed,
         )
 
     assert handled is True
@@ -467,7 +440,7 @@ def test_command_error_skips_completion_hook_and_reports_error():
     mock_set_resp.assert_called_once_with("msg-1", "Command error: workspace conflict")
 
 
-def test_empty_command_error_still_reports_error():
+async def test_empty_command_error_still_reports_error():
     """An empty string error is still a command failure sentinel."""
     msg = _make_msg(content="/resume abc")
     fake_cmd = MagicMock()
@@ -489,16 +462,14 @@ def test_empty_command_error_still_reports_error():
         ),
         patch("EvoScientist.cli.channel._set_channel_response") as mock_set_resp,
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent=None,
-                thread_id="old-thread",
-                workspace_dir="/old-workspace",
-                checkpointer=None,
-                append_system=MagicMock(),
-                on_cmd_completed=completed,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent=None,
+            thread_id="old-thread",
+            workspace_dir="/old-workspace",
+            checkpointer=None,
+            append_system=MagicMock(),
+            on_cmd_completed=completed,
         )
 
     assert handled is True
@@ -506,7 +477,7 @@ def test_empty_command_error_still_reports_error():
     mock_set_resp.assert_called_once_with("msg-1", "Command error: (no details)")
 
 
-def test_on_cmd_completed_exception_is_absorbed():
+async def test_on_cmd_completed_exception_is_absorbed():
     """A raising hook must NOT prevent the channel response from being set."""
     msg = _make_msg()
     fake_cmd = MagicMock()
@@ -526,23 +497,21 @@ def test_on_cmd_completed_exception_is_absorbed():
         ),
         patch("EvoScientist.cli.channel._set_channel_response") as mock_set_resp,
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent="orig",
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=MagicMock(),
-                on_cmd_completed=_boom,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent="orig",
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=MagicMock(),
+            on_cmd_completed=_boom,
         )
     assert handled is True
     mock_set_resp.assert_called_once()
     assert "Command executed" in mock_set_resp.call_args[0][1]
 
 
-def test_top_level_exception_is_absorbed():
+async def test_top_level_exception_is_absorbed():
     """Last-ditch safety net: if anything inside the dispatch pipeline
     raises unexpectedly (lazy import failure, ChannelCommandUI ctor,
     terminal I/O from append_system, ...), the helper must NOT
@@ -557,15 +526,13 @@ def test_top_level_exception_is_absorbed():
         ),
         patch("EvoScientist.cli.channel._set_channel_response") as mock_set_resp,
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent=None,
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=MagicMock(),
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent=None,
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=MagicMock(),
         )
     assert handled is True
     mock_set_resp.assert_called_once()
@@ -574,7 +541,7 @@ def test_top_level_exception_is_absorbed():
     assert "exploded during resolve" in resp_text
 
 
-def test_cmd_execute_returning_false_falls_through():
+async def test_cmd_execute_returning_false_falls_through():
     """When cmd_manager.execute returns False (empty/unparseable input),
     the helper must return False so the caller falls through to the agent."""
     msg = _make_msg(content="/")
@@ -592,15 +559,13 @@ def test_cmd_execute_returning_false_falls_through():
         ),
         patch("EvoScientist.cli.channel._set_channel_response") as mock_set_resp,
     ):
-        handled = _run(
-            dispatch_channel_slash_command(
-                msg,
-                agent=None,
-                thread_id="t1",
-                workspace_dir=None,
-                checkpointer=None,
-                append_system=append,
-            )
+        handled = await dispatch_channel_slash_command(
+            msg,
+            agent=None,
+            thread_id="t1",
+            workspace_dir=None,
+            checkpointer=None,
+            append_system=append,
         )
     assert handled is False
     mock_set_resp.assert_not_called()

tests/test_cli_tui_dispatch.py

@@ -1,6 +1,5 @@
 """Tests for CLI interactive UI backend dispatch."""
 
-import asyncio
 from types import SimpleNamespace
 
 import pytest
@@ -101,7 +100,7 @@ def test_background_agent_server_starts_even_when_async_subagents_disabled(
     assert calls == [(config, "/tmp/workspace")]
 
 
-def test_resume_workspace_sync_runs_even_when_async_subagents_disabled(
+async def test_resume_workspace_sync_runs_even_when_async_subagents_disabled(
     monkeypatch,
 ):
     import EvoScientist.cli.commands as cmds
@@ -117,11 +116,9 @@ def test_resume_workspace_sync_runs_even_when_async_subagents_disabled(
     )
 
     config = SimpleNamespace(enable_async_subagents=False)
-    asyncio.run(
-        cmds._sync_background_agent_server_workspace(
-            config,
-            workspace_dir="/tmp/resumed-workspace",
-        )
+    await cmds._sync_background_agent_server_workspace(
+        config,
+        workspace_dir="/tmp/resumed-workspace",
     )
 
     assert calls == [(config, "/tmp/resumed-workspace")]

tests/test_code_interpreter_middleware.py

@@ -1,4 +1,5 @@
-"""Regression tests for the code_interpreter PTC allowlist.
+"""Regression tests for the code_interpreter PTC allowlist and the
+``EvoCodeInterpreterMiddleware`` subclass shape.
 
 langchain-quickjs >=0.3 reserves the ``task`` sub-agent dispatch tool as the
 top-level REPL global and raises ``ValueError`` if ``task`` appears in the
@@ -8,7 +9,10 @@ allowlist (``task()`` stays reachable as the REPL global, with responseSchema).
 
 from __future__ import annotations
 
+from unittest.mock import MagicMock
+
 import pytest
+from langchain_core.messages import AIMessage, HumanMessage
 
 from EvoScientist.middleware.code_interpreter import (
     _DEFAULT_PTC_ALLOWLIST,
@@ -45,3 +49,290 @@ def test_filter_tools_for_ptc_accepts_default_allowlist():
 
 def test_create_code_interpreter_middleware_builds():
     assert create_code_interpreter_middleware() is not None
+
+
+def test_middleware_uses_thread_mode():
+    """Upstream ``mode="thread"`` (the default) preserves cross-turn REPL
+    state as ``langchain-ai/deepagents#3064`` shipped it. The wire-cost
+    bloat that motivated the earlier ``mode="turn"`` regression guard is
+    fixed at the API serialization layer (``EvoFilteredGraph`` in
+    ``EvoScientist/langgraph_dev/main_graph.py``), not by revoking the
+    persistence feature.
+    """
+    mw = create_code_interpreter_middleware()
+    assert mw._mode == "thread"
+
+
+def test_after_agent_evicts_slot_on_untouched_turn():
+    """Regression guard against reintroducing a conditional-snapshot gate
+    that skips ``after_agent`` on untouched turns.
+
+    Upstream ``after_agent`` in ``langchain_quickjs/middleware.py`` performs
+    two things: snapshot the REPL AND evict the slot (``finally:
+    self._registry.evict(thread_id)``). ``before_agent`` restores the REPL
+    on any turn that follows a touched one via ``self._registry.get`` —
+    which is get-or-create. So if ``after_agent`` returns early without
+    evicting, one ``ThreadWorker`` + QuickJS Runtime leaks per persistent
+    ``thread_id`` that ever went touched → quiet.
+
+    Fix: don't override ``after_agent`` / ``aafter_agent`` at all — inherit
+    upstream's unconditional snapshot+evict behavior. This test creates a
+    slot the way ``before_agent`` would, calls ``after_agent`` with an
+    untouched-state input, and asserts the slot was evicted.
+    """
+    mw = create_code_interpreter_middleware()
+    tid = mw._fallback_thread_id
+
+    # Simulate the slot creation that ``before_agent`` performs when it sees
+    # a prior turn's snapshot payload in state.
+    mw._registry.get(tid)
+    assert len(mw._registry._slots) == 1
+
+    # Untouched-turn state: no ``code_interpreter`` tool call between the
+    # last ``HumanMessage`` and end. Under the earlier buggy gate this
+    # returned ``{}`` without evicting — leaking the slot created above.
+    untouched_state = {
+        "_quickjs_snapshot_payload": b"payload-from-prior-turn",
+        "messages": [
+            HumanMessage(content="thanks"),
+            AIMessage(content="you're welcome"),
+        ],
+    }
+    mw.after_agent(untouched_state, runtime=None)
+
+    assert len(mw._registry._slots) == 0, (
+        "after_agent must evict the slot even on untouched turns, because "
+        "before_agent already restored a REPL that owns a ThreadWorker + "
+        "QuickJS Runtime. Skipping eviction leaks those resources."
+    )
+
+
+def test_evo_filtered_graph_strips_private_snapshot_field():
+    """The ``StateSnapshot`` returned by ``EvoScientist_agent.get_state`` must
+    not contain ``_quickjs_snapshot_payload`` in either ``values`` (the
+    materialized channel payload) or ``metadata['writes']`` (the raw write
+    records surfaced by ``get_state_history``).
+    """
+    from EvoScientist.langgraph_dev.main_graph import _EvoFilteredGraph, _strip_private
+
+    snap = MagicMock()
+    snap.values = {
+        "messages": ["m1"],
+        "_quickjs_snapshot_payload": b"x" * 100,
+        "skills_metadata": [],
+    }
+    snap.metadata = {
+        "source": "loop",
+        "step": 42,
+        "writes": {
+            "CodeInterpreterMiddleware.after_agent": {
+                "_quickjs_snapshot_payload": ("snap", b"y" * 1_400_000),
+                "messages": [],
+            },
+            "model": {"messages": ["m1"]},
+        },
+        "parents": {},
+    }
+    _strip_private(snap)
+    snap._replace.assert_called_once()
+    kwargs = snap._replace.call_args.kwargs
+    assert "_quickjs_snapshot_payload" not in kwargs["values"]
+    assert "messages" in kwargs["values"]
+    assert "skills_metadata" in kwargs["values"]
+    scrubbed_writes = kwargs["metadata"]["writes"]
+    assert (
+        "_quickjs_snapshot_payload"
+        not in scrubbed_writes["CodeInterpreterMiddleware.after_agent"]
+    )
+    assert "messages" in scrubbed_writes["CodeInterpreterMiddleware.after_agent"]
+    assert scrubbed_writes["model"] == {"messages": ["m1"]}
+    # Non-writes metadata keys are preserved.
+    assert kwargs["metadata"]["source"] == "loop"
+    assert kwargs["metadata"]["step"] == 42
+    # Sanity: the class exists and inherits from CompiledStateGraph.
+    from langgraph.graph.state import CompiledStateGraph
+
+    assert issubclass(_EvoFilteredGraph, CompiledStateGraph)
+
+
+def test_strip_private_handles_missing_metadata_writes():
+    """``metadata['writes']`` can be missing or ``None`` on some snapshots
+    (e.g. initial state). The filter must not crash and must still strip
+    values.
+    """
+    from EvoScientist.langgraph_dev.main_graph import _strip_private
+
+    snap = MagicMock()
+    snap.values = {"_quickjs_snapshot_payload": b"x", "messages": []}
+    snap.metadata = {"source": "input", "step": -1, "writes": None}
+    snap.tasks = ()
+    _strip_private(snap)
+    kwargs = snap._replace.call_args.kwargs
+    assert "_quickjs_snapshot_payload" not in kwargs["values"]
+    # writes was None, metadata passes through unchanged.
+    assert kwargs["metadata"]["writes"] is None
+
+
+def test_strip_private_scrubs_task_result_snapshot_blob():
+    """``tasks[*].result`` is where ``after_agent``'s return dict lands.
+    When the middleware snapshots, ``result`` carries
+    ``{"_quickjs_snapshot_payload": ("snap", ~1.4 MB bytes)}``. Verified
+    on live history: this is the dominant per-response leak, larger than
+    ``values`` and ``metadata.writes`` combined for anchor checkpoints.
+    """
+    from EvoScientist.langgraph_dev.main_graph import _strip_private
+
+    class FakeTask:
+        def __init__(self, id_, result):
+            self.id = id_
+            self.name = "CodeInterpreterMiddleware.after_agent"
+            self.result = result
+
+        def _replace(self, **kwargs):
+            for k, v in kwargs.items():
+                setattr(self, k, v)
+            return self
+
+    leaking_task = FakeTask(
+        "t1", {"_quickjs_snapshot_payload": ("snap", b"z" * 1_400_000), "messages": []}
+    )
+    clean_task = FakeTask("t2", {"messages": ["hi"]})
+    snap = MagicMock()
+    snap.values = {}
+    snap.metadata = {"source": "loop", "step": 5}
+    snap.tasks = (leaking_task, clean_task)
+    _strip_private(snap)
+    kwargs = snap._replace.call_args.kwargs
+    tasks_after = kwargs["tasks"]
+    assert "_quickjs_snapshot_payload" not in tasks_after[0].result
+    assert "messages" in tasks_after[0].result
+    # Clean task is passed through untouched.
+    assert tasks_after[1] is clean_task
+
+
+def test_agent_uses_filtered_graph_class():
+    """The ``__class__`` swap in ``main_graph.py`` is the load-bearing wiring
+    that makes ``_strip_private`` reach the langgraph-api endpoints.
+    ``_strip_private`` and ``_EvoFilteredGraph`` in isolation don't prove the
+    swap ran; every other test in this file passes even if someone drops the
+    swap line. This asserts the compiled agent is actually the filtered
+    subclass at module-load time, and that the subclass survives
+    ``Pregel.copy(update=...)`` — the call langgraph-api makes in
+    ``get_graph`` before yielding the graph to endpoint handlers.
+    """
+    from EvoScientist.langgraph_dev.main_graph import (
+        EvoScientist_agent,
+        _EvoFilteredGraph,
+    )
+
+    assert isinstance(EvoScientist_agent, _EvoFilteredGraph)
+    assert isinstance(EvoScientist_agent.copy(update={}), _EvoFilteredGraph)
+
+
+def test_all_registered_graphs_use_filtered_graph_class():
+    """Every graph registered in ``langgraph.json`` (main + all subagents)
+    gets the ``__class__`` swap via ``_apply_filter_to_all_registered_graphs``.
+    Iterating the config directly matches the auto-detect refactor: adding
+    a new subagent to ``langgraph.json`` should not require a corresponding
+    test update.
+
+    Subagents get ``create_code_interpreter_middleware`` unconditionally
+    (``EvoScientist.py:_build_middleware_stack``), so they can touch the
+    QuickJS REPL and write ``_quickjs_snapshot_payload`` on their own
+    checkpoint namespace. Async subagents also get their own ``thread_id``
+    and their ``/threads/{id}/state`` endpoint runs on their own compiled
+    graph — without the swap on those graphs, our filter would miss that
+    endpoint entirely.
+    """
+    import json
+    from importlib import import_module
+    from pathlib import Path
+
+    # Import triggers ``main_graph``'s swap loop.
+    from EvoScientist.langgraph_dev import main_graph
+    from EvoScientist.langgraph_dev.main_graph import _EvoFilteredGraph
+
+    config_path = Path(main_graph.__file__).parent / "langgraph.json"
+    config = json.loads(config_path.read_text())
+    for name, path in config["graphs"].items():
+        module_path, attr = path.rsplit(":", 1)
+        graph = getattr(import_module(module_path), attr)
+        assert isinstance(graph, _EvoFilteredGraph), (
+            f"graph {name!r} ({path}) did not receive the class swap"
+        )
+
+
+def test_strip_private_recurses_into_nested_subgraph_state():
+    """When ``subgraphs=True``, ``PregelTask.state`` holds a nested
+    ``StateSnapshot`` for the subgraph. Its ``values`` (and its own nested
+    tasks) can carry ``_quickjs_snapshot_payload`` just like the parent.
+    Recursion covers the compound leak path CodeRabbit flagged.
+    """
+    from langgraph.types import StateSnapshot
+
+    from EvoScientist.langgraph_dev.main_graph import _strip_private
+
+    nested_snap = StateSnapshot(
+        values={"_quickjs_snapshot_payload": b"n" * 1_400_000, "messages": []},
+        next=(),
+        config={},
+        metadata={"source": "loop", "step": 3},
+        created_at="2026-07-01T12:00:00Z",
+        parent_config=None,
+        tasks=(),
+        interrupts=(),
+    )
+
+    class FakeTask:
+        def __init__(self, state):
+            self.id = "sub-1"
+            self.name = "subgraph"
+            self.result = None
+            self.state = state
+
+        def _replace(self, **kwargs):
+            for k, v in kwargs.items():
+                setattr(self, k, v)
+            return self
+
+    task_with_nested = FakeTask(nested_snap)
+    task_with_config_state = FakeTask({"configurable": {"thread_id": "t"}})
+    snap = MagicMock()
+    snap.values = {}
+    snap.metadata = {"source": "loop", "step": 5}
+    snap.tasks = (task_with_nested, task_with_config_state)
+    _strip_private(snap)
+    kwargs = snap._replace.call_args.kwargs
+    tasks_after = kwargs["tasks"]
+    # Nested StateSnapshot got recursively scrubbed.
+    assert "_quickjs_snapshot_payload" not in tasks_after[0].state.values
+    assert "messages" in tasks_after[0].state.values
+    # A dict (RunnableConfig-shaped) state passes through unchanged — we only
+    # recurse into ``StateSnapshot`` instances.
+    assert tasks_after[1].state == {"configurable": {"thread_id": "t"}}
+
+
+def test_strip_private_scrubs_delta_counters():
+    """``metadata['counters_since_delta_snapshot']`` is a small
+    ``{channel: [count, superstep]}`` bookkeeping map. Not a size problem,
+    but leaks the channel name — strip for consistency with the private
+    annotation.
+    """
+    from EvoScientist.langgraph_dev.main_graph import _strip_private
+
+    snap = MagicMock()
+    snap.values = {}
+    snap.metadata = {
+        "source": "loop",
+        "step": 5,
+        "counters_since_delta_snapshot": {
+            "_quickjs_snapshot_payload": [1, 14],
+            "messages": [3, 14],
+        },
+    }
+    snap.tasks = ()
+    _strip_private(snap)
+    kwargs = snap._replace.call_args.kwargs
+    counters = kwargs["metadata"]["counters_since_delta_snapshot"]
+    assert "_quickjs_snapshot_payload" not in counters
+    assert "messages" in counters

tests/test_compact_command.py

@@ -4,13 +4,12 @@ from types import SimpleNamespace
 from unittest.mock import AsyncMock, MagicMock, patch
 
 from EvoScientist.gateway import GraphTarget
-from tests.conftest import run_async as _run
 from tests.fakes import FakeCommandUI, FakeGraphGateway
 
 _TARGET = GraphTarget()
 
 
-def _compact(
+async def _compact(
     graph_gateway: FakeGraphGateway,
     *,
     thread_id: str = "tid-1",
@@ -18,30 +17,28 @@ def _compact(
 ):
     from EvoScientist.cli.commands import compact_conversation
 
-    return _run(
-        compact_conversation(
-            graph_gateway=graph_gateway,
-            thread_id=thread_id,
-            target=_TARGET,
-            input_tokens_hint=input_tokens_hint,
-        )
+    return await compact_conversation(
+        graph_gateway=graph_gateway,
+        thread_id=thread_id,
+        target=_TARGET,
+        input_tokens_hint=input_tokens_hint,
     )
 
 
 class TestCompactGuards:
     """Guard conditions that return early without touching the middleware."""
 
-    def test_empty_messages(self):
+    async def test_empty_messages(self):
         graph_gateway = FakeGraphGateway(state_values={"messages": []})
 
-        result = _compact(graph_gateway)
+        result = await _compact(graph_gateway)
         assert result.status == "noop"
         assert "no messages" in result.message
 
-    def test_state_read_failure(self):
+    async def test_state_read_failure(self):
         graph_gateway = FakeGraphGateway(state_error=RuntimeError("DB gone"))
 
-        result = _compact(graph_gateway)
+        result = await _compact(graph_gateway)
         assert result.status == "error"
         assert "Failed to read state" in result.message
 
@@ -49,7 +46,7 @@ class TestCompactGuards:
 class TestCompactCutoffZero:
     """When cutoff == 0, conversation is within retention budget."""
 
-    def test_nothing_to_compact_short_conversation(self):
+    async def test_nothing_to_compact_short_conversation(self):
         msgs = [MagicMock() for _ in range(3)]
         graph_gateway = FakeGraphGateway(state_values={"messages": msgs})
 
@@ -79,7 +76,7 @@ class TestCompactCutoffZero:
                 return_value=500,
             ),
         ):
-            result = _compact(graph_gateway)
+            result = await _compact(graph_gateway)
 
         assert result.status == "noop"
         assert "within the retention budget" in result.message
@@ -89,7 +86,7 @@ class TestCompactCutoffZero:
 class TestCompactNegligibleSavings:
     """When cutoff > 0 but savings are too small to be worth it."""
 
-    def test_skip_when_few_messages_and_low_tokens(self):
+    async def test_skip_when_few_messages_and_low_tokens(self):
         msgs = [MagicMock() for _ in range(15)]
         graph_gateway = FakeGraphGateway(
             state_values={"messages": msgs, "_summarization_event": None}
@@ -126,14 +123,14 @@ class TestCompactNegligibleSavings:
                 side_effect=lambda x: next(token_values),
             ),
         ):
-            result = _compact(graph_gateway)
+            result = await _compact(graph_gateway)
 
         assert result.status == "noop"
         assert "not worth" in result.message
         # No LLM call should have been made
         mock_middleware_inst._acreate_summary.assert_not_called()
 
-    def test_still_compacts_when_few_messages_but_high_tokens(self):
+    async def test_still_compacts_when_few_messages_but_high_tokens(self):
         """2 messages but they account for >2% of tokens — should compact."""
         from langchain_core.messages import HumanMessage
 
@@ -178,7 +175,7 @@ class TestCompactNegligibleSavings:
                 side_effect=lambda x: next(token_values),
             ),
         ):
-            result = _compact(graph_gateway)
+            result = await _compact(graph_gateway)
 
         assert result.status == "ok"
         assert len(graph_gateway.updated_states) == 1
@@ -187,7 +184,7 @@ class TestCompactNegligibleSavings:
 class TestCompactSuccess:
     """Normal compaction flow."""
 
-    def test_manual_threshold_blocks_low_context_compaction(self):
+    async def test_manual_threshold_blocks_low_context_compaction(self):
         msgs = [MagicMock() for _ in range(20)]
         graph_gateway = FakeGraphGateway(
             state_values={"messages": msgs, "_summarization_event": None}
@@ -217,7 +214,7 @@ class TestCompactSuccess:
                 return_value=30_000,
             ),
         ):
-            result = _compact(graph_gateway)
+            result = await _compact(graph_gateway)
 
         assert result.status == "noop"
         assert "40%" in result.message
@@ -225,7 +222,7 @@ class TestCompactSuccess:
         mock_middleware_inst._determine_cutoff_index.assert_not_called()
         mock_middleware_inst._acreate_summary.assert_not_called()
 
-    def test_successful_compaction(self):
+    async def test_successful_compaction(self):
         from langchain_core.messages import HumanMessage
 
         msgs = [MagicMock() for _ in range(20)]
@@ -273,7 +270,7 @@ class TestCompactSuccess:
                 side_effect=lambda x: next(token_values),
             ),
         ):
-            result = _compact(graph_gateway)
+            result = await _compact(graph_gateway)
 
         assert result.status == "ok"
         assert result.messages_compacted == 15
@@ -291,7 +288,7 @@ class TestCompactSuccess:
         assert "_summarization_event" in event_data
         assert event_data["_summarization_event"]["cutoff_index"] == 15
 
-    def test_offload_failure_non_fatal(self):
+    async def test_offload_failure_non_fatal(self):
         """Offload failure should not prevent compaction."""
         from langchain_core.messages import HumanMessage
 
@@ -335,7 +332,7 @@ class TestCompactSuccess:
                 return_value=1000,
             ),
         ):
-            result = _compact(graph_gateway)
+            result = await _compact(graph_gateway)
 
         assert result.status == "ok"
         assert len(graph_gateway.updated_states) == 1
@@ -377,7 +374,7 @@ class TestRenderCompactResult:
 class TestCompactCommandUI:
     """TUI-specific compact progress indicator behavior."""
 
-    def test_command_uses_tui_indicator_when_available(self):
+    async def test_command_uses_tui_indicator_when_available(self):
         from EvoScientist.cli.commands import CompactResult
         from EvoScientist.commands.base import CommandContext
         from EvoScientist.commands.implementation.session import CompactCommand
@@ -413,7 +410,7 @@ class TestCompactCommandUI:
                 return_value="summary-panel",
             ),
         ):
-            _run(CompactCommand().execute(ctx, []))
+            await CompactCommand().execute(ctx, [])
 
         assert ui.started == 1
         assert ui.stopped == 1

tests/test_config.py

@@ -52,12 +52,9 @@ def _restore_dangerous_env():
 def temp_config_dir(tmp_path, monkeypatch):
     """Use a temporary directory for config during tests."""
     config_dir = tmp_path / "evoscientist"
+    monkeypatch.delenv("EVOSCIENTIST_CONFIG_DIR", raising=False)
+    monkeypatch.delenv("EVOSCIENTIST_HOME", raising=False)
     monkeypatch.setenv("XDG_CONFIG_HOME", str(tmp_path))
-    # Prevent load_dotenv from loading the project's real .env file
-    monkeypatch.setattr(
-        "EvoScientist.config.settings.find_dotenv",
-        lambda *a, **k: str(tmp_path / ".env"),
-    )
     # Also clear any API keys from environment
     for key in [
         "ANTHROPIC_API_KEY",
@@ -77,6 +74,9 @@ def temp_config_dir(tmp_path, monkeypatch):
         "EVOSCIENTIST_AUXILIARY_MODEL",
         "EVOSCIENTIST_AUXILIARY_PROVIDER",
         "EVOSCIENTIST_OPENROUTER_ANTHROPIC_PROMPT_CACHE",
+        "EVOSCIENTIST_OPENROUTER_HTTP_REFERER",
+        "EVOSCIENTIST_OPENROUTER_APP_TITLE",
+        "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES",
         "EVOSCIENTIST_DANGEROUS_MODE",
     ]:
         monkeypatch.delenv(key, raising=False)
@@ -104,6 +104,9 @@ def clean_env(monkeypatch):
         "EVOSCIENTIST_AUXILIARY_MODEL",
         "EVOSCIENTIST_AUXILIARY_PROVIDER",
         "EVOSCIENTIST_OPENROUTER_ANTHROPIC_PROMPT_CACHE",
+        "EVOSCIENTIST_OPENROUTER_HTTP_REFERER",
+        "EVOSCIENTIST_OPENROUTER_APP_TITLE",
+        "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES",
         "EVOSCIENTIST_DANGEROUS_MODE",
     ]:
         monkeypatch.delenv(key, raising=False)
@@ -129,8 +132,13 @@ class TestEvoScientistConfig:
         assert config.show_thinking is True
         assert config.ui_backend == "tui"
         assert config.log_level == "warning"
-        assert config.reasoning_effort == "high"
+        assert config.reasoning_effort == ""
         assert config.openrouter_anthropic_prompt_cache is True
+        assert config.openrouter_http_referer == (
+            "https://github.com/EvoScientist/EvoScientist"
+        )
+        assert config.openrouter_app_title == "EvoScientist"
+        assert config.openrouter_app_categories == "creative-writing,personal-agent"
         assert config.memory_profile_enabled is True
         assert config.memory_observations_enabled is True
         assert config.memory_observation_writer == MemoryObservationWriter.ALL
@@ -145,6 +153,8 @@ class TestEvoScientistConfig:
         assert config.channel_debug_tracing is False
         assert config.imessage_enabled is False
         assert config.imessage_allowed_senders == ""
+        assert config.repetitive_tool_call_threshold == 2
+        assert config.max_consecutive_tool_errors == 3
 
     def test_auth_mode_default(self):
         """Test that anthropic_auth_mode defaults to api_key."""
@@ -192,6 +202,18 @@ class TestEvoScientistConfig:
         assert config.dangerous_mode is True
         assert config.auto_approve is True
 
+    @pytest.mark.parametrize(
+        "kwargs",
+        [
+            {"repetitive_tool_call_threshold": -1},
+            {"max_consecutive_tool_errors": -1},
+            {"max_consecutive_tool_errors": True},
+        ],
+    )
+    def test_tool_guard_thresholds_must_be_non_negative_integers(self, kwargs):
+        with pytest.raises(ValueError, match="non-negative integer"):
+            EvoScientistConfig(**kwargs)
+
 
 # =============================================================================
 # Test config path functions
@@ -199,14 +221,34 @@ class TestEvoScientistConfig:
 
 
 class TestConfigPaths:
+    def test_get_config_dir_with_explicit_override(self, monkeypatch, tmp_path):
+        """An explicit config directory has the highest priority."""
+        config_dir = tmp_path / "gateway-config"
+        monkeypatch.setenv("EVOSCIENTIST_CONFIG_DIR", str(config_dir))
+        monkeypatch.setenv("EVOSCIENTIST_HOME", str(tmp_path / "runtime-home"))
+
+        assert get_config_dir() == config_dir.resolve()
+
+    def test_get_config_dir_with_evoscientist_home(self, monkeypatch, tmp_path):
+        """Runtime home keeps configuration and data under one root."""
+        home = tmp_path / "runtime-home"
+        monkeypatch.delenv("EVOSCIENTIST_CONFIG_DIR", raising=False)
+        monkeypatch.setenv("EVOSCIENTIST_HOME", str(home))
+
+        assert get_config_dir() == home.resolve() / "config"
+
     def test_get_config_dir_with_xdg(self, monkeypatch, tmp_path):
         """Test config dir uses XDG_CONFIG_HOME when set."""
+        monkeypatch.delenv("EVOSCIENTIST_CONFIG_DIR", raising=False)
+        monkeypatch.delenv("EVOSCIENTIST_HOME", raising=False)
         monkeypatch.setenv("XDG_CONFIG_HOME", str(tmp_path))
         config_dir = get_config_dir()
         assert config_dir == tmp_path / "evoscientist"
 
     def test_get_config_dir_default(self, monkeypatch):
         """Test config dir defaults to ~/.config/evoscientist."""
+        monkeypatch.delenv("EVOSCIENTIST_CONFIG_DIR", raising=False)
+        monkeypatch.delenv("EVOSCIENTIST_HOME", raising=False)
         monkeypatch.delenv("XDG_CONFIG_HOME", raising=False)
         config_dir = get_config_dir()
         assert config_dir == Path.home() / ".config" / "evoscientist"
@@ -242,6 +284,23 @@ class TestLoadSaveReset:
         assert data["provider"] == "openai"
         assert data["model"] == "gpt-4o"
 
+    def test_save_restricts_config_permissions(self, temp_config_dir, clean_env):
+        """Config file permissions should not depend on the process umask."""
+        original_umask = os.umask(0)
+        try:
+            save_config(EvoScientistConfig(anthropic_api_key="test-key"))
+        finally:
+            os.umask(original_umask)
+
+        config_path = get_config_path()
+        if os.name == "nt":
+            assert config_path.exists()
+            # Windows reports pseudo-permission bits, so we don't test them here.
+            return
+
+        assert config_path.parent.stat().st_mode & 0o777 == 0o700
+        assert config_path.stat().st_mode & 0o777 == 0o600
+
     def test_load_reads_saved_config(self, temp_config_dir, clean_env):
         """Test that load reads previously saved config."""
         original = EvoScientistConfig(
@@ -655,6 +714,22 @@ class TestPriorityChain:
         config = get_effective_config()
         assert config.openrouter_anthropic_prompt_cache is False
 
+    def test_env_openrouter_app_attribution_override(
+        self, temp_config_dir, monkeypatch
+    ):
+        """OpenRouter app-attribution env vars should override file config."""
+        save_config(EvoScientistConfig())
+        monkeypatch.setenv("EVOSCIENTIST_OPENROUTER_HTTP_REFERER", "https://acme.test")
+        monkeypatch.setenv("EVOSCIENTIST_OPENROUTER_APP_TITLE", "Acme")
+        monkeypatch.setenv(
+            "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES", "cli-agent,programming-app"
+        )
+
+        config = get_effective_config()
+        assert config.openrouter_http_referer == "https://acme.test"
+        assert config.openrouter_app_title == "Acme"
+        assert config.openrouter_app_categories == "cli-agent,programming-app"
+
     def test_set_openrouter_anthropic_prompt_cache(self, temp_config_dir, clean_env):
         """Test OpenRouter Anthropic prompt cache can be set through config."""
         save_config(EvoScientistConfig())
@@ -715,6 +790,60 @@ class TestApplyConfigToEnv:
             "false"
         )
 
+    def test_openrouter_app_attribution_applied_to_env(self, clean_env, monkeypatch):
+        """Config app-attribution values are exported to env for models.py."""
+        for env in (
+            "EVOSCIENTIST_OPENROUTER_HTTP_REFERER",
+            "EVOSCIENTIST_OPENROUTER_APP_TITLE",
+            "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES",
+        ):
+            monkeypatch.delenv(env, raising=False)
+        config = EvoScientistConfig(
+            openrouter_http_referer="https://acme.test",
+            openrouter_app_title="Acme",
+            openrouter_app_categories="cli-agent,programming-app",
+        )
+        apply_config_to_env(config)
+
+        assert os.environ.get("EVOSCIENTIST_OPENROUTER_HTTP_REFERER") == (
+            "https://acme.test"
+        )
+        assert os.environ.get("EVOSCIENTIST_OPENROUTER_APP_TITLE") == "Acme"
+        assert os.environ.get("EVOSCIENTIST_OPENROUTER_APP_CATEGORIES") == (
+            "cli-agent,programming-app"
+        )
+
+    def test_openrouter_app_attribution_env_not_overwritten(
+        self, clean_env, monkeypatch
+    ):
+        """apply_config_to_env must not clobber an already-set attribution env var."""
+        monkeypatch.setenv("EVOSCIENTIST_OPENROUTER_APP_TITLE", "existing-title")
+        config = EvoScientistConfig(openrouter_app_title="config-title")
+        apply_config_to_env(config)
+
+        assert os.environ.get("EVOSCIENTIST_OPENROUTER_APP_TITLE") == "existing-title"
+
+    def test_openrouter_app_attribution_empty_config_not_applied(
+        self, clean_env, monkeypatch
+    ):
+        """Empty-string attribution config must not create env vars."""
+        for env in (
+            "EVOSCIENTIST_OPENROUTER_HTTP_REFERER",
+            "EVOSCIENTIST_OPENROUTER_APP_TITLE",
+            "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES",
+        ):
+            monkeypatch.delenv(env, raising=False)
+        config = EvoScientistConfig(
+            openrouter_http_referer="",
+            openrouter_app_title="",
+            openrouter_app_categories="",
+        )
+        apply_config_to_env(config)
+
+        assert os.environ.get("EVOSCIENTIST_OPENROUTER_HTTP_REFERER") is None
+        assert os.environ.get("EVOSCIENTIST_OPENROUTER_APP_TITLE") is None
+        assert os.environ.get("EVOSCIENTIST_OPENROUTER_APP_CATEGORIES") is None
+
     def test_dangerous_mode_round_trips_to_env(self, clean_env, monkeypatch):
         """dangerous_mode set via CLI override must survive a fresh re-read.
 
@@ -832,3 +961,30 @@ def test_scheduler_config_defaults_and_env(monkeypatch):
     assert eff2.memory_skill_synthesis_mode == MemorySkillSynthesisMode.AUTO
     assert eff2.memory_skill_synthesis_cadence == MemorySkillSynthesisCadence.MONTHLY
     assert eff2.memory_skill_synthesis_time == "04:30"
+
+
+# =============================================================================
+# Dotenv isolation (issue #322)
+# =============================================================================
+
+
+class TestDotenvIsolation:
+    def test_env_file_not_leaked_into_process_env(self, tmp_path, monkeypatch):
+        """A .env in cwd must not leak into os.environ during tests.
+
+        Without the suite-wide ``_isolate_dotenv`` fixture,
+        ``get_effective_config`` loads the developer's real .env with
+        ``override=True``; an empty-valued line like ``MINIMAX_BASE_URL=``
+        then poisons ``os.environ.get(key, default)`` lookups for every
+        test that runs afterwards in the same process.
+        """
+        monkeypatch.setenv("XDG_CONFIG_HOME", str(tmp_path))
+        repro_dir = tmp_path / "repro"
+        repro_dir.mkdir()
+        (repro_dir / ".env").write_text("MINIMAX_BASE_URL=\n")
+        monkeypatch.chdir(repro_dir)
+        monkeypatch.delenv("MINIMAX_BASE_URL", raising=False)
+
+        get_effective_config()
+
+        assert "MINIMAX_BASE_URL" not in os.environ

tests/test_configurable_model_middleware.py

@@ -15,7 +15,6 @@ from EvoScientist.middleware.configurable_model import (
     ConfigurableModelMiddleware,
     _read_model_override,
 )
-from tests.conftest import run_async as _run
 
 
 @contextmanager
@@ -134,7 +133,7 @@ class TestPassThrough:
         handler.assert_called_once_with(req)
         req.override.assert_not_called()
 
-    def test_async_no_override_passes_request_unchanged(self):
+    async def test_async_no_override_passes_request_unchanged(self):
         mw = ConfigurableModelMiddleware()
         req = _make_request()
 
@@ -143,7 +142,7 @@ class TestPassThrough:
             return "ok"
 
         with _patched_config({}):
-            result = _run(mw.awrap_model_call(req, handler))
+            result = await mw.awrap_model_call(req, handler)
         assert result == "ok"
         req.override.assert_not_called()
 
@@ -185,7 +184,7 @@ class TestModelOverride:
         assert called_with is not req
         assert called_with.model is new_model
 
-    def test_async_override_path_parity(self):
+    async def test_async_override_path_parity(self):
         mw = ConfigurableModelMiddleware()
         req = _make_request()
         new_model = MagicMock()
@@ -202,7 +201,7 @@ class TestModelOverride:
                 "EvoScientist.llm.get_chat_model", return_value=new_model
             ) as mock_get,
         ):
-            result = _run(mw.awrap_model_call(req, handler))
+            result = await mw.awrap_model_call(req, handler)
 
         assert result == "ok"
         mock_get.assert_called_once_with(model="claude-opus-4-8", provider="anthropic")
@@ -316,7 +315,7 @@ class TestResolveFailure:
         handler.assert_called_once_with(req)
         req.override.assert_not_called()
 
-    def test_async_falls_back_when_resolve_raises(self):
+    async def test_async_falls_back_when_resolve_raises(self):
         mw = ConfigurableModelMiddleware()
         req = _make_request()
 
@@ -333,7 +332,7 @@ class TestResolveFailure:
                 side_effect=ValueError("unknown model"),
             ),
         ):
-            result = _run(mw.awrap_model_call(req, handler))
+            result = await mw.awrap_model_call(req, handler)
 
         assert result == "ok"
         assert called == [req]

tests/test_context_overflow_middleware.py

@@ -66,7 +66,6 @@ def test_wrap_model_call_raises_context_overflow():
     assert handler.call_count == 1
 
 
-@pytest.mark.anyio
 async def test_awrap_model_call_raises_context_overflow():
     # Setup mocks
     msgs = [HumanMessage(content=f"msg {i}") for i in range(10)]
@@ -91,7 +90,6 @@ async def test_awrap_model_call_raises_context_overflow():
     assert handler.call_count == 1
 
 
-@pytest.mark.anyio
 async def test_awrap_model_call_passes_through_other_errors():
     request = ModelRequest(
         messages=[],

tests/test_current_command.py

@@ -2,11 +2,9 @@
 
 from unittest.mock import MagicMock
 
-from tests.conftest import run_async as _run
-
 
 class TestCurrentCommand:
-    def test_prints_thread_workspace_and_memory(self):
+    async def test_prints_thread_workspace_and_memory(self):
         from EvoScientist.commands.base import CommandContext
         from EvoScientist.commands.implementation.general import CurrentCommand
 
@@ -17,14 +15,14 @@ class TestCurrentCommand:
             ui=ui,
             workspace_dir="/tmp/ws",
         )
-        _run(CurrentCommand().execute(ctx, []))
+        await CurrentCommand().execute(ctx, [])
         # Three append_system calls: Thread, Workspace, Memory dir.
         calls = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Thread: abc123" in s for s in calls)
         assert any("Workspace:" in s for s in calls)
         assert any("Memory dir:" in s for s in calls)
 
-    def test_skips_workspace_when_none(self):
+    async def test_skips_workspace_when_none(self):
         from EvoScientist.commands.base import CommandContext
         from EvoScientist.commands.implementation.general import CurrentCommand
 
@@ -35,7 +33,7 @@ class TestCurrentCommand:
             ui=ui,
             workspace_dir=None,
         )
-        _run(CurrentCommand().execute(ctx, []))
+        await CurrentCommand().execute(ctx, [])
         calls = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Thread: abc123" in s for s in calls)
         assert not any("Workspace:" in s for s in calls)

tests/test_delete_command.py

@@ -2,7 +2,6 @@
 
 from unittest.mock import AsyncMock, MagicMock
 
-from tests.conftest import run_async as _run
 from tests.fakes import FakeGraphGateway, FakeThreadStore
 
 
@@ -21,17 +20,17 @@ def _ctx(thread_id="current", thread_store=None):
 
 
 class TestDeleteCommand:
-    def test_refuses_to_delete_current(self):
+    async def test_refuses_to_delete_current(self):
         from EvoScientist.commands.implementation.session import DeleteCommand
 
         thread_store = FakeThreadStore(resolved_thread_id="current", deleted=True)
         ctx, ui = _ctx(thread_id="current", thread_store=thread_store)
-        _run(DeleteCommand().execute(ctx, ["current"]))
+        await DeleteCommand().execute(ctx, ["current"])
         assert ("delete_thread", "current") not in thread_store.calls
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Cannot delete the current session" in m for m in msgs)
 
-    def test_happy_path_success(self):
+    async def test_happy_path_success(self):
         from EvoScientist.commands.implementation.session import DeleteCommand
 
         ctx, ui = _ctx(
@@ -41,45 +40,45 @@ class TestDeleteCommand:
                 deleted=True,
             ),
         )
-        _run(DeleteCommand().execute(ctx, ["other-thread"]))
+        await DeleteCommand().execute(ctx, ["other-thread"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Deleted session other-thread" in m for m in msgs)
 
-    def test_not_found(self):
+    async def test_not_found(self):
         from EvoScientist.commands.implementation.session import DeleteCommand
 
         ctx, ui = _ctx()
-        _run(DeleteCommand().execute(ctx, ["missing"]))
+        await DeleteCommand().execute(ctx, ["missing"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("not found" in m for m in msgs)
 
-    def test_ambiguous_prefix(self):
+    async def test_ambiguous_prefix(self):
         from EvoScientist.commands.implementation.session import DeleteCommand
 
         ctx, ui = _ctx(thread_store=FakeThreadStore(matches=["abc-one", "abc-two"]))
-        _run(DeleteCommand().execute(ctx, ["abc"]))
+        await DeleteCommand().execute(ctx, ["abc"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Ambiguous" in m for m in msgs)
 
-    def test_prefix_resolves_to_unique_match(self):
+    async def test_prefix_resolves_to_unique_match(self):
         from EvoScientist.commands.implementation.session import DeleteCommand
 
         ctx, ui = _ctx(
             thread_store=FakeThreadStore(resolved_thread_id="abc-one", deleted=True)
         )
-        _run(DeleteCommand().execute(ctx, ["abc"]))
+        await DeleteCommand().execute(ctx, ["abc"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Deleted session abc-one" in m for m in msgs)
 
-    def test_no_arg_empty_sessions_prints_notice(self):
+    async def test_no_arg_empty_sessions_prints_notice(self):
         from EvoScientist.commands.implementation.session import DeleteCommand
 
         ctx, ui = _ctx()
-        _run(DeleteCommand().execute(ctx, []))
+        await DeleteCommand().execute(ctx, [])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("No sessions to delete" in m for m in msgs)
 
-    def test_no_arg_calls_picker_returns_none(self):
+    async def test_no_arg_calls_picker_returns_none(self):
         """When no arg and picker returns None, nothing is deleted."""
         from EvoScientist.commands.implementation.session import DeleteCommand
 
@@ -96,5 +95,5 @@ class TestDeleteCommand:
         ]
         store = FakeThreadStore(threads=threads)
         ctx.graph_gateway = FakeGraphGateway(thread_store=store)
-        _run(DeleteCommand().execute(ctx, []))
+        await DeleteCommand().execute(ctx, [])
         ui.wait_for_thread_pick.assert_awaited_once()

tests/test_dingtalk_channel.py

@@ -7,7 +7,6 @@ import pytest
 
 from EvoScientist.channels.base import ChannelError, OutboundMessage
 from EvoScientist.channels.dingtalk.channel import DingTalkChannel, DingTalkConfig
-from tests.conftest import run_async as _run
 
 
 class TestDingTalkConfig:
@@ -41,30 +40,30 @@ class TestDingTalkChannel:
         assert channel._running is False
         assert channel.name == "dingtalk"
 
-    def test_start_raises_without_credentials(self):
+    async def test_start_raises_without_credentials(self):
         config = DingTalkConfig(client_id="", client_secret="")
         channel = DingTalkChannel(config)
         with pytest.raises(ChannelError, match="client_id and client_secret"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_start_raises_without_client_id(self):
+    async def test_start_raises_without_client_id(self):
         config = DingTalkConfig(client_id="", client_secret="secret")
         channel = DingTalkChannel(config)
         with pytest.raises(ChannelError, match="client_id and client_secret"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_start_raises_without_client_secret(self):
+    async def test_start_raises_without_client_secret(self):
         config = DingTalkConfig(client_id="id", client_secret="")
         channel = DingTalkChannel(config)
         with pytest.raises(ChannelError, match="client_id and client_secret"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_stop_when_not_running(self):
+    async def test_stop_when_not_running(self):
         config = DingTalkConfig(client_id="test-id", client_secret="test-secret")
         channel = DingTalkChannel(config)
-        _run(channel.stop())
+        await channel.stop()
 
-    def test_send_returns_false_without_client(self):
+    async def test_send_returns_false_without_client(self):
         config = DingTalkConfig(client_id="test-id", client_secret="test-secret")
         channel = DingTalkChannel(config)
         msg = OutboundMessage(
@@ -73,7 +72,7 @@ class TestDingTalkChannel:
             content="hello",
             metadata={"chat_id": "user123"},
         )
-        result = _run(channel.send(msg))
+        result = await channel.send(msg)
         assert result is False
 
     def test_capabilities(self):
@@ -130,20 +129,20 @@ class TestDingTalkWsMessageParsing:
         channel._token_expires = 9999999999
         return channel
 
-    def test_system_ping_ack(self):
+    async def test_system_ping_ack(self):
         channel = self._make_channel()
         data = {
             "type": "SYSTEM",
             "headers": {"topic": "ping", "messageId": "ping-1"},
             "data": "pong-data",
         }
-        _run(channel._on_ws_message(data))
+        await channel._on_ws_message(data)
         channel._ws_session.send_str.assert_called_once()
         sent = json.loads(channel._ws_session.send_str.call_args[0][0])
         assert sent["code"] == 200
         assert sent["data"] == "pong-data"
 
-    def test_callback_text_message(self):
+    async def test_callback_text_message(self):
         channel = self._make_channel()
         channel._enqueue_raw = AsyncMock()
 
@@ -158,14 +157,14 @@ class TestDingTalkWsMessageParsing:
             "headers": {"messageId": "msg-1", "contentType": "application/json"},
             "data": json.dumps(payload),
         }
-        _run(channel._on_ws_message(data))
+        await channel._on_ws_message(data)
         channel._enqueue_raw.assert_called_once()
         raw = channel._enqueue_raw.call_args[0][0]
         assert raw.text == "hello bot"
         assert raw.sender_id == "staff123"
         assert raw.is_group is False
 
-    def test_callback_group_message_mention(self):
+    async def test_callback_group_message_mention(self):
         channel = self._make_channel()
         channel._enqueue_raw = AsyncMock()
 
@@ -181,12 +180,12 @@ class TestDingTalkWsMessageParsing:
             "headers": {"messageId": "msg-2"},
             "data": json.dumps(payload),
         }
-        _run(channel._on_ws_message(data))
+        await channel._on_ws_message(data)
         raw = channel._enqueue_raw.call_args[0][0]
         assert raw.is_group is True
         assert raw.was_mentioned is True
 
-    def test_callback_group_no_mention(self):
+    async def test_callback_group_no_mention(self):
         channel = self._make_channel()
         channel._enqueue_raw = AsyncMock()
 
@@ -201,12 +200,12 @@ class TestDingTalkWsMessageParsing:
             "headers": {"messageId": "msg-3"},
             "data": json.dumps(payload),
         }
-        _run(channel._on_ws_message(data))
+        await channel._on_ws_message(data)
         raw = channel._enqueue_raw.call_args[0][0]
         assert raw.is_group is True
         assert raw.was_mentioned is False
 
-    def test_ignores_non_callback(self):
+    async def test_ignores_non_callback(self):
         channel = self._make_channel()
         channel._enqueue_raw = AsyncMock()
 
@@ -215,10 +214,10 @@ class TestDingTalkWsMessageParsing:
             "headers": {"messageId": "msg-x"},
             "data": "{}",
         }
-        _run(channel._on_ws_message(data))
+        await channel._on_ws_message(data)
         channel._enqueue_raw.assert_not_called()
 
-    def test_ignores_empty_content(self):
+    async def test_ignores_empty_content(self):
         channel = self._make_channel()
         channel._enqueue_raw = AsyncMock()
 
@@ -232,20 +231,20 @@ class TestDingTalkWsMessageParsing:
             "headers": {"messageId": "msg-e"},
             "data": json.dumps(payload),
         }
-        _run(channel._on_ws_message(data))
+        await channel._on_ws_message(data)
         channel._enqueue_raw.assert_not_called()
 
-    def test_non_dict_data_ignored(self):
+    async def test_non_dict_data_ignored(self):
         channel = self._make_channel()
         channel._enqueue_raw = AsyncMock()
-        _run(channel._on_ws_message("not a dict"))
+        await channel._on_ws_message("not a dict")
         channel._enqueue_raw.assert_not_called()
 
 
 class TestDingTalkSendChunk:
     """Test _send_chunk with mocked HTTP client."""
 
-    def test_send_chunk_calls_api(self):
+    async def test_send_chunk_calls_api(self):
         config = DingTalkConfig(client_id="test-app", client_secret="test-secret")
         channel = DingTalkChannel(config)
         channel._access_token = "fake-token"
@@ -256,7 +255,7 @@ class TestDingTalkSendChunk:
         channel._http_client = MagicMock()
         channel._http_client.post = AsyncMock(return_value=mock_response)
 
-        _run(channel._send_chunk("user1", "formatted", "raw text", None, {}))
+        await channel._send_chunk("user1", "formatted", "raw text", None, {})
         channel._http_client.post.assert_called_once()
         call_args = channel._http_client.post.call_args
         body = call_args.kwargs.get("json") or call_args[1].get("json")
@@ -273,21 +272,21 @@ class TestDingTalkChannelRegistration:
 
 
 class TestDingTalkProbe:
-    def test_missing_credentials(self):
+    async def test_missing_credentials(self):
         from EvoScientist.channels.dingtalk.probe import validate_dingtalk
 
-        ok, msg = _run(validate_dingtalk("", ""))
+        ok, msg = await validate_dingtalk("", "")
         assert ok is False
         assert "required" in msg
 
-    def test_missing_client_id(self):
+    async def test_missing_client_id(self):
         from EvoScientist.channels.dingtalk.probe import validate_dingtalk
 
-        ok, _msg = _run(validate_dingtalk("", "secret"))
+        ok, _msg = await validate_dingtalk("", "secret")
         assert ok is False
 
-    def test_missing_client_secret(self):
+    async def test_missing_client_secret(self):
         from EvoScientist.channels.dingtalk.probe import validate_dingtalk
 
-        ok, _msg = _run(validate_dingtalk("id", ""))
+        ok, _msg = await validate_dingtalk("id", "")
         assert ok is False

tests/test_discord_channel.py

@@ -4,7 +4,6 @@ import pytest
 
 from EvoScientist.channels.base import ChannelError
 from EvoScientist.channels.discord.channel import DiscordChannel, DiscordConfig
-from tests.conftest import run_async as _run
 
 
 class TestDiscordChannel:
@@ -14,18 +13,18 @@ class TestDiscordChannel:
         assert channel.config is config
         assert channel._running is False
 
-    def test_start_raises_without_token_or_library(self):
+    async def test_start_raises_without_token_or_library(self):
         config = DiscordConfig(bot_token="")
         channel = DiscordChannel(config)
         with pytest.raises(ChannelError):
-            _run(channel.start())
+            await channel.start()
 
-    def test_stop_when_not_running(self):
+    async def test_stop_when_not_running(self):
         config = DiscordConfig(bot_token="test")
         channel = DiscordChannel(config)
-        _run(channel.stop())
+        await channel.stop()
 
-    def test_send_returns_false_without_client(self):
+    async def test_send_returns_false_without_client(self):
         from EvoScientist.channels.base import OutboundMessage
 
         config = DiscordConfig(bot_token="test")
@@ -36,5 +35,5 @@ class TestDiscordChannel:
             content="hello",
             metadata={"chat_id": "123"},
         )
-        result = _run(channel.send(msg))
+        result = await channel.send(msg)
         assert result is False

tests/test_error_normalization_middleware.py

@@ -0,0 +1,440 @@
+"""Tests for ErrorNormalizationMiddleware + ProviderStreamError.
+
+Verifies that provider-SDK exceptions from a chat model call get
+wrapped into a non-dataclass ``ProviderStreamError`` at the model
+boundary, and that non-provider exceptions pass through unchanged.
+The provider tag is derived from ``request.model`` (class + base_url),
+not from the raised exception.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import dataclasses
+from types import SimpleNamespace
+
+import pytest
+
+from EvoScientist.llm.errors import (
+    AgentControlError,
+    ModelToolProtocolError,
+    ProviderStreamError,
+)
+from EvoScientist.middleware.error_normalization import (
+    ErrorNormalizationMiddleware,
+    _normalize,
+)
+
+# ---------------------------------------------------------------------------
+# Test fixtures — fake chat model instances + requests
+# ---------------------------------------------------------------------------
+
+
+def _fake_model(module: str, cls_name: str, **attrs):
+    """Build a fake chat model instance whose ``type(model).__module__``
+    matches *module*, carrying arbitrary attributes for ``base_url`` /
+    ``openai_api_base`` / ``anthropic_api_url`` lookup.
+    """
+    cls = type(cls_name, (), {"__module__": module})
+    inst = cls()
+    for k, v in attrs.items():
+        setattr(inst, k, v)
+    return inst
+
+
+def _request(model):
+    """Fake ``ModelRequest`` with just the ``.model`` attribute the
+    middleware reads.
+    """
+    return SimpleNamespace(model=model)
+
+
+def _openai_model(base_url: str | None = None):
+    return _fake_model(
+        "langchain_openai.chat_models.base",
+        "ChatOpenAI",
+        openai_api_base=base_url,
+    )
+
+
+def _anthropic_model(base_url: str | None = None):
+    return _fake_model(
+        "langchain_anthropic.chat_models",
+        "ChatAnthropic",
+        anthropic_api_url=base_url,
+    )
+
+
+def _openrouter_model():
+    return _fake_model("langchain_openrouter.chat_models", "ChatOpenRouter")
+
+
+def _google_model():
+    return _fake_model("langchain_google_genai.chat_models", "ChatGoogleGenerativeAI")
+
+
+def _make_exc(cls_name: str = "APIError", message: str = "boom", **attrs):
+    """Build a plain-Exception subclass carrying arbitrary attributes
+    (``status_code``, ``code``, ``type``, ``request_id`` …).
+    """
+    cls = type(cls_name, (Exception,), attrs)
+    return cls(message)
+
+
+# ---------------------------------------------------------------------------
+# _normalize — provider inference from ModelRequest.model
+# ---------------------------------------------------------------------------
+
+
+class TestNormalize:
+    def test_openai_native_model_tags_openai(self):
+        req = _request(_openai_model())
+        exc = _make_exc(message="rate limited", status_code=429)
+        wrapped = _normalize(req, exc)
+        assert isinstance(wrapped, ProviderStreamError)
+        assert wrapped.provider == "openai"
+        assert wrapped.status_code == 429
+
+    def test_openai_routed_deepseek_tagged_by_base_url(self):
+        req = _request(_openai_model(base_url="https://api.deepseek.com"))
+        wrapped = _normalize(req, _make_exc(message="quota exceeded"))
+        assert wrapped.provider == "deepseek"
+
+    def test_openai_routed_moonshot_tagged_by_base_url(self):
+        req = _request(_openai_model(base_url="https://api.moonshot.cn/v1"))
+        assert _normalize(req, _make_exc()).provider == "moonshot"
+
+    def test_unknown_openai_compat_host_tagged_openai_compat(self):
+        req = _request(_openai_model(base_url="https://internal.corp/v1"))
+        assert _normalize(req, _make_exc()).provider == "openai_compat"
+
+    def test_anthropic_native_model_tags_anthropic(self):
+        req = _request(_anthropic_model(base_url="https://api.anthropic.com"))
+        assert _normalize(req, _make_exc()).provider == "anthropic"
+
+    def test_anthropic_routed_minimax_tagged_by_base_url(self):
+        req = _request(_anthropic_model(base_url="https://api.minimaxi.com/anthropic"))
+        assert _normalize(req, _make_exc()).provider == "minimax"
+
+    def test_unknown_anthropic_compat_host_tagged_anthropic_compat(self):
+        req = _request(_anthropic_model(base_url="https://internal.corp/v1"))
+        assert _normalize(req, _make_exc()).provider == "anthropic_compat"
+
+    def test_openrouter_tagged_from_class_alone(self):
+        req = _request(_openrouter_model())
+        wrapped = _normalize(req, _make_exc(cls_name="UnauthorizedResponseError"))
+        assert wrapped.provider == "openrouter"
+        assert wrapped.class_qualname.endswith(".UnauthorizedResponseError")
+
+    def test_google_genai_tagged_from_class_alone(self):
+        req = _request(_google_model())
+        assert _normalize(req, _make_exc()).provider == "google_genai"
+
+    def test_unrecognized_model_class_returns_none(self):
+        req = _request(_fake_model("some.other.pkg", "SomeModel"))
+        assert _normalize(req, _make_exc()) is None
+
+    def test_missing_model_on_request_returns_none(self):
+        """If the request has no ``.model`` at all (defensive)."""
+        assert _normalize(SimpleNamespace(), _make_exc()) is None
+
+    def test_already_normalized_exception_passes_through(self):
+        """``ModelFallbackMiddleware`` wraps against the failing model
+        before re-raising. The outer chain's ``_normalize`` must NOT
+        double-wrap — otherwise attribution flips back to the original
+        request's model.
+        """
+        req = _request(_openrouter_model())
+        pre_wrapped = ProviderStreamError(
+            provider="moonshot",
+            class_qualname="openai.RateLimitError",
+            message="quota exceeded",
+        )
+        assert _normalize(req, pre_wrapped) is None
+
+    @pytest.mark.parametrize(
+        "error",
+        [
+            AgentControlError("MODEL_TOOL_LOOP_DETECTED", "loop stopped"),
+            ModelToolProtocolError(
+                "missing_name",
+                provider="openai",
+                model="gpt-example",
+                route_key="route-1",
+            ),
+        ],
+    )
+    def test_platform_control_error_passes_through(self, error):
+        req = _request(_openai_model())
+
+        assert _normalize(req, error) is None
+
+
+# ---------------------------------------------------------------------------
+# _is_provider_error — used by tool selector to distinguish provider
+# failures (surface) from shape / config failures (degrade)
+# ---------------------------------------------------------------------------
+
+
+class TestIsProviderError:
+    def test_openai_module_is_provider_error(self):
+        from EvoScientist.middleware.error_normalization import _is_provider_error
+
+        assert _is_provider_error(_make_exc(__module__="openai"))
+
+    def test_httpx_timeout_is_provider_error(self):
+        from EvoScientist.middleware.error_normalization import _is_provider_error
+
+        assert _is_provider_error(
+            _make_exc(cls_name="TimeoutException", __module__="httpx")
+        )
+
+    def test_langchain_wrapper_module_is_provider_error(self):
+        from EvoScientist.middleware.error_normalization import _is_provider_error
+
+        assert _is_provider_error(
+            _make_exc(
+                cls_name="BadRequestError",
+                __module__="langchain_openai.chat_models",
+            )
+        )
+
+    def test_pydantic_validation_is_not_provider_error(self):
+        """Structured-output shape failures come from pydantic /
+        langchain, NOT from a provider SDK — the tool selector's
+        graceful-degrade path is right for these.
+        """
+        from EvoScientist.middleware.error_normalization import _is_provider_error
+
+        assert not _is_provider_error(
+            _make_exc(cls_name="ValidationError", __module__="pydantic")
+        )
+
+    def test_builtin_is_not_provider_error(self):
+        from EvoScientist.middleware.error_normalization import _is_provider_error
+
+        assert not _is_provider_error(RuntimeError("x"))
+
+
+# ---------------------------------------------------------------------------
+# ProviderStreamError envelope
+# ---------------------------------------------------------------------------
+
+
+class TestProviderStreamErrorEnvelope:
+    def test_envelope_contains_required_fields(self):
+        err = ProviderStreamError(
+            provider="deepseek",
+            class_qualname="openai.RateLimitError",
+            message="quota exceeded",
+            status_code=429,
+            code="insufficient_quota",
+        )
+        env = err.as_envelope()
+        assert env["error"] == "RateLimitError"
+        assert env["class"] == "openai.RateLimitError"
+        assert env["message"] == "quota exceeded"
+        assert env["provider"] == "deepseek"
+        assert env["status_code"] == 429
+        assert env["code"] == "insufficient_quota"
+
+    def test_envelope_omits_absent_optional_fields(self):
+        err = ProviderStreamError(
+            provider="openrouter",
+            class_qualname="openrouter.errors.foo.UnauthorizedResponseError",
+            message="User not found.",
+        )
+        env = err.as_envelope()
+        assert "status_code" not in env
+        assert "code" not in env
+        assert "type" not in env
+        assert "request_id" not in env
+
+    def test_provider_stream_error_is_not_a_dataclass(self):
+        """The whole point of the wrapper — must not be a dataclass so
+        orjson's OPT_SERIALIZE_DATACLASS fast-path doesn't fire.
+        """
+        err = ProviderStreamError("x", "y.Z", "msg")
+        assert not dataclasses.is_dataclass(err)
+        assert not dataclasses.is_dataclass(type(err))
+
+    def test_model_dump_returns_envelope(self):
+        """Upstream ``serde.default`` calls ``model_dump()`` before its
+        exception branch — the hook that lets us skip the serde patch.
+        """
+        err = ProviderStreamError(
+            provider="openrouter",
+            class_qualname="openrouter.errors.foo.UnauthorizedResponseError",
+            message="User not found.",
+            status_code=401,
+        )
+        assert err.model_dump() == err.as_envelope()
+
+
+# ---------------------------------------------------------------------------
+# Middleware behavior
+# ---------------------------------------------------------------------------
+
+
+class TestMiddleware:
+    def _run_awrap(self, mw, request, handler):
+        async def _go():
+            return await mw.awrap_model_call(request=request, handler=handler)
+
+        return asyncio.run(_go())
+
+    def test_awrap_normalizes_provider_exception(self):
+        raised = _make_exc(cls_name="UnauthorizedResponseError", message="boom")
+
+        async def handler(_req):
+            raise raised
+
+        req = _request(_openrouter_model())
+        mw = ErrorNormalizationMiddleware()
+        with pytest.raises(ProviderStreamError) as excinfo:
+            self._run_awrap(mw, req, handler)
+        assert excinfo.value.provider == "openrouter"
+        assert excinfo.value.__cause__ is raised
+
+    def test_awrap_passes_through_non_provider_model_exception(self):
+        """If the model isn't a recognized provider SDK, the exception
+        passes through unwrapped — same as any non-model exception.
+        """
+        raised = _make_exc(message="boom")
+
+        async def handler(_req):
+            raise raised
+
+        req = _request(_fake_model("some.other.pkg", "SomeModel"))
+        mw = ErrorNormalizationMiddleware()
+        with pytest.raises(Exception, match="boom") as excinfo:
+            self._run_awrap(mw, req, handler)
+        assert excinfo.value is raised
+
+    def _langgraph_error_samples(self):
+        """Instances covering both branches of ``_should_pass_through``:
+        control-flow (``GraphBubbleUp`` + subclasses) and structural
+        errors. Constructor signatures vary — some need positional
+        args — so build each explicitly.
+        """
+        from langgraph.errors import (
+            EmptyInputError,
+            GraphBubbleUp,
+            GraphInterrupt,
+            InvalidUpdateError,
+            NodeTimeoutError,
+            TaskNotFound,
+        )
+
+        return [
+            GraphBubbleUp(),
+            GraphInterrupt(),
+            InvalidUpdateError("bad update"),
+            EmptyInputError("no input"),
+            TaskNotFound(),
+            NodeTimeoutError("node-x", 1.5, kind="run", run_timeout=1.0),
+        ]
+
+    def test_awrap_passes_through_langgraph_errors(self):
+        """Exceptions from ``langgraph.errors.*`` must propagate
+        untouched even when the model is a recognized provider —
+        they're either control-flow signals (interrupts, HITL) or
+        graph-level structural errors, neither is a provider incident.
+        """
+        req = _request(_openrouter_model())  # recognized — would normally wrap
+        mw = ErrorNormalizationMiddleware()
+
+        for raised in self._langgraph_error_samples():
+
+            async def handler(_req, _r=raised):
+                raise _r
+
+            with pytest.raises(type(raised)) as excinfo:
+                self._run_awrap(mw, req, handler)
+            assert excinfo.value is raised, (
+                f"{type(raised).__name__} got wrapped instead of propagated"
+            )
+
+    def test_awrap_passes_through_context_overflow_error(self):
+        """``ContextOverflowError`` is a cross-layer control signal:
+        deepagents' ``SummarizationMiddleware`` sits outside our stack
+        and catches it by type to compress history and retry. Wrapping
+        it here would change the type and break that self-healing
+        fallback — regressing to a user-visible ``ProviderStreamError``
+        on any long conversation.
+        """
+        from langchain_core.exceptions import ContextOverflowError
+
+        raised = ContextOverflowError("context length exceeded")
+
+        async def handler(_req):
+            raise raised
+
+        req = _request(_openrouter_model())  # recognized — would normally wrap
+        mw = ErrorNormalizationMiddleware()
+        with pytest.raises(ContextOverflowError) as excinfo:
+            self._run_awrap(mw, req, handler)
+        assert excinfo.value is raised
+
+    def test_awrap_preserves_model_tool_protocol_error_identity(self):
+        raised = ModelToolProtocolError(
+            "missing_name",
+            provider="openai",
+            model="gpt-example",
+            route_key="route-1",
+        )
+
+        async def handler(_req):
+            raise raised
+
+        req = _request(_openai_model())
+        with pytest.raises(ModelToolProtocolError) as excinfo:
+            self._run_awrap(ErrorNormalizationMiddleware(), req, handler)
+
+        assert excinfo.value is raised
+        assert excinfo.value.code == "MODEL_TOOL_PROTOCOL_INVALID"
+        assert excinfo.value.fallbackable is True
+
+    def test_awrap_wraps_any_exception_from_recognized_model(self):
+        """Any exception raised inside a call to a provider-recognized
+        model gets wrapped — including builtins like ``RuntimeError``.
+        Rationale: at the middleware boundary we can tell the model is
+        a provider, but not the exception's origin (SDK vs
+        langchain-wrapper vs httpx vs our code). Wrapping uniformly
+        gives the WebUI a consistent envelope; upstream's
+        ``RuntimeError``-whitelist would emit ``{"error":
+        "RuntimeError", "message": str(exc)}`` which isn't more
+        useful.
+        """
+        raised = RuntimeError("internal glitch")
+
+        async def handler(_req):
+            raise raised
+
+        req = _request(_openai_model())
+        mw = ErrorNormalizationMiddleware()
+        with pytest.raises(ProviderStreamError) as excinfo:
+            self._run_awrap(mw, req, handler)
+        assert excinfo.value.provider == "openai"
+        assert excinfo.value.__cause__ is raised
+        assert excinfo.value.class_qualname == "builtins.RuntimeError"
+
+    def test_sync_wrap_normalizes_provider_exception(self):
+        raised = _make_exc(message="boom")
+
+        def handler(_req):
+            raise raised
+
+        req = _request(_openrouter_model())
+        mw = ErrorNormalizationMiddleware()
+        with pytest.raises(ProviderStreamError) as excinfo:
+            mw.wrap_model_call(request=req, handler=handler)
+        assert excinfo.value.provider == "openrouter"
+
+    def test_success_path_returns_handler_result(self):
+        async def handler(_req):
+            return "ok"
+
+        req = _request(_openrouter_model())
+        mw = ErrorNormalizationMiddleware()
+        assert self._run_awrap(mw, req, handler) == "ok"

tests/test_evoskills_command.py

@@ -2,8 +2,6 @@
 
 from unittest.mock import AsyncMock, MagicMock, patch
 
-from tests.conftest import run_async as _run
-
 
 def _ctx(supports_interactive=True):
     from EvoScientist.commands.base import CommandContext
@@ -31,7 +29,7 @@ _INDEX = [
 
 
 class TestInstallSkills:
-    def test_picker_cancel_no_install(self):
+    async def test_picker_cancel_no_install(self):
         from EvoScientist.commands.implementation.skills import InstallSkills
 
         ctx, ui = _ctx()
@@ -45,12 +43,12 @@ class TestInstallSkills:
                 "EvoScientist.tools.skills_manager.install_skill",
             ) as install_mock,
         ):
-            _run(InstallSkills().execute(ctx, []))
+            await InstallSkills().execute(ctx, [])
         install_mock.assert_not_called()
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Browse cancelled" in m for m in msgs)
 
-    def test_picker_returns_selections_installs_each(self):
+    async def test_picker_returns_selections_installs_each(self):
         from EvoScientist.commands.implementation.skills import InstallSkills
 
         ctx, ui = _ctx()
@@ -68,10 +66,10 @@ class TestInstallSkills:
                 return_value={"success": True, "name": "x"},
             ) as install_mock,
         ):
-            _run(InstallSkills().execute(ctx, []))
+            await InstallSkills().execute(ctx, [])
         assert install_mock.call_count == 2
 
-    def test_channel_auto_install_on_tag(self):
+    async def test_channel_auto_install_on_tag(self):
         """Non-interactive UI + tag arg → auto-installs matching skills."""
         from EvoScientist.commands.implementation.skills import InstallSkills
 
@@ -86,12 +84,12 @@ class TestInstallSkills:
                 return_value={"success": True, "name": "x"},
             ) as install_mock,
         ):
-            _run(InstallSkills().execute(ctx, ["core"]))
+            await InstallSkills().execute(ctx, ["core"])
         # "core" matches research-ideation only → 1 install, no picker call
         assert install_mock.call_count == 1
         ui.wait_for_skill_browse.assert_not_called()
 
-    def test_fetch_failure_prints_error(self):
+    async def test_fetch_failure_prints_error(self):
         from EvoScientist.commands.implementation.skills import InstallSkills
 
         ctx, ui = _ctx()
@@ -99,6 +97,6 @@ class TestInstallSkills:
             "EvoScientist.tools.skills_manager.fetch_remote_skill_index",
             side_effect=RuntimeError("network fail"),
         ):
-            _run(InstallSkills().execute(ctx, []))
+            await InstallSkills().execute(ctx, [])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Failed to fetch" in m for m in msgs)

tests/test_exit_command.py

@@ -2,11 +2,9 @@
 
 from unittest.mock import MagicMock
 
-from tests.conftest import run_async as _run
-
 
 class TestExitCommand:
-    def test_execute_calls_force_quit(self):
+    async def test_execute_calls_force_quit(self):
         from EvoScientist.commands.base import CommandContext
         from EvoScientist.commands.implementation.session import ExitCommand
 
@@ -17,7 +15,7 @@ class TestExitCommand:
             ui=ui,
         )
         cmd = ExitCommand()
-        _run(cmd.execute(ctx, []))
+        await cmd.execute(ctx, [])
         ui.force_quit.assert_called_once()
 
     def test_aliases_registered(self):

tests/test_feishu_channel.py

@@ -14,7 +14,6 @@ from EvoScientist.channels.feishu.channel import (
     _parse_inline_elements,
     _parse_inline_text,
 )
-from tests.conftest import run_async as _run
 
 
 class TestFeishuConfig:
@@ -57,24 +56,24 @@ class TestFeishuChannel:
         assert channel._running is False
         assert channel.name == "feishu"
 
-    def test_start_raises_without_app_id(self):
+    async def test_start_raises_without_app_id(self):
         config = FeishuConfig(app_id="", app_secret="test-secret")
         channel = FeishuChannel(config)
         with pytest.raises(ChannelError, match="app_id"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_start_raises_without_app_secret(self):
+    async def test_start_raises_without_app_secret(self):
         config = FeishuConfig(app_id="test-id", app_secret="")
         channel = FeishuChannel(config)
         with pytest.raises(ChannelError, match="app_secret"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_stop_when_not_running(self):
+    async def test_stop_when_not_running(self):
         config = FeishuConfig(app_id="test-id", app_secret="test-secret")
         channel = FeishuChannel(config)
-        _run(channel.stop())
+        await channel.stop()
 
-    def test_send_returns_false_without_client(self):
+    async def test_send_returns_false_without_client(self):
         config = FeishuConfig(app_id="test-id", app_secret="test-secret")
         channel = FeishuChannel(config)
         msg = OutboundMessage(
@@ -83,7 +82,7 @@ class TestFeishuChannel:
             content="hello",
             metadata={"chat_id": "oc_test"},
         )
-        result = _run(channel.send(msg))
+        result = await channel.send(msg)
         assert result is False
 
     def test_capabilities(self):
@@ -201,7 +200,7 @@ class TestFeishuWebhookEvent:
         channel._enqueue_raw = AsyncMock()
         return channel
 
-    def test_text_message_v2(self):
+    async def test_text_message_v2(self):
         channel = self._make_channel()
         event = {
             "sender": {
@@ -217,7 +216,7 @@ class TestFeishuWebhookEvent:
                 "create_time": "1700000000000",
             },
         }
-        _run(channel._on_message(event))
+        await channel._on_message(event)
         channel._enqueue_raw.assert_called_once()
         raw = channel._enqueue_raw.call_args[0][0]
         assert raw.text == "hello feishu"
@@ -225,7 +224,7 @@ class TestFeishuWebhookEvent:
         assert raw.chat_id == "oc_chat1"
         assert raw.is_group is False
 
-    def test_group_message_with_mention(self):
+    async def test_group_message_with_mention(self):
         channel = self._make_channel()
         event = {
             "sender": {
@@ -242,13 +241,13 @@ class TestFeishuWebhookEvent:
                 "mentions": [{"key": "@_user_1", "id": {}}],
             },
         }
-        _run(channel._on_message(event))
+        await channel._on_message(event)
         raw = channel._enqueue_raw.call_args[0][0]
         assert raw.is_group is True
         assert raw.was_mentioned is True
         assert channel._mention_names == ["@_user_1"]
 
-    def test_group_message_no_mention(self):
+    async def test_group_message_no_mention(self):
         channel = self._make_channel()
         event = {
             "sender": {
@@ -264,12 +263,12 @@ class TestFeishuWebhookEvent:
                 "create_time": "1700000000000",
             },
         }
-        _run(channel._on_message(event))
+        await channel._on_message(event)
         raw = channel._enqueue_raw.call_args[0][0]
         assert raw.is_group is True
         assert raw.was_mentioned is False
 
-    def test_skips_bot_messages(self):
+    async def test_skips_bot_messages(self):
         channel = self._make_channel()
         event = {
             "sender": {
@@ -283,10 +282,10 @@ class TestFeishuWebhookEvent:
                 "content": json.dumps({"text": "bot reply"}),
             },
         }
-        _run(channel._on_message(event))
+        await channel._on_message(event)
         channel._enqueue_raw.assert_not_called()
 
-    def test_post_message(self):
+    async def test_post_message(self):
         channel = self._make_channel()
         post_content = {
             "zh_cn": {
@@ -308,12 +307,12 @@ class TestFeishuWebhookEvent:
                 "create_time": "1700000000000",
             },
         }
-        _run(channel._on_message(event))
+        await channel._on_message(event)
         raw = channel._enqueue_raw.call_args[0][0]
         assert "Test" in raw.text
         assert "Post body" in raw.text
 
-    def test_unsupported_msg_type_annotation(self):
+    async def test_unsupported_msg_type_annotation(self):
         channel = self._make_channel()
         event = {
             "sender": {
@@ -329,7 +328,7 @@ class TestFeishuWebhookEvent:
                 "create_time": "1700000000000",
             },
         }
-        _run(channel._on_message(event))
+        await channel._on_message(event)
         raw = channel._enqueue_raw.call_args[0][0]
         assert "share_chat" in raw.text
 
@@ -337,7 +336,7 @@ class TestFeishuWebhookEvent:
 class TestFeishuSendChunk:
     """Test _send_chunk with mocked HTTP client."""
 
-    def test_send_chunk_post_format(self):
+    async def test_send_chunk_post_format(self):
         config = FeishuConfig(app_id="test-app", app_secret="test-secret")
         channel = FeishuChannel(config)
         channel._access_token = "fake-token"
@@ -348,14 +347,14 @@ class TestFeishuSendChunk:
         channel._http_client = MagicMock()
         channel._http_client.post = AsyncMock(return_value=mock_response)
 
-        _run(channel._send_chunk("oc_chat1", "formatted", "raw **text**", None, {}))
+        await channel._send_chunk("oc_chat1", "formatted", "raw **text**", None, {})
         channel._http_client.post.assert_called()
         # Should try post format first
         call_args = channel._http_client.post.call_args
         body = call_args.kwargs.get("json") or call_args[1].get("json")
         assert body["receive_id"] == "oc_chat1"
 
-    def test_send_chunk_with_reply(self):
+    async def test_send_chunk_with_reply(self):
         config = FeishuConfig(app_id="test-app", app_secret="test-secret")
         channel = FeishuChannel(config)
         channel._access_token = "fake-token"
@@ -366,7 +365,7 @@ class TestFeishuSendChunk:
         channel._http_client = MagicMock()
         channel._http_client.post = AsyncMock(return_value=mock_response)
 
-        _run(channel._send_chunk("oc_chat1", "reply", "reply text", "om_reply_id", {}))
+        await channel._send_chunk("oc_chat1", "reply", "reply text", "om_reply_id", {})
         # Should call the reply API endpoint
         first_call_url = channel._http_client.post.call_args_list[0][0][0]
         assert "reply" in first_call_url
@@ -484,17 +483,17 @@ class TestFeishuChannelRegistration:
 
 
 class TestFeishuProbe:
-    def test_missing_app_id(self):
+    async def test_missing_app_id(self):
         from EvoScientist.channels.feishu.probe import validate_feishu_credentials
 
-        ok, msg = _run(validate_feishu_credentials("", "secret"))
+        ok, msg = await validate_feishu_credentials("", "secret")
         assert ok is False
         assert "app_id" in msg
 
-    def test_missing_app_secret(self):
+    async def test_missing_app_secret(self):
         from EvoScientist.channels.feishu.probe import validate_feishu_credentials
 
-        ok, msg = _run(validate_feishu_credentials("id", ""))
+        ok, msg = await validate_feishu_credentials("id", "")
         assert ok is False
         assert "app_secret" in msg
 
@@ -510,7 +509,7 @@ class TestFeishuWebSocketMode:
         )
         assert config.subscription_mode == "websocket"
 
-    def test_start_websocket_raises_without_lark_oapi(self):
+    async def test_start_websocket_raises_without_lark_oapi(self):
         config = FeishuConfig(
             app_id="test-id",
             app_secret="test-secret",
@@ -520,9 +519,9 @@ class TestFeishuWebSocketMode:
         # Temporarily hide lark_oapi if it's installed
         with patch.dict(sys.modules, {"lark_oapi": None}):
             with pytest.raises(ChannelError, match="lark-oapi"):
-                _run(channel.start())
+                await channel.start()
 
-    def test_start_webhook_mode_still_works(self):
+    async def test_start_webhook_mode_still_works(self):
         """Ensure subscription_mode='webhook' still validates as before."""
         config = FeishuConfig(
             app_id="",
@@ -531,9 +530,9 @@ class TestFeishuWebSocketMode:
         )
         channel = FeishuChannel(config)
         with pytest.raises(ChannelError, match="app_id"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_invalid_subscription_mode_raises(self):
+    async def test_invalid_subscription_mode_raises(self):
         config = FeishuConfig(
             app_id="test-id",
             app_secret="test-secret",
@@ -541,9 +540,9 @@ class TestFeishuWebSocketMode:
         )
         channel = FeishuChannel(config)
         with pytest.raises(ChannelError, match="Invalid feishu_subscription_mode"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_on_lark_sdk_message_bridges_to_on_message(self):
+    async def test_on_lark_sdk_message_bridges_to_on_message(self):
         """Test that _on_lark_sdk_message enqueues event dict via queue."""
         import queue as queue_mod
 
@@ -594,14 +593,14 @@ class TestFeishuWebSocketMode:
         )
 
         # Verify the consumer processes it correctly
-        _run(channel._on_message(event_dict))
+        await channel._on_message(event_dict)
         channel._enqueue_raw.assert_called_once()
         raw = channel._enqueue_raw.call_args[0][0]
         assert raw.text == "hello from websocket"
         assert raw.sender_id == "ou_test_ws"
         assert raw.is_group is False
 
-    def test_cleanup_websocket_mode(self):
+    async def test_cleanup_websocket_mode(self):
         config = FeishuConfig(
             app_id="test-id",
             app_secret="test-secret",
@@ -617,7 +616,7 @@ class TestFeishuWebSocketMode:
         channel._ws_consumer_task = None
         channel._access_token = "fake-token"
 
-        _run(channel._cleanup())
+        await channel._cleanup()
 
         mock_client.aclose.assert_called_once()
         assert channel._http_client is None

tests/test_gateway_background_runs.py

@@ -1,6 +1,5 @@
 from __future__ import annotations
 
-import asyncio
 from types import SimpleNamespace
 from unittest.mock import MagicMock
 
@@ -179,7 +178,7 @@ def test_launch_background_run_deletes_thread_when_run_creation_fails(monkeypatc
     fake_client.threads.delete.assert_called_once_with("thread-1")
 
 
-def test_async_launch_background_run_deletes_thread_when_run_creation_fails(
+async def test_async_launch_background_run_deletes_thread_when_run_creation_fails(
     monkeypatch,
 ):
     monkeypatch.setattr(
@@ -204,11 +203,8 @@ def test_async_launch_background_run_deletes_thread_when_run_creation_fails(
         lambda **_kwargs: SimpleNamespace(threads=_Threads(), runs=_Runs()),
     )
 
-    async def run() -> None:
-        with pytest.raises(RuntimeError, match="run creation failed"):
-            await background_runs.alaunch_background_run(_request())
-
-    asyncio.run(run())
+    with pytest.raises(RuntimeError, match="run creation failed"):
+        await background_runs.alaunch_background_run(_request())
 
     assert deleted == ["thread-1"]
 
@@ -278,7 +274,7 @@ def test_sync_status_watcher_preserves_thread_on_poll_failure(
     assert deleted == []
 
 
-def test_async_status_watcher_aborts_and_deletes_thread_on_error_status():
+async def test_async_status_watcher_aborts_and_deletes_thread_on_error_status():
     finished: list[background_runs.BackgroundRun] = []
     aborted: list[background_runs.BackgroundRun] = []
     deleted: list[str] = []
@@ -291,29 +287,26 @@ def test_async_status_watcher_aborts_and_deletes_thread_on_error_status():
         async def delete(self, thread_id: str):
             deleted.append(thread_id)
 
-    async def run() -> None:
-        await background_runs.awatch_background_run(
-            SimpleNamespace(runs=_Runs(), threads=_Threads()),
-            thread_id="thread-1",
-            run_id="run-1",
-            name="test worker",
-            hooks=background_runs.BackgroundRunHooks(
-                on_finished=finished.append,
-                on_aborted=aborted.append,
-            ),
-            watcher_config=background_runs.BackgroundRunWatcherConfig(
-                poll_interval_seconds=0,
-            ),
-        )
-
-    asyncio.run(run())
+    await background_runs.awatch_background_run(
+        SimpleNamespace(runs=_Runs(), threads=_Threads()),
+        thread_id="thread-1",
+        run_id="run-1",
+        name="test worker",
+        hooks=background_runs.BackgroundRunHooks(
+            on_finished=finished.append,
+            on_aborted=aborted.append,
+        ),
+        watcher_config=background_runs.BackgroundRunWatcherConfig(
+            poll_interval_seconds=0,
+        ),
+    )
 
     assert finished == []
     assert [run.run_id for run in aborted] == ["run-1"]
     assert deleted == ["thread-1"]
 
 
-def test_async_status_watcher_preserves_run_url():
+async def test_async_status_watcher_preserves_run_url():
     finished: list[background_runs.BackgroundRun] = []
 
     class _Runs:
@@ -324,21 +317,18 @@ def test_async_status_watcher_preserves_run_url():
         async def delete(self, _thread_id: str):
             return None
 
-    async def run() -> None:
-        await background_runs.awatch_background_run(
-            SimpleNamespace(runs=_Runs(), threads=_Threads()),
-            url="http://worker.example",
-            thread_id="thread-1",
-            run_id="run-1",
-            name="test worker",
-            hooks=background_runs.BackgroundRunHooks(
-                on_finished=finished.append,
-            ),
-            watcher_config=background_runs.BackgroundRunWatcherConfig(
-                poll_interval_seconds=0,
-            ),
-        )
-
-    asyncio.run(run())
+    await background_runs.awatch_background_run(
+        SimpleNamespace(runs=_Runs(), threads=_Threads()),
+        url="http://worker.example",
+        thread_id="thread-1",
+        run_id="run-1",
+        name="test worker",
+        hooks=background_runs.BackgroundRunHooks(
+            on_finished=finished.append,
+        ),
+        watcher_config=background_runs.BackgroundRunWatcherConfig(
+            poll_interval_seconds=0,
+        ),
+    )
 
     assert [run.url for run in finished] == ["http://worker.example"]

tests/test_graph_gateway.py

@@ -20,7 +20,6 @@ from EvoScientist.gateway import (
 )
 from EvoScientist.gateway.server import _THREAD_SEARCH_LIMIT
 from EvoScientist.stream import display as display_mod
-from tests.conftest import run_async
 from tests.fakes import (
     FakeGraphGateway,
     FakeLangGraphClient,
@@ -30,7 +29,7 @@ from tests.fakes import (
 )
 
 
-def test_local_gateway_streams_from_injected_streamer():
+async def test_local_gateway_streams_from_injected_streamer():
     seen: dict[str, Any] = {}
 
     async def _streamer(agent, message, thread_id, **kwargs):
@@ -60,7 +59,7 @@ def test_local_gateway_streams_from_injected_streamer():
         return [event async for event in gateway.stream_events(request)]
 
     with patch("EvoScientist.stream.events.stream_agent_events", new=_streamer):
-        events = run_async(_collect())
+        events = await _collect()
 
     assert events == [
         {"type": "text", "content": "hi"},
@@ -75,7 +74,7 @@ def test_local_gateway_streams_from_injected_streamer():
     }
 
 
-def test_local_graph_gateway_delegates_thread_operations():
+async def test_local_graph_gateway_delegates_thread_operations():
     thread_store = FakeThreadStore(
         generated_thread_id="new12345",
         threads=[{"thread_id": "abc12345"}],
@@ -102,7 +101,7 @@ def test_local_graph_gateway_delegates_thread_operations():
             "deleted": await gateway.delete_thread("abc12345"),
         }
 
-    result = run_async(_run())
+    result = await _run()
 
     assert result["created"] == "new12345"
     assert result["threads"] == [{"thread_id": "abc12345"}]
@@ -132,16 +131,14 @@ def test_local_graph_gateway_delegates_thread_operations():
     ]
 
 
-def test_local_graph_gateway_reads_state_values():
+async def test_local_graph_gateway_reads_state_values():
     agent = MagicMock()
     agent.aget_state = AsyncMock(
         return_value=SimpleNamespace(values={"async_tasks": {"task-1": {}}})
     )
     gateway = LocalGraphGateway()
 
-    values = run_async(
-        gateway.get_state_values(GraphTarget(local_graph=agent), "abc12345")
-    )
+    values = await gateway.get_state_values(GraphTarget(local_graph=agent), "abc12345")
 
     assert values == {"async_tasks": {"task-1": {}}}
     agent.aget_state.assert_awaited_once_with(
@@ -149,17 +146,15 @@ def test_local_graph_gateway_reads_state_values():
     )
 
 
-def test_local_graph_gateway_updates_state_values():
+async def test_local_graph_gateway_updates_state_values():
     agent = MagicMock()
     agent.aupdate_state = AsyncMock()
     gateway = LocalGraphGateway()
 
-    run_async(
-        gateway.update_state_values(
-            GraphTarget(local_graph=agent),
-            "abc12345",
-            {"_summarization_event": {"cutoff_index": 2}},
-        )
+    await gateway.update_state_values(
+        GraphTarget(local_graph=agent),
+        "abc12345",
+        {"_summarization_event": {"cutoff_index": 2}},
     )
 
     agent.aupdate_state.assert_awaited_once_with(
@@ -169,7 +164,7 @@ def test_local_graph_gateway_updates_state_values():
     )
 
 
-def test_local_stream_events_delegates_aclose_to_inner():
+async def test_local_stream_events_delegates_aclose_to_inner():
     cleanup_ran = False
 
     async def _streamer(_agent, _message, _thread_id, **_kwargs):
@@ -194,7 +189,7 @@ def test_local_stream_events_delegates_aclose_to_inner():
         assert cleanup_ran is True
 
     with patch("EvoScientist.stream.events.stream_agent_events", new=_streamer):
-        run_async(_run())
+        await _run()
 
 
 def test_run_streaming_can_consume_injected_gateway():
@@ -228,7 +223,7 @@ def test_run_streaming_can_consume_injected_gateway():
     ]
 
 
-def test_resume_command_consumes_context_gateway():
+async def test_resume_command_consumes_context_gateway():
     from EvoScientist.commands.base import CommandContext
     from EvoScientist.commands.implementation.session import ResumeCommand
 
@@ -246,7 +241,7 @@ def test_resume_command_consumes_context_gateway():
         graph_gateway=FakeGraphGateway(thread_store=thread_store),
     )
 
-    run_async(ResumeCommand().execute(ctx, ["abc"]))
+    await ResumeCommand().execute(ctx, ["abc"])
 
     assert ctx.thread_id == "abc12345"
     assert ctx.workspace_dir == "/restored"
@@ -287,7 +282,7 @@ def test_cmd_run_passes_local_graph_gateway(monkeypatch):
     assert seen["gateway"].thread_store is thread_store
 
 
-def test_langgraph_server_thread_store_delegates_to_sdk_threads():
+async def test_langgraph_server_thread_store_delegates_to_sdk_threads():
     threads = FakeLangGraphThreadsClient(
         threads=[
             {
@@ -335,7 +330,7 @@ def test_langgraph_server_thread_store_delegates_to_sdk_threads():
             "deleted": await store.delete_thread("abc12345"),
         }
 
-    result = run_async(_run())
+    result = await _run()
 
     assert result["created"] == "server-thread"
     assert len(threads.created) == 1
@@ -379,7 +374,7 @@ def test_langgraph_server_thread_store_delegates_to_sdk_threads():
     assert threads.deleted == ["abc12345"]
 
 
-def test_langgraph_server_thread_store_limit_zero_pages_all_threads():
+async def test_langgraph_server_thread_store_limit_zero_pages_all_threads():
     rows = [
         {
             "thread_id": f"thread-{index}",
@@ -392,7 +387,7 @@ def test_langgraph_server_thread_store_limit_zero_pages_all_threads():
         client=FakeLangGraphClient(threads),
     )
 
-    result = run_async(store.list_threads(limit=0))
+    result = await store.list_threads(limit=0)
 
     assert [row["thread_id"] for row in result] == [
         f"thread-{index}" for index in range(_THREAD_SEARCH_LIMIT + 1)
@@ -403,7 +398,7 @@ def test_langgraph_server_thread_store_limit_zero_pages_all_threads():
     ]
 
 
-def test_langgraph_server_thread_store_positive_limit_uses_single_search():
+async def test_langgraph_server_thread_store_positive_limit_uses_single_search():
     threads = FakeLangGraphThreadsClient(
         threads=[
             {
@@ -417,7 +412,7 @@ def test_langgraph_server_thread_store_positive_limit_uses_single_search():
         client=FakeLangGraphClient(threads),
     )
 
-    result = run_async(store.list_threads(limit=2))
+    result = await store.list_threads(limit=2)
 
     assert [row["thread_id"] for row in result] == ["thread-0", "thread-1"]
     assert [(search["limit"], search["offset"]) for search in threads.searches] == [
@@ -425,7 +420,7 @@ def test_langgraph_server_thread_store_positive_limit_uses_single_search():
     ]
 
 
-def test_langgraph_server_thread_store_prefix_resolution_skips_exact_lookup():
+async def test_langgraph_server_thread_store_prefix_resolution_skips_exact_lookup():
     threads = FakeLangGraphThreadsClient(
         threads=[
             {
@@ -438,14 +433,14 @@ def test_langgraph_server_thread_store_prefix_resolution_skips_exact_lookup():
         client=FakeLangGraphClient(threads),
     )
 
-    result = run_async(store.resolve_thread_id_prefix("abc"))
+    result = await store.resolve_thread_id_prefix("abc")
 
     assert result == ("abc12345", [])
     assert threads.gets == []
     assert len(threads.searches) == 1
 
 
-def test_langgraph_server_thread_store_prefix_resolution_pages_all_threads():
+async def test_langgraph_server_thread_store_prefix_resolution_pages_all_threads():
     rows = [
         {
             "thread_id": f"thread-{index}",
@@ -464,7 +459,7 @@ def test_langgraph_server_thread_store_prefix_resolution_pages_all_threads():
         client=FakeLangGraphClient(threads),
     )
 
-    result = run_async(store.resolve_thread_id_prefix("older-thread"))
+    result = await store.resolve_thread_id_prefix("older-thread")
 
     assert result == ("older-thread-match", [])
     assert [(search["limit"], search["offset"]) for search in threads.searches] == [
@@ -473,7 +468,7 @@ def test_langgraph_server_thread_store_prefix_resolution_pages_all_threads():
     ]
 
 
-def test_langgraph_server_thread_store_uuid_resolution_uses_exact_lookup():
+async def test_langgraph_server_thread_store_uuid_resolution_uses_exact_lookup():
     thread_id = "019ed9e4-4253-7f62-b50f-f0470a4b3c9f"
     threads = FakeLangGraphThreadsClient(
         threads=[
@@ -487,14 +482,14 @@ def test_langgraph_server_thread_store_uuid_resolution_uses_exact_lookup():
         client=FakeLangGraphClient(threads),
     )
 
-    result = run_async(store.resolve_thread_id_prefix(thread_id))
+    result = await store.resolve_thread_id_prefix(thread_id)
 
     assert result == (thread_id, [])
     assert threads.gets == [thread_id]
     assert threads.searches == []
 
 
-def test_langgraph_server_thread_store_uuid_resolution_filters_graph_id():
+async def test_langgraph_server_thread_store_uuid_resolution_filters_graph_id():
     thread_id = "019ed9e4-4253-7f62-b50f-f0470a4b3c9f"
     threads = FakeLangGraphThreadsClient(
         threads=[
@@ -508,7 +503,7 @@ def test_langgraph_server_thread_store_uuid_resolution_filters_graph_id():
         client=FakeLangGraphClient(threads),
     )
 
-    result = run_async(store.resolve_thread_id_prefix(thread_id))
+    result = await store.resolve_thread_id_prefix(thread_id)
 
     assert result == (None, [])
     assert threads.gets == [thread_id]
@@ -517,7 +512,7 @@ def test_langgraph_server_thread_store_uuid_resolution_filters_graph_id():
     ]
 
 
-def test_langgraph_server_thread_store_clones_thread_with_metadata():
+async def test_langgraph_server_thread_store_clones_thread_with_metadata():
     clone_metadata = {
         "clone_purpose": "memory_extraction",
         "source_thread_id": "source-thread",
@@ -534,8 +529,8 @@ def test_langgraph_server_thread_store_clones_thread_with_metadata():
         client=FakeLangGraphClient(threads),
     )
 
-    cloned_thread_id = run_async(
-        store.clone_thread("source-thread", metadata=clone_metadata)
+    cloned_thread_id = await store.clone_thread(
+        "source-thread", metadata=clone_metadata
     )
 
     assert cloned_thread_id == "source-thread-copy"
@@ -552,7 +547,7 @@ def test_langgraph_server_thread_store_clones_thread_with_metadata():
     }
 
 
-def test_langgraph_server_thread_store_rejects_copy_without_thread_id():
+async def test_langgraph_server_thread_store_rejects_copy_without_thread_id():
     threads = FakeLangGraphThreadsClient(
         threads=[{"thread_id": "source-thread", "metadata": {"graph_id": "agent"}}],
         copy_response=None,
@@ -565,10 +560,10 @@ def test_langgraph_server_thread_store_rejects_copy_without_thread_id():
         await store.clone_thread("source-thread")
 
     with pytest.raises(RuntimeError, match="did not return a cloned thread id"):
-        run_async(_run())
+        await _run()
 
 
-def test_langgraph_server_gateway_clones_thread():
+async def test_langgraph_server_gateway_clones_thread():
     threads = FakeLangGraphThreadsClient(
         threads=[{"thread_id": "source-thread", "metadata": {"graph_id": "agent"}}]
     )
@@ -578,12 +573,10 @@ def test_langgraph_server_gateway_clones_thread():
         )
     )
 
-    cloned_thread_id = run_async(
-        gateway.clone_thread(
-            "source-thread",
-            metadata={"clone_purpose": "manual"},
-            target=GraphTarget(graph_id="agent"),
-        )
+    cloned_thread_id = await gateway.clone_thread(
+        "source-thread",
+        metadata={"clone_purpose": "manual"},
+        target=GraphTarget(graph_id="agent"),
     )
 
     assert cloned_thread_id == "source-thread-copy"
@@ -592,12 +585,12 @@ def test_langgraph_server_gateway_clones_thread():
     ]
 
 
-def test_local_graph_gateway_clone_thread_is_explicitly_unsupported():
+async def test_local_graph_gateway_clone_thread_is_explicitly_unsupported():
     async def _run():
         await LocalGraphGateway().clone_thread("source-thread")
 
     with pytest.raises(NotImplementedError, match="does not support thread cloning"):
-        run_async(_run())
+        await _run()
 
 
 def test_runtime_gateways_can_use_langgraph_server_backend():
@@ -616,7 +609,7 @@ def test_runtime_gateways_can_use_langgraph_server_backend():
     assert gateway.thread_store is runtime_gateways.thread_store
 
 
-def test_langgraph_server_gateway_reads_state_values():
+async def test_langgraph_server_gateway_reads_state_values():
     threads = FakeLangGraphThreadsClient(
         threads=[{"thread_id": "abc12345", "metadata": {"graph_id": "EvoScientist"}}],
         states={"abc12345": {"values": {"async_tasks": {"task-1": {}}}}},
@@ -627,12 +620,12 @@ def test_langgraph_server_gateway_reads_state_values():
         )
     )
 
-    values = run_async(gateway.get_state_values(GraphTarget(), "abc12345"))
+    values = await gateway.get_state_values(GraphTarget(), "abc12345")
 
     assert values == {"async_tasks": {"task-1": {}}}
 
 
-def test_langgraph_server_gateway_messages_apply_summarization_event():
+async def test_langgraph_server_gateway_messages_apply_summarization_event():
     threads = FakeLangGraphThreadsClient(
         threads=[{"thread_id": "abc12345", "metadata": {"graph_id": "EvoScientist"}}],
         states={
@@ -658,7 +651,7 @@ def test_langgraph_server_gateway_messages_apply_summarization_event():
         )
     )
 
-    messages = run_async(gateway.get_thread_messages("abc12345"))
+    messages = await gateway.get_thread_messages("abc12345")
 
     assert len(messages) == 2
     assert isinstance(messages[0], AIMessage)
@@ -667,7 +660,7 @@ def test_langgraph_server_gateway_messages_apply_summarization_event():
     assert messages[1].content == "third"
 
 
-def test_langgraph_server_gateway_updates_state_values():
+async def test_langgraph_server_gateway_updates_state_values():
     threads = FakeLangGraphThreadsClient(
         threads=[{"thread_id": "abc12345", "metadata": {"graph_id": "EvoScientist"}}],
     )
@@ -677,12 +670,10 @@ def test_langgraph_server_gateway_updates_state_values():
         )
     )
 
-    run_async(
-        gateway.update_state_values(
-            GraphTarget(),
-            "abc12345",
-            {"_summarization_event": {"cutoff_index": 2}},
-        )
+    await gateway.update_state_values(
+        GraphTarget(),
+        "abc12345",
+        {"_summarization_event": {"cutoff_index": 2}},
     )
 
     assert threads.state_updates == [
@@ -690,7 +681,7 @@ def test_langgraph_server_gateway_updates_state_values():
     ]
 
 
-def test_langgraph_server_gateway_streams_root_protocol_events():
+async def test_langgraph_server_gateway_streams_root_protocol_events():
     stream = FakeLangGraphThreadStream(
         "abc12345",
         events=[
@@ -737,7 +728,7 @@ def test_langgraph_server_gateway_streams_root_protocol_events():
             )
         ]
 
-    events = run_async(_collect())
+    events = await _collect()
 
     assert len(threads.created) == 1
     assert threads.created[0]["thread_id"] == "abc12345"
@@ -804,7 +795,7 @@ def _root_message_finish() -> dict[str, object]:
     }
 
 
-def _collect_server_gateway_stream(
+async def _collect_server_gateway_stream(
     events: list[dict[str, object]],
     *,
     state_messages: list[dict[str, object]] | None = None,
@@ -832,11 +823,11 @@ def _collect_server_gateway_stream(
             )
         ]
 
-    return run_async(_collect())
+    return await _collect()
 
 
-def test_langgraph_server_gateway_streams_value_message_snapshots():
-    events = _collect_server_gateway_stream(
+async def test_langgraph_server_gateway_streams_value_message_snapshots():
+    events = await _collect_server_gateway_stream(
         [
             _value_snapshot([_OLD_AI, _HUMAN]),
             _value_snapshot([_OLD_AI, _HUMAN, _NEW_AI]),
@@ -850,8 +841,8 @@ def test_langgraph_server_gateway_streams_value_message_snapshots():
     ]
 
 
-def test_langgraph_server_gateway_values_do_not_duplicate_message_stream():
-    events = _collect_server_gateway_stream(
+async def test_langgraph_server_gateway_values_do_not_duplicate_message_stream():
+    events = await _collect_server_gateway_stream(
         [
             _root_text_delta("new"),
             _root_message_finish(),
@@ -866,8 +857,8 @@ def test_langgraph_server_gateway_values_do_not_duplicate_message_stream():
     ]
 
 
-def test_langgraph_server_gateway_ignores_non_root_value_messages():
-    events = _collect_server_gateway_stream(
+async def test_langgraph_server_gateway_ignores_non_root_value_messages():
+    events = await _collect_server_gateway_stream(
         [
             _value_snapshot(
                 [{"type": "ai", "content": "subagent text", "id": "subagent-ai"}],
@@ -880,7 +871,7 @@ def test_langgraph_server_gateway_ignores_non_root_value_messages():
     assert events[-1] == {"type": "done", "content": "", "response": ""}
 
 
-def test_langgraph_server_gateway_emits_state_interrupt_before_done():
+async def test_langgraph_server_gateway_emits_state_interrupt_before_done():
     stream = FakeLangGraphThreadStream(
         "abc12345",
         events=[],
@@ -930,9 +921,15 @@ def test_langgraph_server_gateway_emits_state_interrupt_before_done():
             )
         ]
 
-    events = run_async(_collect())
+    events = await _collect()
 
     assert events == [
+        {
+            "type": "tool_call",
+            "name": "execute",
+            "args": {"command": "echo hello"},
+            "id": "tool-1",
+        },
         {
             "type": "interrupt",
             "interrupt_id": "interrupt-1",
@@ -954,7 +951,7 @@ def test_langgraph_server_gateway_emits_state_interrupt_before_done():
     ]
 
 
-def test_langgraph_server_gateway_streams_subagent_protocol_events():
+async def test_langgraph_server_gateway_streams_subagent_protocol_events():
     stream = FakeLangGraphThreadStream(
         "abc12345",
         events=[
@@ -1003,7 +1000,7 @@ def test_langgraph_server_gateway_streams_subagent_protocol_events():
             )
         ]
 
-    events = run_async(_collect())
+    events = await _collect()
 
     assert events == [
         {
@@ -1028,7 +1025,7 @@ def test_langgraph_server_gateway_streams_subagent_protocol_events():
     ]
 
 
-def test_langgraph_server_gateway_resumes_interrupt_with_thread_stream():
+async def test_langgraph_server_gateway_resumes_interrupt_with_thread_stream():
     from langgraph.types import Command
 
     stream = FakeLangGraphThreadStream(
@@ -1058,7 +1055,7 @@ def test_langgraph_server_gateway_resumes_interrupt_with_thread_stream():
             )
         ]
 
-    events = run_async(_collect())
+    events = await _collect()
 
     assert stream.run.starts == []
     assert stream.run.responses == [

tests/test_hitl.py

@@ -378,7 +378,7 @@ class TestHitlConfig:
 
 
 class TestInterruptEventParsing:
-    def test_interrupt_from_updates_mode(self):
+    async def test_interrupt_from_updates_mode(self):
         """__interrupt__ in updates mode yields interrupt event."""
         interrupt_data = {
             "__interrupt__": [
@@ -405,7 +405,7 @@ class TestInterruptEventParsing:
                 protocol_event("updates", interrupt_data),
             ]
         )
-        events = collect_events(agent, message="test", thread_id="thread-1")
+        events = await collect_events(agent, message="test", thread_id="thread-1")
 
         types = [e["type"] for e in events]
         assert "interrupt" in types
@@ -415,14 +415,14 @@ class TestInterruptEventParsing:
         assert interrupt_ev["action_requests"][0]["name"] == "execute"
         assert interrupt_ev["interrupt_id"] == "main"
 
-    def test_updates_without_interrupt_skipped(self):
+    async def test_updates_without_interrupt_skipped(self):
         """Regular updates mode data is skipped as before."""
         agent = FakeV3Agent(
             [
                 protocol_event("updates", {"some_node": {"key": "value"}}),
             ]
         )
-        events = collect_events(agent, message="test", thread_id="thread-1")
+        events = await collect_events(agent, message="test", thread_id="thread-1")
 
         types = [e["type"] for e in events]
         assert "interrupt" not in types

tests/test_host_metering_extensions.py

@@ -0,0 +1,13 @@
+from EvoScientist.llm.errors import AgentControlError
+from EvoScientist.middleware.model_fallback import _is_non_fallbackable
+
+
+def test_agent_control_error_is_non_fallbackable():
+    error = AgentControlError(
+        "INSUFFICIENT_BALANCE",
+        "balance unavailable",
+        status_code=403,
+    )
+
+    assert "platform control error" in (_is_non_fallbackable(error) or "")
+    assert error.model_dump()["code"] == "INSUFFICIENT_BALANCE"

tests/test_install_skill_command.py

@@ -2,8 +2,6 @@
 
 from unittest.mock import MagicMock, patch
 
-from tests.conftest import run_async as _run
-
 
 def _ctx():
     from EvoScientist.commands.base import CommandContext
@@ -14,15 +12,15 @@ def _ctx():
 
 
 class TestInstallSkill:
-    def test_usage_message_when_no_args(self):
+    async def test_usage_message_when_no_args(self):
         from EvoScientist.commands.implementation.skills import InstallSkill
 
         ctx, ui = _ctx()
-        _run(InstallSkill().execute(ctx, []))
+        await InstallSkill().execute(ctx, [])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Usage:" in m for m in msgs)
 
-    def test_happy_path(self):
+    async def test_happy_path(self):
         from EvoScientist.commands.implementation.skills import InstallSkill
 
         ctx, ui = _ctx()
@@ -35,21 +33,21 @@ class TestInstallSkill:
                 "path": "/tmp/demo",
             },
         ):
-            _run(InstallSkill().execute(ctx, ["./some-path"]))
+            await InstallSkill().execute(ctx, ["./some-path"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Installed: demo-skill" in m for m in msgs)
 
 
 class TestUninstallSkill:
-    def test_usage_message_when_no_args(self):
+    async def test_usage_message_when_no_args(self):
         from EvoScientist.commands.implementation.skills import UninstallSkill
 
         ctx, ui = _ctx()
-        _run(UninstallSkill().execute(ctx, []))
+        await UninstallSkill().execute(ctx, [])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Usage:" in m for m in msgs)
 
-    def test_uninstall_success(self):
+    async def test_uninstall_success(self):
         from EvoScientist.commands.implementation.skills import UninstallSkill
 
         ctx, ui = _ctx()
@@ -57,11 +55,11 @@ class TestUninstallSkill:
             "EvoScientist.tools.skills_manager.uninstall_skill",
             return_value={"success": True},
         ):
-            _run(UninstallSkill().execute(ctx, ["demo-skill"]))
+            await UninstallSkill().execute(ctx, ["demo-skill"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Uninstalled: demo-skill" in m for m in msgs)
 
-    def test_uninstall_failure(self):
+    async def test_uninstall_failure(self):
         from EvoScientist.commands.implementation.skills import UninstallSkill
 
         ctx, ui = _ctx()
@@ -69,6 +67,6 @@ class TestUninstallSkill:
             "EvoScientist.tools.skills_manager.uninstall_skill",
             return_value={"success": False, "error": "not found"},
         ):
-            _run(UninstallSkill().execute(ctx, ["missing"]))
+            await UninstallSkill().execute(ctx, ["missing"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Failed: not found" in m for m in msgs)

tests/test_json_sink.py

@@ -19,7 +19,7 @@ async def _agen(items):
         yield item
 
 
-def test_writes_each_event_as_one_jsonl_line(run_async):
+async def test_writes_each_event_as_one_jsonl_line():
     """Each event dict is serialized to exactly one JSON line, in order."""
     events = [
         {"type": "thinking", "content": "hmm", "id": 0},
@@ -34,7 +34,7 @@ def test_writes_each_event_as_one_jsonl_line(run_async):
     ]
     out = io.StringIO()
 
-    run_async(write_events_as_json(_agen(events), out))
+    await write_events_as_json(_agen(events), out)
 
     lines = out.getvalue().splitlines()
     assert len(lines) == len(events)
@@ -43,7 +43,7 @@ def test_writes_each_event_as_one_jsonl_line(run_async):
     assert parsed[2]["args"] == {"path": "a.md"}
 
 
-def test_returns_final_response_from_done_event(run_async):
+async def test_returns_final_response_from_done_event():
     """The sink returns the response text carried by the terminal `done` event."""
     events = [
         {"type": "text", "content": "partial"},
@@ -51,12 +51,12 @@ def test_returns_final_response_from_done_event(run_async):
     ]
     out = io.StringIO()
 
-    result = run_async(write_events_as_json(_agen(events), out))
+    result = await write_events_as_json(_agen(events), out)
 
     assert result == "the answer"
 
 
-def test_non_serializable_arg_does_not_crash_the_stream(run_async):
+async def test_non_serializable_arg_does_not_crash_the_stream():
     """A non-JSON-serializable value degrades to its str form instead of raising."""
 
     class Weird:
@@ -72,7 +72,7 @@ def test_non_serializable_arg_does_not_crash_the_stream(run_async):
     ]
     out = io.StringIO()
 
-    run_async(write_events_as_json(_agen(events), out))
+    await write_events_as_json(_agen(events), out)
 
     lines = out.getvalue().splitlines()
     # Both lines must be valid JSON; the non-serializable value falls back to str.
@@ -80,7 +80,7 @@ def test_non_serializable_arg_does_not_crash_the_stream(run_async):
     assert first["args"]["obj"] == "WEIRD"
 
 
-def test_stream_json_sources_events_from_gateway(run_async):
+async def test_stream_json_sources_events_from_gateway():
     """stream_json pulls events from gateway.stream_events(request) and serializes
     them — it does not reach past the gateway abstraction."""
     seen: dict[str, object] = {}
@@ -98,7 +98,7 @@ def test_stream_json_sources_events_from_gateway(run_async):
             return _agen(events)
 
     out = io.StringIO()
-    result = run_async(stream_json(_FakeGateway(), object(), out=out))
+    result = await stream_json(_FakeGateway(), object(), out=out)
 
     assert result == "hi"
     assert "request" in seen  # the request was forwarded to the gateway
@@ -106,7 +106,7 @@ def test_stream_json_sources_events_from_gateway(run_async):
     assert types == ["text", "done"]
 
 
-def test_stream_json_propagates_gateway_errors(run_async):
+async def test_stream_json_propagates_gateway_errors():
     """An error from the gateway stream propagates out of stream_json so the CLI
     dispatch can turn it into a clean exit."""
 
@@ -124,4 +124,4 @@ def test_stream_json_propagates_gateway_errors(run_async):
 
     out = io.StringIO()
     with pytest.raises(RuntimeError, match="boom"):
-        run_async(stream_json(_FakeGateway(), object(), out=out))
+        await stream_json(_FakeGateway(), object(), out=out)

tests/test_langgraph_manager.py

@@ -8,6 +8,7 @@ to be available.
 from __future__ import annotations
 
 import dataclasses
+import sys
 from types import SimpleNamespace
 from unittest.mock import MagicMock, patch
 
@@ -32,6 +33,77 @@ def reset_module_state():
     manager._LOG_OFFSET_AT_START = 0
 
 
+# =============================================================================
+# langgraph CLI resolution
+# =============================================================================
+
+
+class TestLanggraphCliResolution:
+    def _make_executable(self, path):
+        path.write_text("#!/bin/sh\n", encoding="utf-8")
+        path.chmod(0o755)
+
+    def test_prefers_current_python_environment_over_path(self, tmp_path, monkeypatch):
+        local_bin = tmp_path / "local" / "bin"
+        local_bin.mkdir(parents=True)
+        local_langgraph = local_bin / "langgraph"
+        self._make_executable(local_langgraph)
+
+        path_bin = tmp_path / "path" / "bin"
+        path_bin.mkdir(parents=True)
+        path_langgraph = path_bin / "langgraph"
+        self._make_executable(path_langgraph)
+
+        monkeypatch.setattr(sys, "executable", str(local_bin / "python"))
+        monkeypatch.setattr(
+            manager.shutil,
+            "which",
+            lambda command: str(path_langgraph) if command == "langgraph" else None,
+        )
+
+        assert manager._langgraph_exe() == str(local_langgraph)
+
+    def test_falls_back_to_path_when_environment_binary_missing(
+        self, tmp_path, monkeypatch
+    ):
+        path_bin = tmp_path / "path" / "bin"
+        path_bin.mkdir(parents=True)
+        path_langgraph = path_bin / "langgraph"
+        self._make_executable(path_langgraph)
+
+        monkeypatch.setattr(
+            sys, "executable", str(tmp_path / "local" / "bin" / "python")
+        )
+        monkeypatch.setattr(
+            manager.shutil,
+            "which",
+            lambda command: str(path_langgraph) if command == "langgraph" else None,
+        )
+
+        assert manager._langgraph_exe() == str(path_langgraph)
+
+    def test_checks_windows_suffix_next_to_current_python(self, tmp_path, monkeypatch):
+        scripts_dir = tmp_path / "Scripts"
+        scripts_dir.mkdir()
+        local_langgraph = scripts_dir / "langgraph.exe"
+        self._make_executable(local_langgraph)
+
+        path_bin = tmp_path / "path" / "bin"
+        path_bin.mkdir(parents=True)
+        path_langgraph = path_bin / "langgraph.exe"
+        self._make_executable(path_langgraph)
+
+        monkeypatch.setattr(sys, "executable", str(scripts_dir / "python.exe"))
+        monkeypatch.setattr(manager.os, "name", "nt", raising=False)
+        monkeypatch.setattr(
+            manager.shutil,
+            "which",
+            lambda command: str(path_langgraph) if command == "langgraph" else None,
+        )
+
+        assert manager._langgraph_exe() == str(local_langgraph)
+
+
 # =============================================================================
 # is_langgraph_dev_running
 # =============================================================================

tests/test_langgraph_schema_generator_silenced.py

@@ -0,0 +1,152 @@
+"""Regression tests for the langgraph_api SchemaGenerator silencing patch.
+
+Reproducer: mounting our ``/api/models`` custom Starlette app makes
+langgraph_api call ``update_openapi_spec`` at startup, which iterates
+EVERY route (ours + upstream's). Endpoints whose docstrings aren't
+valid YAML hit a warning + traceback in the deploy log — purely noise,
+since the existing fallback path already produces a usable schema
+entry. The patch keeps the fallback shape but silences the log spam.
+"""
+
+from __future__ import annotations
+
+import os
+
+# ``langgraph_api.config`` reads several required env vars at import
+# time via starlette's ``Config(...)`` helper. We don't actually use the
+# DB or Redis here — any non-empty string keeps the loader happy.
+os.environ.setdefault("DATABASE_URI", "sqlite:///:memory:")
+os.environ.setdefault("REDIS_URI", "redis://localhost:6379")
+
+# Importing patches.py applies the eager module-level monkey-patch.
+import langgraph_api.utils as _lgapi_utils
+
+import EvoScientist.llm.patches as _patches
+
+# Re-invoke the patch after env vars are set. Required because earlier test
+# modules (e.g. test_llm.py) import patches.py *without* DATABASE_URI/
+# REDIS_URI, which makes ``langgraph_api.utils`` fail to import inside the
+# patch's bare ``except``; the loader swallows it and the flag stays False
+# forever (Python won't re-run module-level code on subsequent imports).
+# The patch function is idempotent (early-return on the flag), so calling
+# it here is a no-op when the patch already landed and a successful retry
+# when the prior import failed.
+_patches._patch_langgraph_schema_generator_silence_warnings()
+
+
+class _FakeEndpoint:
+    """Minimal Starlette-like endpoint info for the schema generator."""
+
+    def __init__(self, path: str, method: str, func):
+        self.path = path
+        self.http_method = method
+        self.func = func
+
+
+class _DocstringFixture:
+    """The kinds of docstrings the patched generator must handle."""
+
+    @staticmethod
+    def prose_with_colon():
+        """Endpoint summary.
+
+        Query params:
+            id: The thing you want.
+        """
+
+    @staticmethod
+    def valid_yaml():
+        """
+        summary: A valid YAML docstring.
+        description: Stays structured.
+        """
+
+    @staticmethod
+    def no_docstring():
+        pass
+
+
+def _generator():
+    return _lgapi_utils.SchemaGenerator(
+        {"openapi": "3.1.0", "info": {"title": "test", "version": "0"}}
+    )
+
+
+def test_prose_docstring_no_longer_logs_warning():
+    """The patched ``parse_docstring`` must silence upstream's structlog
+    WARNING when ``yaml.safe_load`` fails on a prose docstring.
+
+    Inverts the patch first to prove the fixture actually trips
+    ``yaml.safe_load`` — without this baseline assertion the test would
+    pass vacuously if the fixture stopped triggering the failure path
+    (e.g. if upstream changed how docstrings are pre-processed).
+    """
+    from structlog.testing import capture_logs
+
+    gen = _generator()
+    endpoint = _FakeEndpoint("/x", "get", _DocstringFixture.prose_with_colon)
+    gen.get_endpoints = lambda _routes: [endpoint]
+
+    patched_parse = _lgapi_utils.SchemaGenerator.parse_docstring
+    # Phase 1: baseline. Drop the subclass override so MRO falls through
+    # to Starlette's BaseSchemaGenerator.parse_docstring, which is what
+    # production hits before our patch installs.
+    del _lgapi_utils.SchemaGenerator.parse_docstring
+    try:
+        with capture_logs() as baseline_records:
+            gen.get_schema([])
+    finally:
+        _lgapi_utils.SchemaGenerator.parse_docstring = patched_parse
+
+    baseline_warnings = [r for r in baseline_records if r.get("log_level") == "warning"]
+    assert any(
+        "Unable to parse docstring" in r.get("event", "") for r in baseline_warnings
+    ), "fixture no longer trips parse_docstring — test would pass vacuously"
+
+    # Phase 2: with the patch reinstated, the same call must emit no
+    # warning records.
+    with capture_logs() as patched_records:
+        schema = gen.get_schema([])
+
+    assert [r for r in patched_records if r.get("log_level") == "warning"] == []
+
+    # Schema still has the fallback shape — fixture's prose becomes the
+    # description verbatim (with leading/trailing whitespace from the
+    # docstring preserved by upstream's fallback path).
+    entry = schema["paths"]["/x"]["get"]
+    assert "description" in entry
+    assert "Query params" in entry["description"]
+
+
+def test_valid_yaml_docstring_keeps_structured_parse():
+    """Endpoints with parseable YAML keep their structured metadata —
+    we only changed the failure branch, not the success path.
+    """
+    gen = _generator()
+    endpoint = _FakeEndpoint("/y", "get", _DocstringFixture.valid_yaml)
+    gen.get_endpoints = lambda _routes: [endpoint]
+    schema = gen.get_schema([])
+    entry = schema["paths"]["/y"]["get"]
+    assert entry.get("summary") == "A valid YAML docstring."
+    assert entry.get("description") == "Stays structured."
+
+
+def test_no_docstring_still_handled():
+    """Endpoints with ``__doc__ = None`` must not raise — fallback uses
+    empty string for ``description``.
+    """
+    gen = _generator()
+    endpoint = _FakeEndpoint("/z", "get", _DocstringFixture.no_docstring)
+    gen.get_endpoints = lambda _routes: [endpoint]
+    schema = gen.get_schema([])
+    entry = schema["paths"]["/z"]["get"]
+    # Either description="" (fallback path) or structured (if YAML parse
+    # of None happens to succeed somehow — implementation detail).
+    # The contract is just "no exception, entry exists".
+    assert isinstance(entry, dict)
+
+
+def test_patch_flag_set():
+    from EvoScientist.llm.patches import _langgraph_schema_silenced_patched
+
+    assert _langgraph_schema_silenced_patched is True

tests/test_llm.py

@@ -1,5 +1,6 @@
 """Tests for EvoScientist LLM module."""
 
+from types import SimpleNamespace
 from unittest.mock import patch
 
 import pytest
@@ -160,6 +161,68 @@ class TestGetModelInfo:
 
 
 class TestGetChatModel:
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_uses_host_model_resolver(self, mock_init):
+        """An embedding host can provide model routing without a core dependency."""
+        from EvoScientist.runtime_integrations import (
+            configure_runtime_integrations,
+            reset_runtime_integrations,
+        )
+
+        mock_init.return_value = "mock_model"
+        resolved = SimpleNamespace(
+            provider_name="relay-a",
+            model_id="model-a",
+            protocol="openai",
+            api_key="sk-host",
+            base_url="https://relay.example/v1/",
+            params={"max_tokens": 8192, "_default_headers": {"X-Relay": "a"}},
+            supports_reasoning=False,
+        )
+        configure_runtime_integrations(model_resolver=lambda model, provider: resolved)
+        try:
+            assert get_chat_model("alias-a") == "mock_model"
+        finally:
+            reset_runtime_integrations()
+
+        call_kwargs = mock_init.call_args.kwargs
+        assert call_kwargs["model"] == "model-a"
+        assert call_kwargs["model_provider"] == "openai"
+        assert call_kwargs["api_key"] == "sk-host"
+        assert call_kwargs["base_url"] == "https://relay.example/v1"
+        assert call_kwargs["max_tokens"] == 8192
+        assert call_kwargs["default_headers"] == {"X-Relay": "a"}
+        assert "reasoning" not in call_kwargs
+
+    @patch("EvoScientist.llm.models._patch_openai_compat_content")
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_host_openai_provider_with_custom_base_uses_compat_patch(
+        self, mock_init, mock_compat
+    ):
+        from EvoScientist.runtime_integrations import (
+            configure_runtime_integrations,
+            reset_runtime_integrations,
+        )
+
+        model_instance = object()
+        mock_init.return_value = model_instance
+        resolved = SimpleNamespace(
+            provider_name="openai",
+            model_id="gpt-5.5",
+            protocol="openai",
+            api_key="sk-host",
+            base_url="https://relay.example/v1",
+            params={},
+            supports_reasoning=True,
+        )
+        configure_runtime_integrations(model_resolver=lambda model, provider: resolved)
+        try:
+            get_chat_model("gpt-5.5", provider="openai")
+        finally:
+            reset_runtime_integrations()
+
+        mock_compat.assert_called_once_with(model_instance, hoist_tool_media=True)
+
     @patch("EvoScientist.llm.models.init_chat_model")
     def test_uses_default_model_when_none(self, mock_init):
         """Test that get_chat_model uses default model when model=None."""
@@ -229,6 +292,39 @@ class TestGetChatModel:
         assert call_kwargs["temperature"] == 0.7
         assert call_kwargs["max_tokens"] == 1000
 
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_drops_unsupported_legacy_model_kwargs(self, mock_init):
+        mock_init.return_value = "mock_model"
+
+        get_chat_model(
+            "gpt-5-nano",
+            provider="openai",
+            sanitize_openai_sdk_headers=True,
+            model_kwargs={"sanitize_openai_sdk_headers": False, "custom": "value"},
+        )
+
+        call_kwargs = mock_init.call_args.kwargs
+        assert "sanitize_openai_sdk_headers" not in call_kwargs
+        assert call_kwargs["model_kwargs"] == {"custom": "value"}
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_explicit_credentials_override_environment(self, mock_init, monkeypatch):
+        """Host-provided credentials take precedence over process defaults."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENAI_API_KEY", "sk-environment")
+        monkeypatch.setenv("OPENAI_BASE_URL", "https://environment.example/v1")
+
+        get_chat_model(
+            "gpt-5-nano",
+            provider="openai",
+            api_key="sk-explicit",
+            base_url="https://explicit.example/v1",
+        )
+
+        call_kwargs = mock_init.call_args.kwargs
+        assert call_kwargs["api_key"] == "sk-explicit"
+        assert call_kwargs["base_url"] == "https://explicit.example/v1"
+
     @patch("EvoScientist.llm.models.init_chat_model")
     def test_infers_openai_from_gpt_prefix(self, mock_init):
         """Test that OpenAI is inferred from gpt- prefix."""
@@ -439,6 +535,200 @@ class TestThirdPartyRouting:
         call_kwargs = mock_init.call_args[1]
         assert call_kwargs["reasoning"] == {"effort": "medium", "summary": "auto"}
 
+    # --- OpenRouter app attribution (issue #339) ---
+
+    _APP_ATTR_ENV = (
+        "EVOSCIENTIST_OPENROUTER_HTTP_REFERER",
+        "EVOSCIENTIST_OPENROUTER_APP_TITLE",
+        "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES",
+    )
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openrouter_app_attribution_defaults(self, mock_init, monkeypatch):
+        """OpenRouter init should carry EvoScientist's default app attribution."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+        # Isolate from any leaked env overrides so we assert the built-in defaults.
+        for _env in self._APP_ATTR_ENV:
+            monkeypatch.delenv(_env, raising=False)
+
+        get_chat_model("x-ai/grok-4.3", provider="openrouter")
+
+        call_kwargs = mock_init.call_args[1]
+        assert call_kwargs["app_url"] == "https://github.com/EvoScientist/EvoScientist"
+        assert call_kwargs["app_title"] == "EvoScientist"
+        # Must be a list[str] (not the comma string) — langchain-openrouter joins it.
+        assert call_kwargs["app_categories"] == ["creative-writing", "personal-agent"]
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openrouter_app_attribution_from_env(self, mock_init, monkeypatch):
+        """Env vars should override the default app attribution values."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+        monkeypatch.setenv("EVOSCIENTIST_OPENROUTER_HTTP_REFERER", "https://acme.test")
+        monkeypatch.setenv("EVOSCIENTIST_OPENROUTER_APP_TITLE", "Acme")
+        # Include a space to prove each category is stripped.
+        monkeypatch.setenv(
+            "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES", "cli-agent, programming-app"
+        )
+
+        get_chat_model("x-ai/grok-4.3", provider="openrouter")
+
+        call_kwargs = mock_init.call_args[1]
+        assert call_kwargs["app_url"] == "https://acme.test"
+        assert call_kwargs["app_title"] == "Acme"
+        assert call_kwargs["app_categories"] == ["cli-agent", "programming-app"]
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openrouter_app_attribution_user_override_not_clobbered(
+        self, mock_init, monkeypatch
+    ):
+        """Caller-supplied attribution kwargs must beat both env and defaults."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+        # Env is also set, to prove an explicit kwarg outranks the env override
+        # (not just the built-in default).
+        monkeypatch.setenv(
+            "EVOSCIENTIST_OPENROUTER_HTTP_REFERER", "https://env.example"
+        )
+        monkeypatch.setenv("EVOSCIENTIST_OPENROUTER_APP_TITLE", "EnvTitle")
+        monkeypatch.setenv("EVOSCIENTIST_OPENROUTER_APP_CATEGORIES", "env-cat")
+
+        get_chat_model(
+            "x-ai/grok-4.3",
+            provider="openrouter",
+            app_url="https://mine.example",
+            app_title="MyApp",
+            app_categories=["only-this"],
+        )
+
+        call_kwargs = mock_init.call_args[1]
+        assert call_kwargs["app_url"] == "https://mine.example"
+        assert call_kwargs["app_title"] == "MyApp"
+        # An explicit list is preserved verbatim, not re-split.
+        assert call_kwargs["app_categories"] == ["only-this"]
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_non_openrouter_providers_get_no_app_attribution(
+        self, mock_init, monkeypatch
+    ):
+        """Only the openrouter provider should receive app-attribution kwargs."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("ANTHROPIC_API_KEY", "sk-real")
+        monkeypatch.setenv("OLLAMA_BASE_URL", "http://localhost:11434")
+
+        for model, provider in (
+            ("claude-sonnet-4-6", "anthropic"),
+            ("llama3.1:8b", "ollama"),
+        ):
+            get_chat_model(model, provider=provider)
+            call_kwargs = mock_init.call_args[1]
+            assert "app_url" not in call_kwargs
+            assert "app_title" not in call_kwargs
+            assert "app_categories" not in call_kwargs
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openrouter_app_attribution_coexists_with_reasoning_and_cache(
+        self, mock_init, monkeypatch
+    ):
+        """Attribution must not disturb reasoning or Anthropic prompt caching."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+        monkeypatch.delenv("EVOSCIENTIST_REASONING_EFFORT", raising=False)
+        monkeypatch.delenv(
+            "EVOSCIENTIST_OPENROUTER_ANTHROPIC_PROMPT_CACHE", raising=False
+        )
+        for _env in self._APP_ATTR_ENV:
+            monkeypatch.delenv(_env, raising=False)
+
+        get_chat_model("claude-sonnet-4.6", provider="openrouter")
+
+        call_kwargs = mock_init.call_args[1]
+        # Existing behavior intact.
+        assert call_kwargs["reasoning"] == {"effort": "high", "summary": "auto"}
+        assert call_kwargs["model_kwargs"]["cache_control"] == {"type": "ephemeral"}
+        # Attribution added alongside.
+        assert call_kwargs["app_url"] == "https://github.com/EvoScientist/EvoScientist"
+        assert call_kwargs["app_title"] == "EvoScientist"
+        assert call_kwargs["app_categories"] == ["creative-writing", "personal-agent"]
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openrouter_app_categories_env_strips_blank_items(
+        self, mock_init, monkeypatch
+    ):
+        """A messy comma value (stray commas / spaces) yields a clean list."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+        monkeypatch.setenv("EVOSCIENTIST_OPENROUTER_APP_CATEGORIES", "a,,  b  ")
+
+        get_chat_model("x-ai/grok-4.3", provider="openrouter")
+
+        assert mock_init.call_args[1]["app_categories"] == ["a", "b"]
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openrouter_app_categories_capped_to_per_request_limit(
+        self, mock_init, monkeypatch
+    ):
+        """Over-configuring categories caps to the first N and warns the user."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+        monkeypatch.setenv(
+            "EVOSCIENTIST_OPENROUTER_APP_CATEGORIES",
+            "cli-agent,programming-app,personal-agent,writing-assistant",
+        )
+
+        with pytest.warns(UserWarning, match="at most 2 app categories"):
+            get_chat_model("x-ai/grok-4.3", provider="openrouter")
+
+        # OpenRouter honors at most 2 per request, so only the first 2 are sent.
+        assert mock_init.call_args[1]["app_categories"] == [
+            "cli-agent",
+            "programming-app",
+        ]
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openrouter_app_categories_all_separators_omit_kwarg(
+        self, mock_init, monkeypatch
+    ):
+        """A categories value with no real items omits the kwarg entirely."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+        monkeypatch.setenv("EVOSCIENTIST_OPENROUTER_APP_CATEGORIES", " , , ")
+
+        get_chat_model("x-ai/grok-4.3", provider="openrouter")
+
+        # No app_categories kwarg at all — not an empty list (which the library
+        # would reject / send as an empty header).
+        assert "app_categories" not in mock_init.call_args[1]
+
+    def test_openrouter_app_attribution_lands_on_real_model(self, monkeypatch):
+        """Build a REAL ChatOpenRouter (no mock) and assert the attribution
+        values land on the instance rather than being silently dumped into
+        model_kwargs.
+
+        The mocked tests above assert on the kwargs handed to init_chat_model,
+        so they cannot catch a param-name typo or a langchain-openrouter version
+        that accepts these only as passthrough model params (which the library
+        does with a warning, not an error). This test is the guard for both.
+        """
+        from langchain_openrouter import ChatOpenRouter
+
+        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+        for _env in self._APP_ATTR_ENV:
+            monkeypatch.delenv(_env, raising=False)
+
+        model = get_chat_model("x-ai/grok-4.3", provider="openrouter")
+
+        assert isinstance(model, ChatOpenRouter)
+        assert model.app_url == "https://github.com/EvoScientist/EvoScientist"
+        assert model.app_title == "EvoScientist"
+        assert model.app_categories == ["creative-writing", "personal-agent"]
+        # Not silently swallowed into model_kwargs (the passthrough failure mode).
+        model_kwargs = model.model_kwargs or {}
+        assert "app_url" not in model_kwargs
+        assert "app_title" not in model_kwargs
+        assert "app_categories" not in model_kwargs
+
     @patch("EvoScientist.llm.models.init_chat_model")
     def test_openrouter_anthropic_prompt_cache_enabled_by_default(
         self, mock_init, monkeypatch
@@ -957,6 +1247,153 @@ class TestPatchOpenAICompatContent:
         model._astream = AsyncMock()
         return model
 
+    def test_missing_tool_call_ids_are_repaired_without_mutating_history(self):
+        from langchain_core.messages import AIMessage, ToolMessage
+
+        from EvoScientist.llm.patches import _ensure_openai_tool_call_ids
+
+        ai = AIMessage(
+            content=[{"type": "tool_call", "id": "", "name": "execute", "args": {}}],
+            tool_calls=[{"id": "", "name": "execute", "args": {}}],
+        )
+        tool = ToolMessage(content="ok", tool_call_id="")
+
+        normalized = _ensure_openai_tool_call_ids([ai, tool])
+
+        call_id = normalized[0].tool_calls[0]["id"]
+        assert call_id.startswith("call_")
+        assert normalized[0].content[0]["id"] == call_id
+        assert normalized[1].tool_call_id == call_id
+        assert ai.tool_calls[0]["id"] == ""
+        assert tool.tool_call_id == ""
+
+    def test_missing_parallel_tool_call_ids_are_stable_and_ordered(self):
+        from langchain_core.messages import AIMessage, ToolMessage
+
+        from EvoScientist.llm.patches import _ensure_openai_tool_call_ids
+
+        messages = [
+            AIMessage(
+                id="assistant-1",
+                content="",
+                tool_calls=[
+                    {"id": "", "name": "read_file", "args": {}},
+                    {"id": "", "name": "execute", "args": {}},
+                ],
+            ),
+            ToolMessage(content="file", tool_call_id=""),
+            ToolMessage(content="command", tool_call_id=""),
+        ]
+
+        first = _ensure_openai_tool_call_ids(messages)
+        second = _ensure_openai_tool_call_ids(messages)
+        call_ids = [call["id"] for call in first[0].tool_calls]
+
+        assert call_ids == [call["id"] for call in second[0].tool_calls]
+        assert len(set(call_ids)) == 2
+        assert [message.tool_call_id for message in first[1:]] == call_ids
+
+    def test_content_tool_block_is_normalized_to_parsed_call(self):
+        from langchain_core.messages import AIMessage, ToolMessage
+
+        from EvoScientist.llm.patches import _ensure_openai_tool_call_ids
+
+        normalized = _ensure_openai_tool_call_ids(
+            [
+                AIMessage(
+                    content=[
+                        {
+                            "type": "tool_call",
+                            "id": "wrong-id",
+                            "name": "wrong-name",
+                            "args": {},
+                        }
+                    ],
+                    tool_calls=[{"id": "call-1", "name": "execute", "args": {}}],
+                ),
+                ToolMessage(content="ok", tool_call_id="call-1"),
+            ]
+        )
+
+        assert normalized[0].content[0]["id"] == "call-1"
+        assert normalized[0].content[0]["name"] == "execute"
+
+    def test_invalid_tool_call_is_not_replayed_to_responses_api(self):
+        from langchain_core.messages import AIMessage, HumanMessage
+        from langchain_openai.chat_models.base import _construct_responses_api_input
+
+        from EvoScientist.llm.patches import _sanitize_messages
+
+        invalid = AIMessage(
+            content=[
+                {"type": "reasoning", "reasoning": "partial"},
+                {
+                    "type": "tool_call",
+                    "id": None,
+                    "name": "execute",
+                    "args": '{"command":',
+                },
+            ],
+            invalid_tool_calls=[
+                {
+                    "type": "invalid_tool_call",
+                    "id": None,
+                    "name": "execute",
+                    "args": '{"command":',
+                    "error": "Failed to parse tool call arguments as JSON",
+                }
+            ],
+        )
+
+        normalized = _sanitize_messages([invalid, HumanMessage(content="retry")])
+        payload = _construct_responses_api_input(normalized)
+
+        assert all(item.get("type") != "function_call" for item in payload)
+        assert [message.type for message in normalized] == ["human"]
+
+    def test_invalid_tool_call_preserves_replayable_assistant_text(self):
+        from langchain_core.messages import AIMessage
+
+        from EvoScientist.llm.patches import _sanitize_messages
+
+        invalid = AIMessage(
+            content="I could not finish the tool request.",
+            invalid_tool_calls=[
+                {
+                    "type": "invalid_tool_call",
+                    "id": None,
+                    "name": "execute",
+                    "args": "{",
+                    "error": "bad json",
+                }
+            ],
+        )
+
+        normalized = _sanitize_messages([invalid])
+
+        assert len(normalized) == 1
+        assert normalized[0].content == "I could not finish the tool request."
+        assert normalized[0].invalid_tool_calls == []
+
+    def test_orphan_tool_results_and_unanswered_calls_are_removed(self):
+        from langchain_core.messages import AIMessage, HumanMessage, ToolMessage
+
+        from EvoScientist.llm.patches import _sanitize_messages
+
+        messages = [
+            ToolMessage(content="orphan", tool_call_id="missing"),
+            AIMessage(
+                content="waiting",
+                tool_calls=[{"id": "call_unanswered", "name": "execute", "args": {}}],
+            ),
+            HumanMessage(content="continue"),
+        ]
+
+        normalized = _sanitize_messages(messages)
+
+        assert [message.type for message in normalized] == ["ai", "human"]
+        assert normalized[0].tool_calls == []
+
     def test_generate_flattened(self):
         from langchain_core.messages import HumanMessage
 
@@ -972,7 +1409,6 @@ class TestPatchOpenAICompatContent:
         called_msgs = orig.call_args[0][0]
         assert called_msgs[0].content == "hello"
 
-    @pytest.mark.anyio
     async def test_agenerate_flattened(self):
         from langchain_core.messages import HumanMessage
 
@@ -1003,7 +1439,6 @@ class TestPatchOpenAICompatContent:
         called_msgs = orig.call_args[0][0]
         assert called_msgs[0].content == "hello"
 
-    @pytest.mark.anyio
     async def test_astream_flattened(self):
         from langchain_core.messages import HumanMessage
 
@@ -1044,7 +1479,6 @@ class TestPatchOpenAICompatContent:
         called_msgs = orig.call_args[0][0]
         assert called_msgs[0].content == [{"type": "text", "text": "see"}, img]
 
-    @pytest.mark.anyio
     async def test_agenerate_preserves_media(self):
         from langchain_core.messages import HumanMessage
 
@@ -1077,7 +1511,6 @@ class TestPatchOpenAICompatContent:
         called_msgs = orig.call_args[0][0]
         assert called_msgs[0].content == [{"type": "text", "text": "see"}, img]
 
-    @pytest.mark.anyio
     async def test_astream_preserves_media(self):
         from langchain_core.messages import HumanMessage
 
@@ -1550,7 +1983,6 @@ class TestNoVisionFallback:
         assert out == ["x", "y"]
         assert len(calls) == 2
 
-    @pytest.mark.anyio
     async def test_astream_falls_back(self):
         from unittest.mock import MagicMock
 
@@ -2243,6 +2675,38 @@ class TestPatchOpenrouterStripResponsesReasoning:
 
 
 class TestAutoConfig:
+    @pytest.fixture(autouse=True)
+    def _clear_reasoning_effort_env(self, monkeypatch):
+        """Keep auto-config tests independent of the developer environment."""
+        monkeypatch.delenv("EVOSCIENTIST_REASONING_EFFORT", raising=False)
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_internal_sentinels_disable_auto_reasoning(self, mock_init, monkeypatch):
+        """Internal callers can disable reasoning without leaking sentinels."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.delenv("ANTHROPIC_BASE_URL", raising=False)
+        monkeypatch.delenv("OPENAI_BASE_URL", raising=False)
+
+        for model, provider in (
+            ("claude-sonnet-4-6", "anthropic"),
+            ("gpt-5-nano", "openai"),
+            ("gemini-2.5-flash", "google-genai"),
+            ("llama3.1:8b", "ollama"),
+        ):
+            mock_init.reset_mock()
+            get_chat_model(
+                model,
+                provider=provider,
+                _disable_reasoning=True,
+                _disable_thinking=True,
+            )
+            call_kwargs = mock_init.call_args.kwargs
+            assert "_disable_reasoning" not in call_kwargs
+            assert "_disable_thinking" not in call_kwargs
+            assert "reasoning" not in call_kwargs
+            assert "thinking" not in call_kwargs
+            assert "include_thoughts" not in call_kwargs
+
     @patch("EvoScientist.llm.models.init_chat_model")
     def test_anthropic_4_5_thinking(self, mock_init, monkeypatch):
         """Anthropic 4-5 models get enabled thinking with budget."""
@@ -2332,6 +2796,7 @@ class TestAutoConfig:
         """gpt-5.4+ and codex models get xhigh reasoning."""
         mock_init.return_value = "mock_model"
         monkeypatch.delenv("OPENAI_BASE_URL", raising=False)
+        monkeypatch.delenv("EVOSCIENTIST_REASONING_EFFORT", raising=False)
 
         get_chat_model("gpt-5.4", provider="openai")
         assert mock_init.call_args[1]["reasoning"] == {
@@ -2351,6 +2816,26 @@ class TestAutoConfig:
             "summary": "auto",
         }
 
+        get_chat_model("gpt-5.6-sol", provider="openai")
+        assert mock_init.call_args[1]["reasoning"] == {
+            "effort": "xhigh",
+            "summary": "auto",
+        }
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openai_reasoning_effort_from_env(self, mock_init, monkeypatch):
+        """Native OpenAI reasoning effort should be configurable via env var."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.delenv("OPENAI_BASE_URL", raising=False)
+        monkeypatch.setenv("EVOSCIENTIST_REASONING_EFFORT", "high")
+
+        get_chat_model("gpt-5.5", provider="openai")
+
+        assert mock_init.call_args[1]["reasoning"] == {
+            "effort": "high",
+            "summary": "auto",
+        }
+
     @patch("EvoScientist.llm.models.init_chat_model")
     def test_openai_reasoning_high_fallback(self, mock_init, monkeypatch):
         """Other OpenAI models get high reasoning effort."""
@@ -2382,8 +2867,8 @@ class TestAutoConfig:
         assert call_kwargs["model_provider"] == "openai"
         assert call_kwargs["base_url"] == "http://127.0.0.1:8000/codex/v1"
         assert call_kwargs["api_key"] == "ccproxy-oauth"
-        # Proxy mode: reasoning skipped (ccproxy untested)
-        assert "reasoning" not in call_kwargs
+        # ccproxy uses the Responses API, so reasoning configuration is valid.
+        assert call_kwargs["reasoning"] == {"effort": "high", "summary": "auto"}
         # Proxy mode: Responses API (bypasses format chain), streaming ON
         assert call_kwargs["use_responses_api"] is True
         assert "streaming" not in call_kwargs
@@ -2416,6 +2901,120 @@ class TestAutoConfig:
         call_kwargs = mock_init.call_args[1]
         assert call_kwargs["reasoning"] == {"effort": "high", "summary": "auto"}
         assert "use_responses_api" not in call_kwargs
+        assert "default_headers" not in call_kwargs
+
+    @patch(
+        "EvoScientist.llm.models._installed_codex_client_version",
+        return_value="0.144.1",
+    )
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openai_ccproxy_codex_client_headers(
+        self, mock_init, mock_installed_version, monkeypatch
+    ):
+        """ccproxy Codex mode sends Codex-CLI-shaped client headers."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENAI_BASE_URL", "http://127.0.0.1:8000/codex/v1")
+        monkeypatch.setenv("OPENAI_API_KEY", "ccproxy-oauth")
+        monkeypatch.delenv("EVOSCIENTIST_CODEX_CLIENT_VERSION", raising=False)
+
+        get_chat_model("gpt-5.5", provider="openai")
+
+        headers = mock_init.call_args[1]["default_headers"]
+        assert headers["originator"] == "codex_cli_rs"
+        assert headers["version"] == "0.144.1"
+        assert headers["User-Agent"].startswith("codex_cli_rs/0.144.1")
+        mock_installed_version.assert_called_once_with()
+        assert mock_init.call_args[1]["reasoning"]["effort"] == "xhigh"
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openai_ccproxy_codex_client_version_env(self, mock_init, monkeypatch):
+        """EVOSCIENTIST_CODEX_CLIENT_VERSION overrides the pinned version."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENAI_BASE_URL", "http://127.0.0.1:8000/codex/v1")
+        monkeypatch.setenv("OPENAI_API_KEY", "ccproxy-oauth")
+        monkeypatch.setenv("EVOSCIENTIST_CODEX_CLIENT_VERSION", "9.9.9")
+
+        get_chat_model("gpt-5.5", provider="openai")
+
+        headers = mock_init.call_args[1]["default_headers"]
+        assert headers["version"] == "9.9.9"
+        assert headers["User-Agent"].startswith("codex_cli_rs/9.9.9")
+
+    @patch("EvoScientist.llm.models.subprocess.run")
+    def test_installed_codex_client_version(self, mock_run):
+        """The advertised version follows the installed Codex CLI."""
+        from EvoScientist.llm.models import _installed_codex_client_version
+
+        mock_run.return_value.returncode = 0
+        mock_run.return_value.stdout = "codex-cli 0.144.1\n"
+        mock_run.return_value.stderr = ""
+        _installed_codex_client_version.cache_clear()
+        try:
+            assert _installed_codex_client_version() == "0.144.1"
+            assert _installed_codex_client_version() == "0.144.1"
+        finally:
+            _installed_codex_client_version.cache_clear()
+        mock_run.assert_called_once_with(
+            ["codex", "--version"],
+            capture_output=True,
+            text=True,
+            timeout=2,
+            check=False,
+        )
+
+    @patch(
+        "EvoScientist.llm.models._installed_codex_client_version",
+        return_value="0.140.0",
+    )
+    def test_older_installed_codex_uses_fallback(
+        self, mock_installed_version, monkeypatch
+    ):
+        """An outdated installed CLI must not undercut the safe fallback."""
+        from EvoScientist.llm.models import (
+            _CODEX_CLIENT_VERSION_FALLBACK,
+            _resolve_codex_client_version,
+        )
+
+        monkeypatch.delenv("EVOSCIENTIST_CODEX_CLIENT_VERSION", raising=False)
+
+        assert _resolve_codex_client_version() == _CODEX_CLIENT_VERSION_FALLBACK
+        mock_installed_version.assert_called_once_with()
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openai_ccproxy_codex_headers_respect_caller(self, mock_init, monkeypatch):
+        """Caller-supplied default_headers keys are not overridden."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENAI_BASE_URL", "http://127.0.0.1:8000/codex/v1")
+        monkeypatch.setenv("OPENAI_API_KEY", "ccproxy-oauth")
+
+        get_chat_model(
+            "gpt-5.5",
+            provider="openai",
+            default_headers={"originator": "codex_vscode", "version": "9.9.9"},
+        )
+
+        headers = mock_init.call_args[1]["default_headers"]
+        assert headers["originator"] == "codex_vscode"
+        assert headers["version"] == "9.9.9"
+        assert headers["User-Agent"].startswith("codex_cli_rs/9.9.9")
+
+    @patch("EvoScientist.llm.models.init_chat_model")
+    def test_openai_ccproxy_codex_none_headers(self, mock_init, monkeypatch):
+        """An explicit default_headers=None is normalized before gap-filling."""
+        mock_init.return_value = "mock_model"
+        monkeypatch.setenv("OPENAI_BASE_URL", "http://127.0.0.1:8000/codex/v1")
+        monkeypatch.setenv("OPENAI_API_KEY", "ccproxy-oauth")
+        monkeypatch.setenv("EVOSCIENTIST_CODEX_CLIENT_VERSION", "9.9.9")
+
+        get_chat_model(
+            "gpt-5.5",
+            provider="openai",
+            default_headers=None,
+        )
+
+        headers = mock_init.call_args[1]["default_headers"]
+        assert headers["originator"] == "codex_cli_rs"
+        assert headers["version"] == "9.9.9"
 
     @patch("EvoScientist.llm.models.init_chat_model")
     def test_openai_ccproxy_key_but_wrong_path_not_ccproxy(

tests/test_logging_config.py

@@ -0,0 +1,129 @@
+import logging
+from datetime import UTC, datetime
+from io import StringIO
+
+from EvoScientist.logging_config import (
+    DailyLogFileHandler,
+    configure_console_logging,
+    configure_daily_file_logging,
+    configure_logging,
+    default_log_dir,
+    resolve_log_level,
+)
+
+
+def test_daily_log_file_handler_uses_dated_active_file(tmp_path):
+    handler = DailyLogFileHandler(tmp_path, prefix="evoscientist", retention_days=30)
+    logger = logging.getLogger("tests.daily_log_file_handler")
+    logger.handlers.clear()
+    logger.propagate = False
+    logger.setLevel(logging.INFO)
+    logger.addHandler(handler)
+
+    logger.info("hello")
+    handler.close()
+
+    today = datetime.now().strftime("%Y-%m-%d")
+    assert (tmp_path / f"evoscientist-{today}.log").read_text(encoding="utf-8").strip()
+
+
+def test_daily_log_file_handler_keeps_latest_retention_days(tmp_path):
+    for day in range(1, 33):
+        (tmp_path / f"evoscientist-2026-01-{day:02d}.log").write_text(
+            "x", encoding="utf-8"
+        )
+
+    handler = DailyLogFileHandler(tmp_path, prefix="evoscientist", retention_days=30)
+    handler._delete_expired_logs()
+    handler.close()
+
+    remaining = sorted(path.name for path in tmp_path.glob("evoscientist-*.log"))
+    assert len(remaining) == 30
+    assert remaining[0] == "evoscientist-2026-01-03.log"
+
+
+def test_configure_daily_file_logging_replaces_matching_handler(tmp_path):
+    logger = logging.getLogger("tests.configure_daily_file_logging")
+    logger.handlers.clear()
+    logger.propagate = False
+
+    first = configure_daily_file_logging(logger, log_dir=tmp_path)
+    second = configure_daily_file_logging(logger, log_dir=tmp_path)
+
+    try:
+        handlers = [h for h in logger.handlers if isinstance(h, DailyLogFileHandler)]
+        assert handlers == [second]
+        assert first.stream is None
+    finally:
+        for handler in logger.handlers[:]:
+            logger.removeHandler(handler)
+            handler.close()
+
+
+def test_configure_logging_replaces_only_managed_handlers(tmp_path):
+    logger = logging.getLogger("tests.configure_logging")
+    logger.handlers.clear()
+    logger.propagate = False
+    external = logging.NullHandler()
+    logger.addHandler(external)
+
+    configure_logging(logger, log_dir=tmp_path, level="debug")
+    configure_logging(logger, log_dir=tmp_path, level="info")
+
+    try:
+        daily_handlers = [h for h in logger.handlers if isinstance(h, DailyLogFileHandler)]
+        stream_handlers = [
+            h
+            for h in logger.handlers
+            if isinstance(h, logging.StreamHandler)
+            and not isinstance(h, DailyLogFileHandler)
+        ]
+        assert external in logger.handlers
+        assert len(daily_handlers) == 1
+        assert len(stream_handlers) == 1
+        assert logger.level == logging.INFO
+    finally:
+        for handler in logger.handlers[:]:
+            logger.removeHandler(handler)
+            handler.close()
+
+
+def test_configure_console_logging_emits_to_stream():
+    logger = logging.getLogger("tests.configure_console_logging")
+    logger.handlers.clear()
+    logger.propagate = False
+    stream = StringIO()
+
+    configure_console_logging(logger, level="INFO", stream=stream)
+    try:
+        logger.info("hello")
+        assert "tests.configure_console_logging: hello" in stream.getvalue()
+    finally:
+        for handler in logger.handlers[:]:
+            logger.removeHandler(handler)
+            handler.close()
+
+
+def test_resolve_log_level_accepts_alias_numeric_and_fallback():
+    assert resolve_log_level("warn") == logging.WARNING
+    assert resolve_log_level("10") == logging.DEBUG
+    assert resolve_log_level("", default=logging.ERROR) == logging.ERROR
+    assert resolve_log_level("not-a-level", default=logging.CRITICAL) == logging.CRITICAL
+
+
+def test_daily_log_file_handler_supports_utc(tmp_path):
+    handler = DailyLogFileHandler(tmp_path, utc=True)
+    try:
+        today_utc = datetime.now(UTC).strftime("%Y-%m-%d")
+        assert handler.active_log_path.name == f"evoscientist-{today_utc}.log"
+    finally:
+        handler.close()
+
+
+def test_default_log_dir_uses_current_data_dir(monkeypatch, tmp_path):
+    import EvoScientist.paths as paths
+
+    monkeypatch.delenv("EVOSCIENTIST_LOG_DIR", raising=False)
+    monkeypatch.setattr(paths, "DATA_DIR", tmp_path / "data")
+
+    assert default_log_dir() == tmp_path / "data" / "logs"

tests/test_mcp_client.py

@@ -1399,9 +1399,7 @@ class TestLoadToolsProgressCallback:
 
         monkeypatch.setattr(lc_client, "MultiServerMCPClient", _FakeClient)
 
-    def test_success_emits_start_then_success_with_tool_count(self, monkeypatch):
-        import asyncio
-
+    async def test_success_emits_start_then_success_with_tool_count(self, monkeypatch):
         from EvoScientist.mcp.client import _load_tools
 
         events: list[tuple[str, str, str]] = []
@@ -1415,36 +1413,45 @@ class TestLoadToolsProgressCallback:
         def record(event, name, detail):
             events.append((event, name, detail))
 
-        asyncio.run(_load_tools(config, on_progress=record))
+        await _load_tools(config, on_progress=record)
 
         assert events == [
             ("start", "srv", ""),
             ("success", "srv", "3"),
         ]
 
-    def test_failure_emits_start_then_error_with_detail(self, monkeypatch):
-        import asyncio
-
-        from EvoScientist.mcp.client import _load_tools
+    async def test_failure_emits_start_then_error_with_detail(self, monkeypatch):
+        from EvoScientist.mcp import client as mcp_client
 
         events: list[tuple[str, str, str]] = []
         self._patch_client(monkeypatch, {"srv": RuntimeError("boom")})
+        monkeypatch.setattr(mcp_client, "_MCP_SERVER_ERRORS", {})
 
         config = {"srv": {"transport": "stdio", "command": "demo"}}
 
         def record(event, name, detail):
             events.append((event, name, detail))
 
-        asyncio.run(_load_tools(config, on_progress=record))
+        await mcp_client._load_tools(config, on_progress=record)
 
         assert events == [
             ("start", "srv", ""),
             ("error", "srv", "boom"),
         ]
+        assert mcp_client.get_mcp_server_errors() == {"srv": "boom"}
 
-    def test_mixed_fleet_reports_each_server_independently(self, monkeypatch):
-        import asyncio
+    async def test_success_clears_previous_server_error(self, monkeypatch):
+        from EvoScientist.mcp import client as mcp_client
 
+        self._patch_client(monkeypatch, {"srv": ["tool1"]})
+        monkeypatch.setattr(mcp_client, "_MCP_SERVER_ERRORS", {"srv": "old error"})
+
+        config = {"srv": {"transport": "stdio", "command": "demo"}}
+        await mcp_client._load_tools(config)
+
+        assert mcp_client.get_mcp_server_errors() == {}
+
+    async def test_mixed_fleet_reports_each_server_independently(self, monkeypatch):
         from EvoScientist.mcp.client import _load_tools
 
         events: list[tuple[str, str, str]] = []
@@ -1464,7 +1471,7 @@ class TestLoadToolsProgressCallback:
         def record(event, name, detail):
             events.append((event, name, detail))
 
-        asyncio.run(_load_tools(config, on_progress=record))
+        await _load_tools(config, on_progress=record)
 
         by_server = {}
         for ev, name, detail in events:
@@ -1472,9 +1479,7 @@ class TestLoadToolsProgressCallback:
         assert by_server["ok_srv"] == [("start", ""), ("success", "1")]
         assert by_server["bad_srv"] == [("start", ""), ("error", "refused")]
 
-    def test_callback_errors_do_not_break_the_load(self, monkeypatch):
-        import asyncio
-
+    async def test_callback_errors_do_not_break_the_load(self, monkeypatch):
         from EvoScientist.mcp.client import _load_tools
 
         self._patch_client(monkeypatch, {"srv": ["tool1"]})
@@ -1484,10 +1489,10 @@ class TestLoadToolsProgressCallback:
         def bad_callback(event, name, detail):
             raise RuntimeError("callback bug")
 
-        result = asyncio.run(_load_tools(config, on_progress=bad_callback))
+        result = await _load_tools(config, on_progress=bad_callback)
         assert result == {"srv": ["tool1"]}
 
-    def test_semaphore_caps_concurrent_connections(self, monkeypatch):
+    async def test_semaphore_caps_concurrent_connections(self, monkeypatch):
         """Many configured servers must not all spawn at once."""
         import asyncio
 
@@ -1516,7 +1521,7 @@ class TestLoadToolsProgressCallback:
         config = {
             f"srv{i}": {"transport": "stdio", "command": "demo"} for i in range(10)
         }
-        asyncio.run(mcp_client._load_tools(config))
+        await mcp_client._load_tools(config)
 
         assert inflight["peak"] <= 3
         assert inflight["peak"] > 1  # sanity: we *are* parallelizing

tests/test_mcp_command.py

@@ -2,8 +2,6 @@
 
 from unittest.mock import MagicMock, patch
 
-from tests.conftest import run_async as _run
-
 
 def _ctx():
     from EvoScientist.commands.base import CommandContext
@@ -14,16 +12,16 @@ def _ctx():
 
 
 class TestMCPCommandDispatch:
-    def test_no_args_lists(self):
+    async def test_no_args_lists(self):
         from EvoScientist.commands.implementation.mcp import MCPCommand
 
         ctx, ui = _ctx()
         with patch("EvoScientist.mcp.load_mcp_config", return_value={}):
-            _run(MCPCommand().execute(ctx, []))
+            await MCPCommand().execute(ctx, [])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("No MCP servers configured" in m for m in msgs)
 
-    def test_list_subcommand(self):
+    async def test_list_subcommand(self):
         from EvoScientist.commands.implementation.mcp import MCPCommand
 
         ctx, ui = _ctx()
@@ -31,10 +29,10 @@ class TestMCPCommandDispatch:
             "srv1": {"transport": "stdio", "tools": ["foo"], "expose_to": ["main"]},
         }
         with patch("EvoScientist.mcp.load_mcp_config", return_value=cfg):
-            _run(MCPCommand().execute(ctx, ["list"]))
+            await MCPCommand().execute(ctx, ["list"])
         ui.mount_renderable.assert_called_once()
 
-    def test_add_subcommand_dispatches(self):
+    async def test_add_subcommand_dispatches(self):
         from EvoScientist.commands.implementation.mcp import MCPCommand
 
         ctx, _ui = _ctx()
@@ -48,10 +46,10 @@ class TestMCPCommandDispatch:
                 return_value={"transport": "stdio"},
             ) as add_mock,
         ):
-            _run(MCPCommand().execute(ctx, ["add", "srv1", "python"]))
+            await MCPCommand().execute(ctx, ["add", "srv1", "python"])
         add_mock.assert_called_once()
 
-    def test_edit_subcommand_dispatches(self):
+    async def test_edit_subcommand_dispatches(self):
         from EvoScientist.commands.implementation.mcp import MCPCommand
 
         ctx, _ui = _ctx()
@@ -64,28 +62,28 @@ class TestMCPCommandDispatch:
                 "EvoScientist.mcp.edit_mcp_server",
             ) as edit_mock,
         ):
-            _run(MCPCommand().execute(ctx, ["edit", "srv1", "--tools", "bar"]))
+            await MCPCommand().execute(ctx, ["edit", "srv1", "--tools", "bar"])
         edit_mock.assert_called_once_with("srv1", tools=["bar"])
 
-    def test_remove_subcommand_success(self):
+    async def test_remove_subcommand_success(self):
         from EvoScientist.commands.implementation.mcp import MCPCommand
 
         ctx, ui = _ctx()
         with patch("EvoScientist.mcp.remove_mcp_server", return_value=True):
-            _run(MCPCommand().execute(ctx, ["remove", "srv1"]))
+            await MCPCommand().execute(ctx, ["remove", "srv1"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Removed MCP server: srv1" in m for m in msgs)
 
-    def test_remove_subcommand_not_found(self):
+    async def test_remove_subcommand_not_found(self):
         from EvoScientist.commands.implementation.mcp import MCPCommand
 
         ctx, ui = _ctx()
         with patch("EvoScientist.mcp.remove_mcp_server", return_value=False):
-            _run(MCPCommand().execute(ctx, ["remove", "missing"]))
+            await MCPCommand().execute(ctx, ["remove", "missing"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Server not found" in m for m in msgs)
 
-    def test_install_delegates_to_install_mcp_command(self):
+    async def test_install_delegates_to_install_mcp_command(self):
         """/mcp install should instantiate InstallMCPCommand and execute it."""
         from EvoScientist.commands.implementation.mcp import MCPCommand
 
@@ -101,13 +99,13 @@ class TestMCPCommandDispatch:
 
             instance.execute = fake_execute
             klass.return_value = instance
-            _run(MCPCommand().execute(ctx, ["install", "foo"]))
+            await MCPCommand().execute(ctx, ["install", "foo"])
         klass.assert_called_once()
 
-    def test_unknown_subcommand_prints_help(self):
+    async def test_unknown_subcommand_prints_help(self):
         from EvoScientist.commands.implementation.mcp import MCPCommand
 
         ctx, ui = _ctx()
-        _run(MCPCommand().execute(ctx, ["bogus"]))
+        await MCPCommand().execute(ctx, ["bogus"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("MCP commands:" in m for m in msgs)

tests/test_model_command.py

@@ -5,8 +5,6 @@ from unittest.mock import AsyncMock, MagicMock, patch
 
 import pytest
 
-from tests.conftest import run_async as _run
-
 
 class TestExtractModelAndProvider:
     """Unit tests for the argument parser helper."""
@@ -80,7 +78,7 @@ class TestExtractModelAndProvider:
 class TestModelCommandUnknownModel:
     """Verify error message for unknown models."""
 
-    def test_unknown_model_shows_error(self):
+    async def test_unknown_model_shows_error(self):
         from EvoScientist.commands.implementation.model import ModelCommand
 
         cmd = ModelCommand()
@@ -95,7 +93,7 @@ class TestModelCommandUnknownModel:
             "EvoScientist.EvoScientist._ensure_config",
             return_value=cfg,
         ):
-            _run(cmd.execute(ctx, ["nonexistent-model-xyz"]))
+            await cmd.execute(ctx, ["nonexistent-model-xyz"])
 
         ui.append_system.assert_called_once()
         call_args = ui.append_system.call_args
@@ -106,7 +104,7 @@ class TestModelCommandUnknownModel:
 class TestModelCommandPickerCancelled:
     """Verify no-op when the interactive picker is cancelled."""
 
-    def test_picker_returns_none(self):
+    async def test_picker_returns_none(self):
         from EvoScientist.commands.implementation.model import ModelCommand
 
         cmd = ModelCommand()
@@ -122,7 +120,7 @@ class TestModelCommandPickerCancelled:
             "EvoScientist.EvoScientist._ensure_config",
             return_value=cfg,
         ):
-            _run(cmd.execute(ctx, []))
+            await cmd.execute(ctx, [])
 
         # No model switch should have happened
         ui.append_system.assert_not_called()
@@ -131,7 +129,7 @@ class TestModelCommandPickerCancelled:
 class TestModelCommandSwitch:
     """Verify a successful model switch updates config and rebuilds agent."""
 
-    def test_switch_known_model(self):
+    async def test_switch_known_model(self):
         from EvoScientist.commands.implementation.model import ModelCommand
 
         cmd = ModelCommand()
@@ -158,7 +156,7 @@ class TestModelCommandSwitch:
                 return_value=new_agent,
             ),
         ):
-            _run(cmd.execute(ctx, ["claude-opus-4-8"]))
+            await cmd.execute(ctx, ["claude-opus-4-8"])
 
         # The switch is committed via set_active_config(temp_cfg), not by
         # mutating the original cfg object in place.
@@ -176,7 +174,7 @@ class TestModelCommandSwitch:
         assert "claude-opus-4-8" in msg
         assert "anthropic" in msg
 
-    def test_switch_with_save_flag(self):
+    async def test_switch_with_save_flag(self):
         from EvoScientist.commands.implementation.model import ModelCommand
 
         cmd = ModelCommand()
@@ -203,7 +201,7 @@ class TestModelCommandSwitch:
             ),
             patch("EvoScientist.config.settings.set_config_value") as mock_save,
         ):
-            _run(cmd.execute(ctx, ["claude-opus-4-8", "--save"]))
+            await cmd.execute(ctx, ["claude-opus-4-8", "--save"])
 
         # Config file should be updated
         mock_save.assert_any_call("model", "claude-opus-4-8")
@@ -213,7 +211,7 @@ class TestModelCommandSwitch:
         msg = ui.append_system.call_args[0][0]
         assert "saved to config" in msg
 
-    def test_switch_without_save_flag_does_not_persist(self):
+    async def test_switch_without_save_flag_does_not_persist(self):
         from EvoScientist.commands.implementation.model import ModelCommand
 
         cmd = ModelCommand()
@@ -240,7 +238,7 @@ class TestModelCommandSwitch:
             ),
             patch("EvoScientist.config.settings.set_config_value") as mock_save,
         ):
-            _run(cmd.execute(ctx, ["claude-opus-4-8"]))
+            await cmd.execute(ctx, ["claude-opus-4-8"])
 
         # Config file should NOT be updated
         mock_save.assert_not_called()
@@ -253,7 +251,7 @@ class TestModelCommandSwitch:
 class TestModelCommandFailure:
     """Verify error handling when chat-model construction raises."""
 
-    def test_build_chat_model_error(self):
+    async def test_build_chat_model_error(self):
         from EvoScientist.commands.implementation.model import ModelCommand
 
         cmd = ModelCommand()
@@ -276,7 +274,7 @@ class TestModelCommandFailure:
                 side_effect=RuntimeError("API key missing"),
             ) as mock_build,
         ):
-            _run(cmd.execute(ctx, ["claude-opus-4-8"]))
+            await cmd.execute(ctx, ["claude-opus-4-8"])
 
         mock_build.assert_called_once()
         ui.append_system.assert_called_once()
@@ -446,7 +444,7 @@ class TestApplyModelIntegration:
     pair so we can assert on identity.
     """
 
-    def test_new_agent_is_bound_to_newly_selected_model(self, evo_module_state):
+    async def test_new_agent_is_bound_to_newly_selected_model(self, evo_module_state):
         from EvoScientist.commands.implementation.model import ModelCommand
         from EvoScientist.config.settings import EvoScientistConfig
 
@@ -502,7 +500,7 @@ class TestApplyModelIntegration:
             ),
         ):
             cmd = ModelCommand()
-            _run(cmd._apply_model(ctx, "minimax-m2.7", "openrouter"))
+            await cmd._apply_model(ctx, "minimax-m2.7", "openrouter")
 
         # The agent produced by _apply_model must be bound to the
         # NEWLY requested model, threaded in via chat_model=.
@@ -539,7 +537,9 @@ class TestApplyModelPreservesConfigByReference:
     switch (the held object stops being the active ``_config`` after the first).
     """
 
-    def test_held_config_reference_tracks_repeated_switches(self, evo_module_state):
+    async def test_held_config_reference_tracks_repeated_switches(
+        self, evo_module_state
+    ):
         from EvoScientist.commands.implementation.model import ModelCommand
         from EvoScientist.config.settings import EvoScientistConfig
 
@@ -592,7 +592,7 @@ class TestApplyModelPreservesConfigByReference:
                 ("minimax-m2.7", "openrouter"),
                 ("claude-sonnet-4-6", "anthropic"),
             ]:
-                _run(cmd._apply_model(ctx, model, provider))
+                await cmd._apply_model(ctx, model, provider)
                 # The held reference must reflect the LATEST switch on every
                 # iteration — not just the first — and stay the active config.
                 assert agent_holder["config"].model == model
@@ -610,7 +610,7 @@ class TestModelCommandLoadAgentFailure:
     the ordering could silently regress (e.g. if ``_apply_model`` were
     reordered to call ``set_chat_model`` first)."""
 
-    def test_load_agent_error_is_transactional(self):
+    async def test_load_agent_error_is_transactional(self):
         from EvoScientist.commands.implementation.model import ModelCommand
 
         cmd = ModelCommand()
@@ -646,7 +646,7 @@ class TestModelCommandLoadAgentFailure:
             # Pass ``--save`` to strengthen the assertion: if the ordering
             # ever regresses, ``set_config_value`` would be called with
             # stale data.
-            _run(cmd.execute(ctx, ["claude-opus-4-8", "--save"]))
+            await cmd.execute(ctx, ["claude-opus-4-8", "--save"])
 
         # _load_agent was attempted (transactional first step).
         mock_load.assert_called_once()
@@ -677,7 +677,7 @@ class TestApplyModelLoadAgentFailureTransactional:
     downstream setters never run on failure.
     """
 
-    def test_globals_unchanged_when_load_agent_raises(self, evo_module_state):
+    async def test_globals_unchanged_when_load_agent_raises(self, evo_module_state):
         from EvoScientist.commands.implementation.model import ModelCommand
         from EvoScientist.config.settings import EvoScientistConfig
 
@@ -721,7 +721,7 @@ class TestApplyModelLoadAgentFailureTransactional:
             ),
         ):
             cmd = ModelCommand()
-            _run(cmd._apply_model(ctx, "minimax-m2.7", "openrouter"))
+            await cmd._apply_model(ctx, "minimax-m2.7", "openrouter")
 
         # All four globals are unchanged — nothing was committed.
         assert mod._config is cfg
@@ -753,7 +753,7 @@ class TestModelCommandOllamaPicker:
         ctx.ui = ui
         return ctx, cfg, ui
 
-    def test_picker_entries_include_detected_ollama_models(self):
+    async def test_picker_entries_include_detected_ollama_models(self):
         """When Ollama is reachable, detected models appear in entries with
         provider='ollama' and the Custom sentinel is appended."""
         from EvoScientist.commands.implementation.model import ModelCommand
@@ -773,7 +773,7 @@ class TestModelCommandOllamaPicker:
                 side_effect=fake_discover,
             ),
         ):
-            _run(ModelCommand().execute(ctx, []))
+            await ModelCommand().execute(ctx, [])
 
         entries = ui.wait_for_model_pick.call_args[0][0]
         ollama_rows = [(n, mid, p) for (n, mid, p) in entries if p == "ollama"]
@@ -785,7 +785,7 @@ class TestModelCommandOllamaPicker:
             "ollama",
         ) in ollama_rows
 
-    def test_picker_entries_include_sentinel_when_discovery_empty(self):
+    async def test_picker_entries_include_sentinel_when_discovery_empty(self):
         """Daemon unreachable / no models pulled — sentinel is the user's
         escape hatch and must always be present."""
         from EvoScientist.commands.implementation.model import ModelCommand
@@ -805,7 +805,7 @@ class TestModelCommandOllamaPicker:
                 side_effect=fake_discover,
             ),
         ):
-            _run(ModelCommand().execute(ctx, []))
+            await ModelCommand().execute(ctx, [])
 
         entries = ui.wait_for_model_pick.call_args[0][0]
         ollama_rows = [(n, mid, p) for (n, mid, p) in entries if p == "ollama"]
@@ -813,7 +813,7 @@ class TestModelCommandOllamaPicker:
             ("Custom Ollama model...", "__custom_ollama__", "ollama")
         ]
 
-    def test_picker_skips_ollama_section_when_not_configured(self):
+    async def test_picker_skips_ollama_section_when_not_configured(self):
         """ollama_base_url unset → no discovery call, no ollama entries,
         no sentinel (issue non-goal: no implicit localhost detection)."""
         from EvoScientist.commands.implementation.model import ModelCommand
@@ -832,13 +832,13 @@ class TestModelCommandOllamaPicker:
                 discovery,
             ),
         ):
-            _run(ModelCommand().execute(ctx, []))
+            await ModelCommand().execute(ctx, [])
 
         discovery.assert_not_called()
         entries = ui.wait_for_model_pick.call_args[0][0]
         assert not any(p == "ollama" for (_, _, p) in entries)
 
-    def test_picker_handles_cfg_without_ollama_base_url_attr(self):
+    async def test_picker_handles_cfg_without_ollama_base_url_attr(self):
         """getattr(cfg, 'ollama_base_url', None) fallback: old configs
         (or SimpleNamespace test fixtures) may not carry the attribute
         at all. Must not raise AttributeError, must not probe."""
@@ -864,13 +864,13 @@ class TestModelCommandOllamaPicker:
                 discovery,
             ),
         ):
-            _run(ModelCommand().execute(ctx, []))
+            await ModelCommand().execute(ctx, [])
 
         discovery.assert_not_called()
         entries = ui.wait_for_model_pick.call_args[0][0]
         assert not any(p == "ollama" for (_, _, p) in entries)
 
-    def test_picker_sentinel_result_is_treated_as_cancel(self):
+    async def test_picker_sentinel_result_is_treated_as_cancel(self):
         """Defense-in-depth: if the widget ever returns the sentinel name
         itself (shouldn't happen — it should substitute the typed name),
         dispatch treats it as a cancel and does NOT call _apply_model."""
@@ -893,12 +893,12 @@ class TestModelCommandOllamaPicker:
             ),
             patch("EvoScientist.cli.agent._load_agent") as load_agent,
         ):
-            _run(ModelCommand().execute(ctx, []))
+            await ModelCommand().execute(ctx, [])
 
         load_agent.assert_not_called()
         assert cfg.model == "claude-sonnet-4-6"  # unchanged
 
-    def test_picker_applies_detected_ollama_model(self):
+    async def test_picker_applies_detected_ollama_model(self):
         """User picks a live-detected Ollama model → _apply_model is invoked
         with (name, "ollama") and the agent is rebuilt."""
         from EvoScientist.commands.implementation.model import ModelCommand
@@ -928,7 +928,7 @@ class TestModelCommandOllamaPicker:
                 return_value=MagicMock(),
             ),
         ):
-            _run(ModelCommand().execute(ctx, []))
+            await ModelCommand().execute(ctx, [])
 
         # Committed via set_active_config(temp_cfg); original cfg untouched.
         set_cfg.assert_called_once()

tests/test_model_fallback.py

@@ -6,6 +6,7 @@ fallback chain behaviour via _try_fallbacks / _guard_and_fallback.
 
 from __future__ import annotations
 
+from types import SimpleNamespace
 from unittest.mock import AsyncMock, MagicMock, patch
 
 import pytest
@@ -20,7 +21,6 @@ from EvoScientist.middleware.model_fallback import (
     clear_fallbacks,
     set_ui_emit,
 )
-from tests.conftest import run_async as _run
 
 # ── Helpers ──────────────────────────────────────────────────────
 
@@ -84,6 +84,7 @@ class TestIsNonFallbackable:
             "Error 400: invalid_request_error",
             "400 Bad Request: invalid request body",
             "400: malformed JSON in request",
+            "<400> InvalidParameter: Repetitive tool calls detected in history",
         ],
     )
     def test_malformed_request_400_patterns(self, msg):
@@ -146,7 +147,7 @@ class TestIsNonFallbackable:
 class TestTryFallbacks:
     """End-to-end tests for the fallback chain traversal."""
 
-    def test_first_fallback_succeeds(self):
+    async def test_first_fallback_succeeds(self):
         """When the first fallback model works, return its response."""
         add_fallback("fb-model", "fb-provider")
         req = _fake_request()
@@ -154,13 +155,13 @@ class TestTryFallbacks:
 
         with patch("EvoScientist.llm.models.get_chat_model") as mock_gcm:
             mock_gcm.return_value = MagicMock()
-            result = _run(_try_fallbacks(req, invoke, Exception("503 boom")))
+            result = await _try_fallbacks(req, invoke, Exception("503 boom"))
 
         assert result is AI_RESPONSE
         invoke.assert_awaited_once()
         mock_gcm.assert_called_once_with(model="fb-model", provider="fb-provider")
 
-    def test_skips_failing_fallback_tries_next(self):
+    async def test_skips_failing_fallback_tries_next(self):
         """When the first fallback fails, try the second."""
         add_fallback("fb-bad", "prov-a")
         add_fallback("fb-good", "prov-b")
@@ -177,12 +178,12 @@ class TestTryFallbacks:
 
         with patch("EvoScientist.llm.models.get_chat_model") as mock_gcm:
             mock_gcm.return_value = MagicMock()
-            result = _run(_try_fallbacks(req, _invoke, Exception("503 boom")))
+            result = await _try_fallbacks(req, _invoke, Exception("503 boom"))
 
         assert result is AI_RESPONSE
         assert call_count == 2
 
-    def test_all_fallbacks_exhausted_raises_last(self):
+    async def test_all_fallbacks_exhausted_raises_last(self):
         """When every fallback fails, re-raise the last exception."""
         add_fallback("fb-a", "prov-a")
         add_fallback("fb-b", "prov-b")
@@ -202,11 +203,11 @@ class TestTryFallbacks:
         with patch("EvoScientist.llm.models.get_chat_model") as mock_gcm:
             mock_gcm.return_value = MagicMock()
             with pytest.raises(Exception, match="429 from fb-b") as exc_info:
-                _run(_try_fallbacks(req, _invoke, Exception("503 primary")))
+                await _try_fallbacks(req, _invoke, Exception("503 primary"))
 
         assert exc_info.value is last_error
 
-    def test_non_fallbackable_in_chain_aborts_immediately(self):
+    async def test_non_fallbackable_in_chain_aborts_immediately(self):
         """A non-fallbackable error from a fallback model aborts the chain."""
         add_fallback("fb-a", "prov-a")
         add_fallback("fb-b", "prov-b")  # should never be reached
@@ -218,12 +219,95 @@ class TestTryFallbacks:
         with patch("EvoScientist.llm.models.get_chat_model") as mock_gcm:
             mock_gcm.return_value = MagicMock()
             with pytest.raises(Exception, match="context_length_exceeded"):
-                _run(_try_fallbacks(req, _invoke, Exception("503 primary")))
+                await _try_fallbacks(req, _invoke, Exception("503 primary"))
 
         # get_chat_model should only have been called once (for fb-a),
         # fb-b should never be reached.
         assert mock_gcm.call_count == 1
 
+    async def test_exhausted_fallbacks_attribute_to_last_failing_model(self):
+        """Regression: when every fallback fails, the raised
+        ``ProviderStreamError`` must be attributed to the model that
+        ACTUALLY failed last, not the original ``request.model``.
+        Prevents a ``deepseek → moonshot`` chain from surfacing as
+        ``provider: deepseek`` after moonshot exhausts its quota.
+        """
+        from EvoScientist.llm.errors import ProviderStreamError
+
+        add_fallback("moonshot-model", "moonshot")
+        # Original request's model is openai-shape. Fallback's model
+        # will be openai-shape with a moonshot base_url.
+        req = _fake_request()
+
+        # ChatOpenAI-shape model instance so ``_provider_from_model``
+        # returns a recognized provider.
+        def _make_openai_model(base_url=None):
+            cls = type(
+                "ChatOpenAI",
+                (),
+                {"__module__": "langchain_openai.chat_models.base"},
+            )
+            inst = cls()
+            inst.openai_api_base = base_url
+            return inst
+
+        req.model = _make_openai_model()  # primary
+        fallback_model = _make_openai_model(base_url="https://api.moonshot.cn/v1")
+        # ``request.override(model=...)`` must return the request with the
+        # new model so ``_try_fallbacks`` tracks the failing model.
+        req.override = MagicMock(
+            side_effect=lambda **kw: SimpleNamespace(model=kw.get("model", req.model))
+        )
+
+        async def _invoke(_r):
+            raise Exception("429 quota exceeded")
+
+        with patch("EvoScientist.llm.models.get_chat_model") as mock_gcm:
+            mock_gcm.return_value = fallback_model
+            with pytest.raises(ProviderStreamError) as exc_info:
+                await _try_fallbacks(req, _invoke, Exception("openai primary failed"))
+
+        # Attribution flipped to moonshot (the failing fallback), not
+        # openai (the original request's model).
+        assert exc_info.value.provider == "moonshot"
+        assert "quota exceeded" in exc_info.value.message
+
+    async def test_langgraph_error_at_fallback_raise_point_passes_through(self):
+        """Regression: ``_raise_normalized`` calls ``_normalize``
+        directly, so its ``_should_pass_through`` gate must fire even
+        without the ``ErrorNormalizationMiddleware`` wrap sites' own
+        check. Prevents a ``langgraph.errors.*`` exception hitting the
+        fallback chain from being wrapped as a provider incident.
+        """
+        from langgraph.errors import InvalidUpdateError
+
+        add_fallback("fb-a", "prov-a")
+        req = _fake_request()
+
+        # Use a recognized-provider model so ``_provider_from_model``
+        # wouldn't short-circuit — the guard has to come from
+        # ``_should_pass_through``, not the provider check.
+        cls = type(
+            "ChatOpenAI", (), {"__module__": "langchain_openai.chat_models.base"}
+        )
+        model = cls()
+        model.openai_api_base = None
+        req.model = model
+        req.override = MagicMock(
+            side_effect=lambda **kw: SimpleNamespace(model=kw.get("model", req.model))
+        )
+
+        raised = InvalidUpdateError("state mismatch")
+
+        async def _invoke(_r):
+            raise raised
+
+        with patch("EvoScientist.llm.models.get_chat_model") as mock_gcm:
+            mock_gcm.return_value = model
+            with pytest.raises(InvalidUpdateError) as exc_info:
+                await _try_fallbacks(req, _invoke, Exception("primary failed"))
+        assert exc_info.value is raised
+
 
 # ═════════════════════════════════════════════════════════════════
 # 3. _guard_and_fallback — pre-check before chain walk
@@ -233,43 +317,69 @@ class TestTryFallbacks:
 class TestGuardAndFallback:
     """Verify that non-fallbackable errors are re-raised before trying the chain."""
 
-    def test_context_overflow_raises_immediately(self):
+    async def test_context_overflow_raises_immediately(self):
         add_fallback("fb", "prov")
         req = _fake_request()
         invoke = AsyncMock()
 
         with pytest.raises(ContextOverflowError):
-            _run(_guard_and_fallback(ContextOverflowError("overflow"), req, invoke))
+            await _guard_and_fallback(ContextOverflowError("overflow"), req, invoke)
 
         invoke.assert_not_awaited()
 
-    def test_malformed_400_raises_immediately(self):
+    async def test_context_overflow_with_provider_model_passes_through_unwrapped(self):
+        """Regression: a ``ContextOverflowError`` entering
+        ``_guard_and_fallback`` under a recognized-provider model must
+        come out unwrapped. Otherwise ``_raise_normalized`` →
+        ``_normalize`` would wrap it as a ``ProviderStreamError`` and
+        deepagents' ``SummarizationMiddleware`` (which sits outside
+        the user middleware stack and catches by exact type) would
+        stop compressing history and retrying.
+        """
+        add_fallback("fb", "prov")
+        req = _fake_request()
+        # Recognized provider — without the gate in ``_normalize`` this
+        # would wrap. With the gate, the raw type propagates.
+        cls = type(
+            "ChatOpenAI", (), {"__module__": "langchain_openai.chat_models.base"}
+        )
+        model = cls()
+        model.openai_api_base = None
+        req.model = model
+        invoke = AsyncMock()
+
+        raised = ContextOverflowError("context length exceeded")
+        with pytest.raises(ContextOverflowError) as exc_info:
+            await _guard_and_fallback(raised, req, invoke)
+
+        assert exc_info.value is raised
+        invoke.assert_not_awaited()
+
+    async def test_malformed_400_raises_immediately(self):
         add_fallback("fb", "prov")
         req = _fake_request()
         invoke = AsyncMock()
 
         with pytest.raises(Exception, match="invalid_request_error"):
-            _run(
-                _guard_and_fallback(
-                    Exception("400: invalid_request_error"), req, invoke
-                )
+            await _guard_and_fallback(
+                Exception("400: invalid_request_error"), req, invoke
             )
 
         invoke.assert_not_awaited()
 
-    def test_server_error_proceeds_to_fallback(self):
+    async def test_server_error_proceeds_to_fallback(self):
         add_fallback("fb", "prov")
         req = _fake_request()
         invoke = AsyncMock(return_value=AI_RESPONSE)
 
         with patch("EvoScientist.llm.models.get_chat_model") as mock_gcm:
             mock_gcm.return_value = MagicMock()
-            result = _run(_guard_and_fallback(Exception("503 overloaded"), req, invoke))
+            result = await _guard_and_fallback(Exception("503 overloaded"), req, invoke)
 
         assert result is AI_RESPONSE
         invoke.assert_awaited_once()
 
-    def test_auth_error_proceeds_to_fallback(self):
+    async def test_auth_error_proceeds_to_fallback(self):
         """Auth errors should try the fallback chain (different provider)."""
         add_fallback("fb", "other-prov")
         req = _fake_request()
@@ -277,10 +387,8 @@ class TestGuardAndFallback:
 
         with patch("EvoScientist.llm.models.get_chat_model") as mock_gcm:
             mock_gcm.return_value = MagicMock()
-            result = _run(
-                _guard_and_fallback(
-                    Exception("400 Bad Request: invalid_api_key"), req, invoke
-                )
+            result = await _guard_and_fallback(
+                Exception("400 Bad Request: invalid_api_key"), req, invoke
             )
 
         assert result is AI_RESPONSE
@@ -295,7 +403,7 @@ class TestGuardAndFallback:
 class TestUiEmit:
     """Verify that fallback events are surfaced via the registered callback."""
 
-    def test_emit_captures_messages(self):
+    async def test_emit_captures_messages(self):
         add_fallback("fb", "prov")
         req = _fake_request()
         invoke = AsyncMock(return_value=AI_RESPONSE)
@@ -305,14 +413,14 @@ class TestUiEmit:
 
         with patch("EvoScientist.llm.models.get_chat_model") as mock_gcm:
             mock_gcm.return_value = MagicMock()
-            _run(_try_fallbacks(req, invoke, Exception("503 down")))
+            await _try_fallbacks(req, invoke, Exception("503 down"))
 
         texts = [t for t, _ in messages]
         assert any("Primary model failed" in t for t in texts)
         assert any("Falling back to fb (prov)" in t for t in texts)
         assert any("succeeded" in t for t in texts)
 
-    def test_emit_shows_non_fallbackable_rejection(self):
+    async def test_emit_shows_non_fallbackable_rejection(self):
         add_fallback("fb", "prov")
         req = _fake_request()
         invoke = AsyncMock()
@@ -321,7 +429,7 @@ class TestUiEmit:
         set_ui_emit(lambda text, style: messages.append((text, style)))
 
         with pytest.raises(ContextOverflowError):
-            _run(_guard_and_fallback(ContextOverflowError("overflow"), req, invoke))
+            await _guard_and_fallback(ContextOverflowError("overflow"), req, invoke)
 
         texts = [t for t, _ in messages]
         assert any("not eligible for fallback" in t for t in texts)

tests/test_model_passthrough_patch.py

@@ -16,7 +16,6 @@ from unittest.mock import AsyncMock, MagicMock, patch
 import pytest
 
 from EvoScientist.llm import patches as patches_mod
-from tests.conftest import run_async as _run
 
 # =============================================================================
 # Helpers
@@ -152,7 +151,7 @@ class TestStartAsyncTaskInjection:
             "configurable": {"model": "gpt-5", "model_provider": "openai"}
         }
 
-    def test_async_start_injects_config(self, restore_model_passthrough_patch):
+    async def test_async_start_injects_config(self, restore_model_passthrough_patch):
         try:
             from deepagents.middleware import async_subagents as ds_mod
         except ImportError:
@@ -176,12 +175,10 @@ class TestStartAsyncTaskInjection:
             "EvoScientist.EvoScientist._ensure_config",
             return_value=_stub_cfg(model="claude-haiku-4-5", provider="anthropic"),
         ):
-            _run(
-                tool.coroutine(
-                    description="hi",
-                    subagent_type="writing-agent",
-                    runtime=_runtime_stub(),
-                )
+            await tool.coroutine(
+                description="hi",
+                subagent_type="writing-agent",
+                runtime=_runtime_stub(),
             )
 
         runs_async.create.assert_awaited_once()
@@ -267,7 +264,7 @@ class TestUpdateAsyncTaskInjection:
             "last_updated_at": "2026-05-07T00:00:00Z",
         }
 
-    def test_async_update_injects_config(self, restore_model_passthrough_patch):
+    async def test_async_update_injects_config(self, restore_model_passthrough_patch):
         """The async coroutine path must inject config too."""
         try:
             from deepagents.middleware import async_subagents as ds_mod
@@ -296,12 +293,10 @@ class TestUpdateAsyncTaskInjection:
             "EvoScientist.EvoScientist._ensure_config",
             return_value=_stub_cfg(model="gpt-5", provider="openai"),
         ):
-            _run(
-                tool.coroutine(
-                    task_id="thread-001",
-                    message="follow up async",
-                    runtime=runtime,
-                )
+            await tool.coroutine(
+                task_id="thread-001",
+                message="follow up async",
+                runtime=runtime,
             )
 
         runs_async.create.assert_awaited_once()

tests/test_new_command.py

@@ -2,11 +2,9 @@
 
 from unittest.mock import AsyncMock, MagicMock
 
-from tests.conftest import run_async as _run
-
 
 class TestNewCommand:
-    def test_execute_calls_start_new_session(self):
+    async def test_execute_calls_start_new_session(self):
         from EvoScientist.commands.base import CommandContext
         from EvoScientist.commands.implementation.session import NewCommand
 
@@ -18,7 +16,7 @@ class TestNewCommand:
             ui=ui,
             workspace_dir="/old/ws",
         )
-        _run(NewCommand().execute(ctx, []))
+        await NewCommand().execute(ctx, [])
         ui.start_new_session.assert_awaited_once()
 
     def test_requires_agent_false(self):
@@ -26,7 +24,7 @@ class TestNewCommand:
 
         assert NewCommand().requires_agent is False
 
-    def test_no_agent_access(self):
+    async def test_no_agent_access(self):
         """Command body must not touch ctx.agent (it's still loading)."""
         from EvoScientist.commands.base import CommandContext
         from EvoScientist.commands.implementation.session import NewCommand
@@ -35,4 +33,4 @@ class TestNewCommand:
         ui.start_new_session = AsyncMock()
         ctx = CommandContext(agent=None, thread_id="tid", ui=ui)
         # No AttributeError even though ctx.agent is None
-        _run(NewCommand().execute(ctx, []))
+        await NewCommand().execute(ctx, [])

tests/test_observation_memory.py

@@ -1655,9 +1655,7 @@ def test_turn_compaction_uses_latest_user_turn_only():
     ]
 
 
-def test_lifecycle_schedules_turn_worker_without_awaiting(
-    tmp_path, monkeypatch, run_async
-):
+async def test_lifecycle_schedules_turn_worker_without_awaiting(tmp_path, monkeypatch):
     memory_dir = tmp_path / "memories"
     workspace_dir = tmp_path / "workspace"
     calls = []
@@ -1682,21 +1680,18 @@ def test_lifecycle_schedules_turn_worker_without_awaiting(
     )
     runtime = _runtime("thread-1")
 
-    async def run():
-        state: AgentState[object] = {
-            "messages": [
-                HumanMessage("previous turn"),
-                AIMessage("previous answer"),
-                HumanMessage("hi"),
-                AIMessage("done"),
-            ]
-        }
-        await middleware.aafter_agent(
-            state,
-            runtime,
-        )
-
-    run_async(run())
+    state: AgentState[object] = {
+        "messages": [
+            HumanMessage("previous turn"),
+            AIMessage("previous answer"),
+            HumanMessage("hi"),
+            AIMessage("done"),
+        ]
+    }
+    await middleware.aafter_agent(
+        state,
+        runtime,
+    )
 
     assert len(calls) == 1
     request, hooks = calls[0]
@@ -2182,10 +2177,9 @@ def test_observation_linker_does_not_launch_when_observations_disabled(
     launch_call.assert_not_called()
 
 
-def test_async_observation_linker_does_not_launch_when_observations_disabled(
+async def test_async_observation_linker_does_not_launch_when_observations_disabled(
     tmp_path,
     monkeypatch,
-    run_async,
 ):
     context = _linker_context(
         memory_dir=tmp_path / "memories",
@@ -2200,7 +2194,7 @@ def test_async_observation_linker_does_not_launch_when_observations_disabled(
     launch_call = MagicMock()
     monkeypatch.setattr(memory_launch, "alaunch_background_run", launch_call)
 
-    run = run_async(memory_launch.alaunch_observation_linker(context))
+    run = await memory_launch.alaunch_observation_linker(context)
 
     assert run is None
     launch_call.assert_not_called()
@@ -2300,7 +2294,11 @@ def test_memory_worker_observation_writer_modes(
         observation_writer=observation_writer,
     )
 
-    assert type(middleware[0]).__name__ == "ToolErrorHandlerMiddleware"
+    # ErrorNormalizationMiddleware wraps outermost so provider-SDK
+    # exceptions from the auxiliary model call get normalized before
+    # the tool-error handler sees them.
+    assert type(middleware[0]).__name__ == "ErrorNormalizationMiddleware"
+    assert type(middleware[1]).__name__ == "ToolErrorHandlerMiddleware"
     assert _memory_tool_names(middleware) == expected_tools
 
 
@@ -2335,8 +2333,8 @@ def test_sync_memory_worker_watcher_untracks_without_counting_on_poll_abort(
     assert status.observations_recorded == 0
 
 
-def test_async_memory_worker_watcher_untracks_without_counting_on_poll_abort(
-    tmp_path, monkeypatch, run_async
+async def test_async_memory_worker_watcher_untracks_without_counting_on_poll_abort(
+    tmp_path, monkeypatch
 ):
     memory_dir = tmp_path / "memories"
     _mark_worker_started(memory_dir)
@@ -2348,14 +2346,12 @@ def test_async_memory_worker_watcher_untracks_without_counting_on_poll_abort(
         async def get(self, **_kwargs):
             raise RuntimeError("poll failed")
 
-    run_async(
-        background_runs.awatch_background_run(
-            SimpleNamespace(runs=_Runs()),
-            thread_id="worker-thread",
-            run_id="run-1",
-            hooks=memory_launch._memory_worker_launch_hooks(memory_dir),
-            watcher_config=_fast_watcher_config(max_poll_failures=1),
-        )
+    await background_runs.awatch_background_run(
+        SimpleNamespace(runs=_Runs()),
+        thread_id="worker-thread",
+        run_id="run-1",
+        hooks=memory_launch._memory_worker_launch_hooks(memory_dir),
+        watcher_config=_fast_watcher_config(max_poll_failures=1),
     )
     status = worker_activity.memory_worker_status()
     assert status.is_running is False
@@ -2363,8 +2359,8 @@ def test_async_memory_worker_watcher_untracks_without_counting_on_poll_abort(
     assert status.observations_recorded == 0
 
 
-def test_async_memory_worker_watcher_counts_completion_under_blockbuster(
-    tmp_path, run_async
+async def test_async_memory_worker_watcher_counts_completion_under_blockbuster(
+    tmp_path,
 ):
     memory_dir = tmp_path / "memories"
     _mark_worker_started(memory_dir)
@@ -2376,20 +2372,17 @@ def test_async_memory_worker_watcher_counts_completion_under_blockbuster(
         async def get(self, **_kwargs):
             return {"status": "success"}
 
-    async def run():
-        blocker = BlockBuster(scanned_modules=[memory_worker, worker_activity])
-        blocker.activate()
-        try:
-            await background_runs.awatch_background_run(
-                SimpleNamespace(runs=_Runs()),
-                thread_id="worker-thread",
-                run_id="run-1",
-                hooks=memory_launch._memory_worker_launch_hooks(memory_dir),
-            )
-        finally:
-            blocker.deactivate()
-
-    run_async(run())
+    blocker = BlockBuster(scanned_modules=[memory_worker, worker_activity])
+    blocker.activate()
+    try:
+        await background_runs.awatch_background_run(
+            SimpleNamespace(runs=_Runs()),
+            thread_id="worker-thread",
+            run_id="run-1",
+            hooks=memory_launch._memory_worker_launch_hooks(memory_dir),
+        )
+    finally:
+        blocker.deactivate()
     status = worker_activity.memory_worker_status()
     assert status.is_running is False
     assert status.profile_updates == 1
@@ -2527,7 +2520,7 @@ def test_memory_worker_marks_active_status(tmp_path, monkeypatch):
     assert status.observations_recorded == 1
 
 
-def test_async_memory_worker_offloads_blocking_work(tmp_path, monkeypatch, run_async):
+async def test_async_memory_worker_offloads_blocking_work(tmp_path, monkeypatch):
     monkeypatch.setattr(
         background_runs, "default_background_run_url", lambda: "http://x"
     )
@@ -2569,22 +2562,18 @@ def test_async_memory_worker_offloads_blocking_work(tmp_path, monkeypatch, run_a
 
     spawned: list[background_runs.BackgroundRun] = []
 
-    async def run():
-        event_loop_thread = threading.get_ident()
-        context = _memory_source_context(
-            memory_dir=tmp_path / "memories",
-            workspace_dir=tmp_path / "workspace",
-            trajectory=[{"role": "human", "content": "hi"}],
-        )
-        request = memory_launch.memory_worker_launch_request(context)
-        await background_runs.alaunch_background_run(
-            request,
-            hooks=memory_launch._memory_worker_launch_hooks(tmp_path / "memories"),
-            spawn_status_watcher=spawned.append,
-        )
-        return event_loop_thread
-
-    event_loop_thread = run_async(run())
+    event_loop_thread = threading.get_ident()
+    context = _memory_source_context(
+        memory_dir=tmp_path / "memories",
+        workspace_dir=tmp_path / "workspace",
+        trajectory=[{"role": "human", "content": "hi"}],
+    )
+    request = memory_launch.memory_worker_launch_request(context)
+    await background_runs.alaunch_background_run(
+        request,
+        hooks=memory_launch._memory_worker_launch_hooks(tmp_path / "memories"),
+        spawn_status_watcher=spawned.append,
+    )
     assert [name for name, _thread_id in call_threads] == ["health", "snapshot"]
     assert all(thread_id != event_loop_thread for _name, thread_id in call_threads)
     assert worker_activity.memory_worker_status().is_running is True

tests/test_ollama_discovery.py

@@ -17,7 +17,6 @@ from EvoScientist.llm.ollama_discovery import (
     discover_ollama_models,
     validate_ollama_connection,
 )
-from tests.conftest import run_async as _run
 
 
 class TestValidateOllamaConnection:
@@ -71,17 +70,17 @@ class TestValidateOllamaConnection:
 class TestDiscoverOllamaModels:
     """Async probe — contract: never raise, return list[str]."""
 
-    def test_empty_base_url_returns_empty_without_http(self):
+    async def test_empty_base_url_returns_empty_without_http(self):
         # No HTTP call should be made for an empty base_url — verified by
         # the fact that no mock is set up and the test completes.
-        names = _run(discover_ollama_models(""))
+        names = await discover_ollama_models("")
         assert names == []
 
-    def test_none_base_url_returns_empty(self):
-        names = _run(discover_ollama_models(None))
+    async def test_none_base_url_returns_empty(self):
+        names = await discover_ollama_models(None)
         assert names == []
 
-    def test_200_returns_names(self):
+    async def test_200_returns_names(self):
         async def fake_get(self, url):
             resp = MagicMock()
             resp.status_code = 200
@@ -93,10 +92,10 @@ class TestDiscoverOllamaModels:
             return resp
 
         with patch.object(httpx.AsyncClient, "get", fake_get):
-            names = _run(discover_ollama_models("http://localhost:11434"))
+            names = await discover_ollama_models("http://localhost:11434")
         assert names == ["llama3.3:latest", "qwen3:8b"]
 
-    def test_strips_entries_without_name(self):
+    async def test_strips_entries_without_name(self):
         async def fake_get(self, url):
             resp = MagicMock()
             resp.status_code = 200
@@ -112,36 +111,36 @@ class TestDiscoverOllamaModels:
             return resp
 
         with patch.object(httpx.AsyncClient, "get", fake_get):
-            names = _run(discover_ollama_models("http://localhost:11434"))
+            names = await discover_ollama_models("http://localhost:11434")
         assert names == ["llama3.3"]
 
-    def test_timeout_returns_empty(self):
+    async def test_timeout_returns_empty(self):
         async def fake_get(self, url):
             raise httpx.TimeoutException("timed out")
 
         with patch.object(httpx.AsyncClient, "get", fake_get):
-            names = _run(discover_ollama_models("http://localhost:11434"))
+            names = await discover_ollama_models("http://localhost:11434")
         assert names == []
 
-    def test_connect_error_returns_empty(self):
+    async def test_connect_error_returns_empty(self):
         async def fake_get(self, url):
             raise httpx.ConnectError("refused")
 
         with patch.object(httpx.AsyncClient, "get", fake_get):
-            names = _run(discover_ollama_models("http://localhost:11434"))
+            names = await discover_ollama_models("http://localhost:11434")
         assert names == []
 
-    def test_non_200_returns_empty(self):
+    async def test_non_200_returns_empty(self):
         async def fake_get(self, url):
             resp = MagicMock()
             resp.status_code = 500
             return resp
 
         with patch.object(httpx.AsyncClient, "get", fake_get):
-            names = _run(discover_ollama_models("http://localhost:11434"))
+            names = await discover_ollama_models("http://localhost:11434")
         assert names == []
 
-    def test_malformed_json_returns_empty(self):
+    async def test_malformed_json_returns_empty(self):
         async def fake_get(self, url):
             resp = MagicMock()
             resp.status_code = 200
@@ -149,10 +148,10 @@ class TestDiscoverOllamaModels:
             return resp
 
         with patch.object(httpx.AsyncClient, "get", fake_get):
-            names = _run(discover_ollama_models("http://localhost:11434"))
+            names = await discover_ollama_models("http://localhost:11434")
         assert names == []
 
-    def test_missing_models_key_returns_empty(self):
+    async def test_missing_models_key_returns_empty(self):
         async def fake_get(self, url):
             resp = MagicMock()
             resp.status_code = 200
@@ -160,10 +159,10 @@ class TestDiscoverOllamaModels:
             return resp
 
         with patch.object(httpx.AsyncClient, "get", fake_get):
-            names = _run(discover_ollama_models("http://localhost:11434"))
+            names = await discover_ollama_models("http://localhost:11434")
         assert names == []
 
-    def test_trailing_slash_stripped_from_url(self):
+    async def test_trailing_slash_stripped_from_url(self):
         called = {}
 
         async def fake_get(self, url):
@@ -174,7 +173,7 @@ class TestDiscoverOllamaModels:
             return resp
 
         with patch.object(httpx.AsyncClient, "get", fake_get):
-            _run(discover_ollama_models("http://localhost:11434/"))
+            await discover_ollama_models("http://localhost:11434/")
         assert called["url"] == "http://localhost:11434/api/tags"
 
 

tests/test_onboard.py

@@ -184,6 +184,38 @@ class TestSharedConstantsAlignment:
                 )
 
 
+class TestOAuthModeReconcile:
+    def test_reconcile_preserves_auxiliary_openai_oauth(self):
+        from EvoScientist.config.onboard.wizard import _reconcile_oauth_modes
+
+        config = EvoScientistConfig(
+            provider="minimax",
+            auxiliary_provider="openai",
+            auxiliary_model="gpt-5.5",
+            openai_auth_mode="oauth",
+            anthropic_auth_mode="oauth",
+        )
+
+        _reconcile_oauth_modes(config)
+
+        assert config.openai_auth_mode == "oauth"
+        assert config.anthropic_auth_mode == "api_key"
+
+    def test_reconcile_preserves_auxiliary_provider_without_model(self):
+        from EvoScientist.config.onboard.wizard import _reconcile_oauth_modes
+
+        config = EvoScientistConfig(
+            provider="minimax",
+            auxiliary_provider="openai",
+            auxiliary_model="",
+            openai_auth_mode="oauth",
+        )
+
+        _reconcile_oauth_modes(config)
+
+        assert config.openai_auth_mode == "oauth"
+
+
 # =============================================================================
 # Test render_progress
 # =============================================================================
@@ -386,6 +418,109 @@ class TestStepProvider:
                 _step_provider(config)
 
 
+class TestStepOAuthAuthMode:
+    @pytest.mark.parametrize(
+        (
+            "step_name",
+            "config_attr",
+            "provider_label",
+            "oauth_choice_label",
+            "ccproxy_provider",
+            "status_label",
+            "question_label",
+            "login_prompt",
+        ),
+        [
+            (
+                "_step_anthropic_auth_mode",
+                "anthropic_auth_mode",
+                "Anthropic",
+                "Claude Code OAuth",
+                "claude_api",
+                "OAuth",
+                "Authentication mode",
+                "Log in to Claude now?",
+            ),
+            (
+                "_step_openai_auth_mode",
+                "openai_auth_mode",
+                "OpenAI",
+                "Codex OAuth",
+                "codex",
+                "Codex OAuth",
+                "OpenAI authentication mode",
+                "Log in to Codex now?",
+            ),
+        ],
+    )
+    def test_oauth_wrappers_use_provider_specific_ccproxy_flow(
+        self,
+        step_name,
+        config_attr,
+        provider_label,
+        oauth_choice_label,
+        ccproxy_provider,
+        status_label,
+        question_label,
+        login_prompt,
+    ):
+        """Anthropic/OpenAI wrappers share flow but keep provider-specific IDs."""
+        from EvoScientist.config.onboard import steps as onboard_steps
+
+        config = EvoScientistConfig(**{config_attr: "oauth"})
+        select_question = MagicMock()
+        select_question.ask.return_value = "oauth"
+        confirm_question = MagicMock()
+        confirm_question.ask.return_value = True
+
+        with (
+            patch(
+                "EvoScientist.ccproxy_manager.is_ccproxy_available", return_value=True
+            ),
+            patch(
+                "EvoScientist.ccproxy_manager.check_ccproxy_auth",
+                return_value=(False, "not authenticated"),
+            ) as mock_check_auth,
+            patch(
+                "EvoScientist.config.onboard.prompter.install_navigation_keys"
+            ) as mock_nav,
+            patch(
+                "EvoScientist.config.onboard.steps.questionary.select",
+                return_value=select_question,
+            ) as mock_select,
+            patch(
+                "EvoScientist.config.onboard.steps.questionary.confirm",
+                return_value=confirm_question,
+            ) as mock_confirm,
+            patch(
+                "EvoScientist.config.onboard.steps._prompt_ccproxy_port"
+            ) as mock_port,
+            patch("EvoScientist.config.onboard.steps._run_ccproxy_login") as mock_login,
+        ):
+            result = getattr(onboard_steps, step_name)(config)
+
+        assert result == "oauth"
+        mock_nav.assert_called_once_with(select_question, with_back=True)
+        mock_port.assert_called_once_with(config)
+        mock_check_auth.assert_called_once_with(ccproxy_provider)
+        mock_login.assert_called_once_with(ccproxy_provider, status_label)
+
+        select_call = mock_select.call_args
+        assert select_call.args[0] == f"{question_label}  [Esc/← to go back]:"
+        assert select_call.kwargs["default"] == "oauth"
+        choice_titles = [
+            choice.title
+            for choice in select_call.kwargs["choices"]
+            if getattr(choice, "value", None) in {"api_key", "oauth"}
+        ]
+        assert choice_titles == [
+            f"API Key (direct {provider_label} access)",
+            f"{oauth_choice_label} (via ccproxy — no API key needed)",
+        ]
+        mock_confirm.assert_called_once()
+        assert mock_confirm.call_args.args[0] == login_prompt
+
+
 class TestStepModel:
     def test_returns_selected_model(self):
         """Test that _step_model returns selected model."""
@@ -1266,6 +1401,7 @@ class TestRunOnboard:
                 "claude-sonnet-4-6",  # Model
                 "assemble",  # Auxiliary: Assemble
                 "openai",  # Auxiliary provider (a different company)
+                "api_key",  # Auxiliary OpenAI auth mode
                 "gpt-5.5",  # Auxiliary model
                 "daemon",  # Workspace mode
                 True,  # Show thinking
@@ -1289,12 +1425,185 @@ class TestRunOnboard:
         final_config = mock_save.call_args_list[-1].args[0]
         assert final_config.auxiliary_provider == "openai"
         assert final_config.auxiliary_model == "gpt-5.5"
+        assert final_config.openai_auth_mode == "api_key"
         # The auxiliary provider's key is stored in its per-provider field.
         assert final_config.openai_api_key == "sk-aux-openai"
         # Main agent is untouched.
         assert final_config.provider == "anthropic"
         assert final_config.model == "claude-sonnet-4-6"
 
+    def test_auxiliary_same_provider_reuses_main_credentials(self):
+        """Same-provider co-pilot should not imply separate credentials exist."""
+        from EvoScientist.config.onboard.wizard import run_onboard
+
+        mock_q = MagicMock()
+        with (
+            _patch_all_questionary(mock_q),
+            patch("EvoScientist.config.onboard.wizard.load_config") as mock_load,
+            patch("EvoScientist.config.onboard.wizard.save_config") as mock_save,
+            patch("EvoScientist.config.onboard.wizard.console"),
+            patch("EvoScientist.config.onboard.steps.console"),
+        ):
+            mock_load.return_value = EvoScientistConfig(
+                provider="openai",
+                model="gpt-5.5",
+                openai_api_key="sk-main-openai",
+            )
+            mock_q.select.return_value.ask.side_effect = [
+                "assemble",  # Auxiliary: Assemble
+                "openai",  # Same provider as the main model
+                "gpt-5.5",  # Auxiliary model
+            ]
+            mock_q.confirm.return_value.ask.side_effect = [True]  # Save config
+
+            result = run_onboard(
+                skip_validation=True, only_sections={"auxiliary_model"}
+            )
+
+        assert result is True
+        final_config = mock_save.call_args_list[-1].args[0]
+        assert final_config.auxiliary_provider == "openai"
+        assert final_config.auxiliary_model == "gpt-5.5"
+        assert final_config.openai_api_key == "sk-main-openai"
+        mock_q.password.assert_not_called()
+        assert mock_q.select.return_value.ask.call_count == 3
+
+    def test_auxiliary_same_provider_prompts_when_shared_key_missing(self):
+        """Same-provider reuse should not hide a missing shared API key."""
+        from EvoScientist.config.onboard.wizard import run_onboard
+
+        mock_q = MagicMock()
+        with (
+            _patch_all_questionary(mock_q),
+            patch("EvoScientist.config.onboard.wizard.load_config") as mock_load,
+            patch("EvoScientist.config.onboard.wizard.save_config") as mock_save,
+            patch("EvoScientist.config.onboard.wizard.console"),
+            patch("EvoScientist.config.onboard.steps.console"),
+            patch("EvoScientist.config.onboard.helpers.console"),
+            patch(
+                "EvoScientist.ccproxy_manager.is_ccproxy_available", return_value=True
+            ),
+        ):
+            mock_load.return_value = EvoScientistConfig(
+                provider="openai",
+                model="gpt-5.5",
+                openai_auth_mode="api_key",
+                openai_api_key="",
+            )
+            mock_q.select.return_value.ask.side_effect = [
+                "assemble",  # Auxiliary: Assemble
+                "openai",  # Same provider as the main model
+                "api_key",  # Shared OpenAI auth mode
+                "gpt-5.5",  # Auxiliary model
+            ]
+            mock_q.password.return_value.ask.side_effect = [
+                "sk-shared-openai",
+            ]
+            mock_q.confirm.return_value.ask.side_effect = [True]  # Save config
+
+            result = run_onboard(
+                skip_validation=True, only_sections={"auxiliary_model"}
+            )
+
+        assert result is True
+        final_config = mock_save.call_args_list[-1].args[0]
+        assert final_config.auxiliary_provider == "openai"
+        assert final_config.auxiliary_model == "gpt-5.5"
+        assert final_config.openai_auth_mode == "api_key"
+        assert final_config.openai_api_key == "sk-shared-openai"
+        mock_q.password.assert_called_once()
+        assert mock_q.select.return_value.ask.call_count == 4
+
+    def test_auxiliary_openai_oauth_skips_api_key(self):
+        """Auxiliary OpenAI now uses the shared auth flow and skips keys on OAuth."""
+        from EvoScientist.config.onboard.wizard import run_onboard
+
+        mock_q = MagicMock()
+        with (
+            _patch_all_questionary(mock_q),
+            patch("EvoScientist.config.onboard.wizard.load_config") as mock_load,
+            patch("EvoScientist.config.onboard.wizard.save_config") as mock_save,
+            patch("EvoScientist.config.onboard.wizard.console"),
+            patch("EvoScientist.config.onboard.steps.console"),
+            patch("EvoScientist.config.onboard.helpers.console"),
+            patch(
+                "EvoScientist.ccproxy_manager.is_ccproxy_available", return_value=True
+            ),
+            patch(
+                "EvoScientist.ccproxy_manager.check_ccproxy_auth",
+                return_value=(False, "not authenticated"),
+            ) as mock_auth,
+        ):
+            mock_load.return_value = EvoScientistConfig()
+            mock_q.select.return_value.ask.side_effect = [
+                "assemble",  # Auxiliary: Assemble
+                "openai",  # Auxiliary provider
+                "oauth",  # OpenAI auth mode
+                "gpt-5.5",  # Auxiliary model
+            ]
+            mock_q.text.return_value.ask.side_effect = [
+                "",  # ccproxy port (keep default)
+            ]
+            mock_q.confirm.return_value.ask.side_effect = [
+                False,  # Do not log in to Codex now
+                True,  # Save config
+            ]
+
+            result = run_onboard(
+                skip_validation=True, only_sections={"auxiliary_model"}
+            )
+
+        assert result is True
+        final_config = mock_save.call_args_list[-1].args[0]
+        assert final_config.auxiliary_provider == "openai"
+        assert final_config.auxiliary_model == "gpt-5.5"
+        assert final_config.openai_auth_mode == "oauth"
+        assert final_config.openai_api_key == ""
+        mock_q.password.assert_not_called()
+        mock_auth.assert_called_once_with("codex")
+
+    def test_auxiliary_reconfigure_clears_unused_openai_oauth(self):
+        """Switching co-pilot away from OpenAI clears stale OpenAI OAuth mode."""
+        from EvoScientist.config.onboard.wizard import run_onboard
+
+        mock_q = MagicMock()
+        with (
+            _patch_all_questionary(mock_q),
+            patch("EvoScientist.config.onboard.wizard.load_config") as mock_load,
+            patch("EvoScientist.config.onboard.wizard.save_config") as mock_save,
+            patch("EvoScientist.config.onboard.wizard.console"),
+            patch("EvoScientist.config.onboard.steps.console"),
+            patch("EvoScientist.config.onboard.helpers.console"),
+        ):
+            mock_load.return_value = EvoScientistConfig(
+                provider="anthropic",
+                model="claude-sonnet-4-6",
+                anthropic_auth_mode="oauth",
+                auxiliary_provider="openai",
+                auxiliary_model="gpt-5.5",
+                openai_auth_mode="oauth",
+            )
+            mock_q.select.return_value.ask.side_effect = [
+                "assemble",  # Auxiliary: Assemble
+                "minimax",  # Auxiliary provider no longer uses OpenAI
+                "global",  # MiniMax region
+                "minimax-m2",  # Auxiliary model
+            ]
+            mock_q.password.return_value.ask.side_effect = [
+                "sk-minimax",  # MiniMax API key
+            ]
+            mock_q.confirm.return_value.ask.side_effect = [True]  # Save config
+
+            result = run_onboard(
+                skip_validation=True, only_sections={"auxiliary_model"}
+            )
+
+        assert result is True
+        final_config = mock_save.call_args_list[-1].args[0]
+        assert final_config.auxiliary_provider == "minimax"
+        assert final_config.openai_auth_mode == "api_key"
+        assert final_config.anthropic_auth_mode == "oauth"
+
     def test_auxiliary_custom_provider_collects_base_url(self):
         """Regression for the custom-provider fix: a custom auxiliary provider
         collects its base URL (provider -> base URL -> key -> model order)."""

tests/test_paths.py

@@ -21,6 +21,7 @@ def _restore_paths():
         "GLOBAL_MEMORIES_DIR": paths.GLOBAL_MEMORIES_DIR,
         "USER_SKILLS_DIR": paths.USER_SKILLS_DIR,
         "_active_workspace": paths._active_workspace,
+        "_EVOSCIENTIST_DATA_ROOT": paths._EVOSCIENTIST_DATA_ROOT,
     }
     yield
     paths.WORKSPACE_ROOT = orig["WORKSPACE_ROOT"]
@@ -32,6 +33,7 @@ def _restore_paths():
     paths.GLOBAL_MEMORIES_DIR = orig["GLOBAL_MEMORIES_DIR"]
     paths.USER_SKILLS_DIR = orig["USER_SKILLS_DIR"]
     paths._active_workspace = orig["_active_workspace"]
+    paths._EVOSCIENTIST_DATA_ROOT = orig["_EVOSCIENTIST_DATA_ROOT"]
 
 
 class TestSetWorkspaceRoot:
@@ -140,6 +142,63 @@ class TestDataDir:
         assert paths.GLOBAL_MEMORIES_DIR == paths.DATA_DIR / "memories"
 
 
+class TestGatewayDataDirs:
+    def test_evoscientist_root_prefers_home_override(self, tmp_path, monkeypatch):
+        home = tmp_path / "runtime-home"
+        monkeypatch.setenv("EVOSCIENTIST_HOME", str(home))
+
+        assert paths.evoscientist_root() == home.resolve()
+
+    def test_evoscientist_root_falls_back_to_data_dir(self, tmp_path, monkeypatch):
+        data_dir = tmp_path / "app-data"
+        monkeypatch.delenv("EVOSCIENTIST_HOME", raising=False)
+        monkeypatch.setattr(paths, "DATA_DIR", data_dir)
+
+        assert paths.evoscientist_root() == data_dir.resolve()
+
+    def test_data_root_respects_environment_override(self, tmp_path, monkeypatch):
+        data_root = tmp_path / "web-data"
+        monkeypatch.setenv("EVOSCIENTIST_DATA_ROOT", str(data_root))
+        paths._EVOSCIENTIST_DATA_ROOT = None
+
+        assert paths._data_root() == data_root.resolve()
+
+    def test_user_thread_and_global_dirs_are_created(self, tmp_path, monkeypatch):
+        data_root = tmp_path / "web-data"
+        monkeypatch.setenv("EVOSCIENTIST_DATA_ROOT", str(data_root))
+        paths._EVOSCIENTIST_DATA_ROOT = None
+
+        user_dir = paths.user_data_dir("user-a")
+        thread_dir = paths.thread_data_dir("user-a", "thread-1")
+        shared_dir = paths.global_data_dir("user-a")
+
+        assert user_dir == data_root / "user-a"
+        assert thread_dir == user_dir / "thread-1"
+        assert shared_dir == user_dir / "__global__"
+        assert user_dir.is_dir()
+        assert thread_dir.is_dir()
+        assert shared_dir.is_dir()
+
+    def test_iter_user_data_dirs_yields_directories_only(self, tmp_path, monkeypatch):
+        data_root = tmp_path / "web-data"
+        monkeypatch.setenv("EVOSCIENTIST_DATA_ROOT", str(data_root))
+        paths._EVOSCIENTIST_DATA_ROOT = None
+        paths.user_data_dir("user-a")
+        paths.user_data_dir("user-b")
+        (data_root / "metadata.json").write_text("{}", encoding="utf-8")
+
+        assert {path.name for path in paths.iter_user_data_dirs()} == {
+            "user-a",
+            "user-b",
+        }
+
+    def test_uploads_dir_uses_evoscientist_root(self, tmp_path, monkeypatch):
+        home = tmp_path / "runtime-home"
+        monkeypatch.setenv("EVOSCIENTIST_HOME", str(home))
+
+        assert paths.uploads_dir() == home.resolve() / "uploads"
+
+
 class TestLegacySessionsDbMigration:
     """Tests for migrate_legacy_sessions_db() — transitional helper.
 

tests/test_pick_skills_interactive.py

@@ -79,14 +79,13 @@ class TestPickSkillsInteractive:
 class TestInstallSkillsHandlesEmpty:
     """InstallSkills.execute must distinguish None vs [] from the picker."""
 
-    def test_empty_list_suppresses_cancel_message(self):
+    async def test_empty_list_suppresses_cancel_message(self):
         """When picker returns [], user should NOT see "Browse cancelled"
         (the picker already printed its own message)."""
         from unittest.mock import AsyncMock
 
         from EvoScientist.commands.base import CommandContext
         from EvoScientist.commands.implementation.skills import InstallSkills
-        from tests.conftest import run_async as _run
 
         ui = MagicMock()
         ui.supports_interactive = True
@@ -97,18 +96,17 @@ class TestInstallSkillsHandlesEmpty:
             "EvoScientist.tools.skills_manager.fetch_remote_skill_index",
             return_value=_INDEX,
         ):
-            _run(InstallSkills().execute(ctx, []))
+            await InstallSkills().execute(ctx, [])
 
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert not any("Browse cancelled" in m for m in msgs)
 
-    def test_none_shows_cancel_message(self):
+    async def test_none_shows_cancel_message(self):
         """When picker returns None (actual cancel), user sees the message."""
         from unittest.mock import AsyncMock
 
         from EvoScientist.commands.base import CommandContext
         from EvoScientist.commands.implementation.skills import InstallSkills
-        from tests.conftest import run_async as _run
 
         ui = MagicMock()
         ui.supports_interactive = True
@@ -119,7 +117,7 @@ class TestInstallSkillsHandlesEmpty:
             "EvoScientist.tools.skills_manager.fetch_remote_skill_index",
             return_value=_INDEX,
         ):
-            _run(InstallSkills().execute(ctx, []))
+            await InstallSkills().execute(ctx, [])
 
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Browse cancelled" in m for m in msgs)

tests/test_profile_memory_middleware.py

@@ -344,9 +344,7 @@ def test_profile_memory_uses_path_pointers_when_profiles_exceed_budget(
     )
 
 
-def test_profile_memory_async_path_bootstraps_and_injects(
-    tmp_path, monkeypatch, run_async
-):
+async def test_profile_memory_async_path_bootstraps_and_injects(tmp_path, monkeypatch):
     memories = tmp_path / "memories"
     workspace = tmp_path / "workspace"
     workspace.mkdir()
@@ -356,7 +354,7 @@ def test_profile_memory_async_path_bootstraps_and_injects(
         return request
 
     middleware = memory_module.create_memory_middleware(str(memories))
-    run_async(middleware.awrap_model_call(_request(), _handler))
+    await middleware.awrap_model_call(_request(), _handler)
 
     assert (memories / "profile" / "USER_PROFILE.md").exists()
 
@@ -399,8 +397,8 @@ def test_profile_memory_read_failure_uses_path_pointers_without_overwriting(
     assert soul_path.read_bytes() == original_bytes
 
 
-def test_profile_memory_async_path_inlines_content_under_blockbuster(
-    tmp_path, monkeypatch, run_async
+async def test_profile_memory_async_path_inlines_content_under_blockbuster(
+    tmp_path, monkeypatch
 ):
     memories = tmp_path / "memories"
     workspace = tmp_path / "workspace"
@@ -425,17 +423,13 @@ def test_profile_memory_async_path_inlines_content_under_blockbuster(
 
     monkeypatch.setattr(middleware, "_read_profile_memory", tracked_read_profile_memory)
 
-    async def run():
-        event_loop_thread = threading.get_ident()
-        blocker = BlockBuster(scanned_modules=memory_module)
-        blocker.activate()
-        try:
-            modified = await middleware.amodify_request(_request())
-        finally:
-            blocker.deactivate()
-        return event_loop_thread, modified
-
-    event_loop_thread, modified = run_async(run())
+    event_loop_thread = threading.get_ident()
+    blocker = BlockBuster(scanned_modules=memory_module)
+    blocker.activate()
+    try:
+        modified = await middleware.amodify_request(_request())
+    finally:
+        blocker.deactivate()
 
     assert call_threads
     assert all(thread_id != event_loop_thread for thread_id in call_threads)
@@ -534,8 +528,8 @@ def test_profile_memory_uses_explicit_workspace_for_project_profile(
     ).exists()
 
 
-def test_profile_memory_resolves_project_id_once_per_middleware(
-    tmp_path, monkeypatch, run_async
+async def test_profile_memory_resolves_project_id_once_per_middleware(
+    tmp_path, monkeypatch
 ):
     memories = tmp_path / "memories"
     workspace = tmp_path / "workspace"
@@ -552,7 +546,7 @@ def test_profile_memory_resolves_project_id_once_per_middleware(
         str(memories), workspace_dir=str(workspace), max_inline_profile_chars=10
     )
     middleware.modify_request(_request())
-    run_async(middleware.amodify_request(_request()))
+    await middleware.amodify_request(_request())
 
     assert calls == [workspace]
     assert middleware.project_id == "P-cached-project"

tests/test_qq_channel.py

@@ -8,7 +8,6 @@ from EvoScientist.channels.qq.channel import (
     QQConfig,
     _build_qq_keyboard,
 )
-from tests.conftest import run_async as _run
 
 
 class TestQQChannelSend:
@@ -22,7 +21,7 @@ class TestQQChannelSend:
         channel._client.api.post_group_message = AsyncMock()
         return channel
 
-    def test_send_prefers_native_markdown_for_c2c(self):
+    async def test_send_prefers_native_markdown_for_c2c(self):
         channel = self._make_ready_channel()
         msg = OutboundMessage(
             channel="qq",
@@ -35,7 +34,7 @@ class TestQQChannelSend:
             },
         )
 
-        assert _run(channel.send(msg)) is True
+        assert await channel.send(msg) is True
 
         channel._client.api.post_c2c_message.assert_awaited_once()
         sent = channel._client.api.post_c2c_message.await_args.kwargs
@@ -46,7 +45,7 @@ class TestQQChannelSend:
         assert sent["msg_seq"] == 1
         assert "content" not in sent
 
-    def test_send_falls_back_to_plain_text_when_markdown_send_fails(self):
+    async def test_send_falls_back_to_plain_text_when_markdown_send_fails(self):
         channel = self._make_ready_channel()
         channel._trace_event = MagicMock(side_effect=RuntimeError("trace failed"))
         channel._client.api.post_c2c_message = AsyncMock(
@@ -63,7 +62,7 @@ class TestQQChannelSend:
             },
         )
 
-        assert _run(channel.send(msg)) is True
+        assert await channel.send(msg) is True
 
         assert channel._client.api.post_c2c_message.await_count == 2
         first = channel._client.api.post_c2c_message.await_args_list[0].kwargs
@@ -80,7 +79,7 @@ class TestQQChannelSend:
         # trigger "duplicate msg_seq".
         assert second["msg_seq"] == 2
 
-    def test_send_does_not_fallback_on_transport_error(self):
+    async def test_send_does_not_fallback_on_transport_error(self):
         channel = self._make_ready_channel()
 
         async def _send_once(coro_factory, max_retries=3):
@@ -101,13 +100,13 @@ class TestQQChannelSend:
             },
         )
 
-        assert _run(channel.send(msg)) is False
+        assert await channel.send(msg) is False
         channel._client.api.post_c2c_message.assert_awaited_once()
         sent = channel._client.api.post_c2c_message.await_args.kwargs
         assert sent["msg_type"] == 2
         assert "content" not in sent
 
-    def test_send_does_not_fallback_when_transport_error_mentions_markdown(self):
+    async def test_send_does_not_fallback_when_transport_error_mentions_markdown(self):
         """A transport-layer error whose message incidentally contains the word
         "markdown" must NOT be reclassified as a markdown compatibility failure,
         otherwise genuine send failures get silently swallowed as plain-text."""
@@ -133,10 +132,10 @@ class TestQQChannelSend:
             },
         )
 
-        assert _run(channel.send(msg)) is False
+        assert await channel.send(msg) is False
         channel._client.api.post_c2c_message.assert_awaited_once()
 
-    def test_send_falls_back_on_qq_server_error_code(self):
+    async def test_send_falls_back_on_qq_server_error_code(self):
         """QQ server-side markdown errors (e.g. 304014 template not configured)
         should trigger plain-text fallback with a fresh msg_seq."""
         channel = self._make_ready_channel()
@@ -159,7 +158,7 @@ class TestQQChannelSend:
             },
         )
 
-        assert _run(channel.send(msg)) is True
+        assert await channel.send(msg) is True
 
         assert channel._client.api.post_c2c_message.await_count == 2
         first = channel._client.api.post_c2c_message.await_args_list[0].kwargs
@@ -231,7 +230,7 @@ class TestQQSendWithButtons:
         channel._client.api.post_group_message = AsyncMock()
         return channel
 
-    def test_c2c_send_attaches_keyboard(self):
+    async def test_c2c_send_attaches_keyboard(self):
         channel = self._make_channel()
         msg = OutboundMessage(
             channel="qq",
@@ -247,7 +246,7 @@ class TestQQSendWithButtons:
                 ],
             },
         )
-        assert _run(channel.send(msg)) is True
+        assert await channel.send(msg) is True
 
         sent = channel._client.api.post_c2c_message.await_args.kwargs
         assert sent["msg_type"] == 2
@@ -256,7 +255,7 @@ class TestQQSendWithButtons:
         assert rows[0]["buttons"][0]["action"]["data"] == "1"
         assert rows[1]["buttons"][0]["action"]["data"] == "2"
 
-    def test_group_send_does_not_attach_keyboard(self):
+    async def test_group_send_does_not_attach_keyboard(self):
         """Group keyboards are out of scope — silently dropped."""
         channel = self._make_channel()
         msg = OutboundMessage(
@@ -270,11 +269,11 @@ class TestQQSendWithButtons:
                 "buttons": [{"text": "Approve", "value": "1"}],
             },
         )
-        assert _run(channel.send(msg)) is True
+        assert await channel.send(msg) is True
         sent = channel._client.api.post_group_message.await_args.kwargs
         assert "keyboard" not in sent
 
-    def test_fallback_appends_button_hint_when_keyboard_present(self):
+    async def test_fallback_appends_button_hint_when_keyboard_present(self):
         """If markdown send fails and we fall back to plain text, the
         keyboard is lost — append a textual hint so the user still has
         a way to reply (the values still pass `_parse_approval_reply`).
@@ -302,7 +301,7 @@ class TestQQSendWithButtons:
                 ],
             },
         )
-        assert _run(channel.send(msg)) is True
+        assert await channel.send(msg) is True
 
         plain_call = channel._client.api.post_c2c_message.await_args_list[1].kwargs
         assert plain_call["msg_type"] == 0
@@ -311,7 +310,7 @@ class TestQQSendWithButtons:
         assert "1=Approve" in plain_call["content"]
         assert "2=Reject" in plain_call["content"]
 
-    def test_fallback_hint_handles_non_string_button_value(self):
+    async def test_fallback_hint_handles_non_string_button_value(self):
         """Regression: integer/None button values must not crash the
         plain-text fallback (the keyboard builder already coerces them)."""
         channel = self._make_channel()
@@ -335,7 +334,7 @@ class TestQQSendWithButtons:
                 ],
             },
         )
-        assert _run(channel.send(msg)) is True
+        assert await channel.send(msg) is True
         plain_call = channel._client.api.post_c2c_message.await_args_list[1].kwargs
         assert "42=OK" in plain_call["content"]
         assert "Cancel=Cancel" in plain_call["content"]
@@ -392,9 +391,9 @@ class TestQQInteractionCallback:
         )
         return interaction
 
-    def test_click_publishes_to_bus_with_button_data(self):
+    async def test_click_publishes_to_bus_with_button_data(self):
         channel = self._make_channel()
-        _run(channel._on_interaction(self._make_interaction("1")))
+        await channel._on_interaction(self._make_interaction("1"))
 
         channel._bus.publish_inbound.assert_awaited_once()
         inbound = channel._bus.publish_inbound.await_args[0][0]
@@ -405,63 +404,61 @@ class TestQQInteractionCallback:
         assert inbound.metadata["button_value"] == "1"
         assert inbound.metadata["msg_type"] == "c2c"
 
-    def test_click_acks_interaction(self):
+    async def test_click_acks_interaction(self):
         channel = self._make_channel()
-        _run(channel._on_interaction(self._make_interaction("1")))
+        await channel._on_interaction(self._make_interaction("1"))
         channel._client.api.on_interaction_result.assert_awaited_once_with("intr_1", 0)
 
-    def test_click_bypasses_debounce(self):
+    async def test_click_bypasses_debounce(self):
         """Click never hits queue_message (debounce buffer)."""
         channel = self._make_channel()
         channel.queue_message = AsyncMock()
-        _run(channel._on_interaction(self._make_interaction("3")))
+        await channel._on_interaction(self._make_interaction("3"))
         channel.queue_message.assert_not_called()
         channel._bus.publish_inbound.assert_awaited_once()
 
-    def test_group_interaction_ignored(self):
+    async def test_group_interaction_ignored(self):
         """No user_openid → group/guild click → don't publish."""
         channel = self._make_channel()
         intr = self._make_interaction(user_openid="")
         intr.group_openid = "group_xxx"
-        _run(channel._on_interaction(intr))
+        await channel._on_interaction(intr)
         channel._bus.publish_inbound.assert_not_called()
         # ACK still fires — it runs first, before the group-skip return.
         channel._client.api.on_interaction_result.assert_awaited_once_with("intr_1", 0)
 
-    def test_click_dropped_when_middleware_rejects(self):
+    async def test_click_dropped_when_middleware_rejects(self):
         channel = self._make_channel()
         channel._build_inbound_async = AsyncMock(return_value=None)
-        _run(channel._on_interaction(self._make_interaction("1")))
+        await channel._on_interaction(self._make_interaction("1"))
         channel._bus.publish_inbound.assert_not_called()
         # ACK still fires (we don't want the user staring at a stuck button)
         channel._client.api.on_interaction_result.assert_awaited_once()
 
-    def test_empty_button_data_falls_back_to_button_id(self):
+    async def test_empty_button_data_falls_back_to_button_id(self):
         channel = self._make_channel()
-        _run(
-            channel._on_interaction(
-                self._make_interaction(button_data="", button_id="btn_3")
-            )
+        await channel._on_interaction(
+            self._make_interaction(button_data="", button_id="btn_3")
         )
         inbound = channel._bus.publish_inbound.await_args[0][0]
         assert inbound.content == "btn_3"
 
-    def test_ack_fires_even_when_handler_throws(self):
+    async def test_ack_fires_even_when_handler_throws(self):
         """ACK must run before downstream processing so the QQ button UI
         stays responsive even if middleware/bus crashes."""
         channel = self._make_channel()
         channel._build_inbound_async = AsyncMock(side_effect=RuntimeError("boom"))
         # Should not raise — handler swallows downstream errors.
-        _run(channel._on_interaction(self._make_interaction("1")))
+        await channel._on_interaction(self._make_interaction("1"))
         channel._client.api.on_interaction_result.assert_awaited_once_with("intr_1", 0)
 
-    def test_button_value_metadata_is_string_coerced(self):
+    async def test_button_value_metadata_is_string_coerced(self):
         """Regression: metadata['button_value'] must be a string (was raw)."""
         channel = self._make_channel()
         resolved = MagicMock(button_id="btn_0", button_data=42, message_id="msg_orig")
         data = MagicMock(type=None, resolved=resolved)
         intr = MagicMock(id="intr_1", user_openid="u_x", group_openid=None, data=data)
-        _run(channel._on_interaction(intr))
+        await channel._on_interaction(intr)
         inbound = channel._bus.publish_inbound.await_args[0][0]
         assert inbound.content == "42"
         assert inbound.metadata["button_value"] == "42"

tests/test_repetitive_tool_guard.py

@@ -0,0 +1,177 @@
+"""Deterministic tool-loop guard and provider projection tests."""
+
+from dataclasses import dataclass, replace
+from typing import Any
+
+import pytest
+from langchain.agents.middleware.types import ModelResponse
+from langchain_core.messages import AIMessage, HumanMessage, ToolMessage
+
+from EvoScientist.llm.errors import AgentControlError
+from EvoScientist.middleware.repetitive_tool_guard import (
+    RepetitiveToolCallGuardMiddleware,
+    collapse_repetitive_tool_rounds,
+)
+
+
+def _round(
+    call_id: str,
+    *,
+    name: str = "execute",
+    command: str = "pwd",
+    content: str = "Error: invalid argument: command rejected by schema",
+    status: str = "error",
+) -> list[Any]:
+    return [
+        AIMessage(
+            content="",
+            tool_calls=[{"id": call_id, "name": name, "args": {"command": command}}],
+        ),
+        ToolMessage(
+            content=content,
+            tool_call_id=call_id,
+            name=name,
+            status=status,
+        ),
+    ]
+
+
+@dataclass(frozen=True)
+class _Request:
+    messages: list[Any]
+    tools: list[Any]
+
+    def override(self, **updates: Any):
+        return replace(self, **updates)
+
+
+def test_provider_projection_keeps_first_and_last_deterministic_error_rounds():
+    messages = [HumanMessage(content="inspect")]
+    for index in range(4):
+        messages.extend(_round(f"call-{index}"))
+    messages.append(HumanMessage(content="continue"))
+
+    repair = collapse_repetitive_tool_rounds(messages, threshold=2)
+
+    assert repair.removed_rounds == 2
+    assert [m.type for m in repair.messages] == [
+        "human",
+        "ai",
+        "tool",
+        "ai",
+        "tool",
+        "human",
+    ]
+    assert repair.messages[1].tool_calls[0]["id"] == "call-0"
+    assert repair.messages[3].tool_calls[0]["id"] == "call-3"
+
+
+def test_successful_repeated_calls_are_never_projected_away():
+    messages = [
+        *_round("call-1", content="ok", status="success"),
+        *_round("call-2", content="ok", status="success"),
+        *_round("call-3", content="ok", status="success"),
+    ]
+    repair = collapse_repetitive_tool_rounds(messages)
+    assert repair.messages == messages
+    assert repair.removed_rounds == 0
+    assert repair.tail_repetitions == 0
+
+
+def test_transient_and_unknown_errors_do_not_count_as_semantic_loop():
+    transient = [
+        *_round("call-1", content="Error: connection timeout"),
+        *_round("call-2", content="Error: connection timeout"),
+    ]
+    unknown = [
+        *_round("call-3", content="Error: something unusual"),
+        *_round("call-4", content="Error: something unusual"),
+    ]
+    assert collapse_repetitive_tool_rounds(transient).tail_repetitions == 0
+    assert collapse_repetitive_tool_rounds(unknown).tail_consecutive_errors == 0
+
+
+def test_generic_raw_execution_error_code_remains_unknown():
+    messages = _round("call-1", content="Error: something unusual")
+    messages[1].additional_kwargs["error_code"] = "TOOL_EXECUTION_FAILED"
+
+    repair = collapse_repetitive_tool_rounds(messages)
+
+    assert repair.tail_consecutive_errors == 0
+
+
+def test_identical_tail_loop_stops_before_next_model_call():
+    request = _Request(
+        messages=[*_round("call-1"), *_round("call-2")],
+        tools=[{"name": "execute"}],
+    )
+    called = False
+
+    def handler(_request):
+        nonlocal called
+        called = True
+        return ModelResponse(result=[AIMessage(content="should not run")])
+
+    with pytest.raises(AgentControlError) as caught:
+        RepetitiveToolCallGuardMiddleware(threshold=2).wrap_model_call(request, handler)
+
+    assert caught.value.code == "MODEL_TOOL_LOOP_DETECTED"
+    assert called is False
+
+
+def test_different_deterministic_errors_hit_consecutive_limit():
+    request = _Request(
+        messages=[
+            *_round("one", name="execute"),
+            *_round("two", name="read_file"),
+            *_round("three", name="search"),
+        ],
+        tools=[],
+    )
+
+    with pytest.raises(AgentControlError) as caught:
+        RepetitiveToolCallGuardMiddleware(
+            threshold=0, max_consecutive_errors=3
+        ).wrap_model_call(request, lambda _request: None)
+
+    assert caught.value.code == "MODEL_TOOL_ERROR_LIMIT"
+
+
+def test_user_message_breaks_tail_loop_but_historical_projection_is_temporary():
+    original = [
+        *_round("call-1"),
+        *_round("call-2"),
+        *_round("call-3"),
+        HumanMessage(content="try a new approach"),
+    ]
+    request = _Request(messages=original, tools=[])
+    captured = []
+
+    def handler(prepared):
+        captured.append(prepared)
+        return ModelResponse(result=[AIMessage(content="continued")])
+
+    RepetitiveToolCallGuardMiddleware().wrap_model_call(request, handler)
+    assert len(captured[0].messages) == 5
+    assert len(original) == 7
+
+
+def test_zero_thresholds_disable_only_semantic_loop_guards():
+    request = _Request(messages=[*_round("one"), *_round("two")], tools=[])
+    captured = []
+    middleware = RepetitiveToolCallGuardMiddleware(
+        threshold=0, max_consecutive_errors=0
+    )
+    middleware.wrap_model_call(
+        request,
+        lambda prepared: (
+            captured.append(prepared) or ModelResponse(result=[AIMessage(content="ok")])
+        ),
+    )
+    assert captured == [request]
+
+
+@pytest.mark.parametrize("kwargs", [{"threshold": -1}, {"max_consecutive_errors": -1}])
+def test_negative_threshold_is_rejected(kwargs):
+    with pytest.raises(ValueError, match="non-negative"):
+        RepetitiveToolCallGuardMiddleware(**kwargs)

tests/test_resume_command.py

@@ -2,7 +2,6 @@
 
 from unittest.mock import AsyncMock, MagicMock
 
-from tests.conftest import run_async as _run
 from tests.fakes import FakeGraphGateway, FakeThreadStore
 
 
@@ -24,7 +23,7 @@ def _ctx(thread_id="current", workspace_dir="/ws", thread_store=None):
 
 
 class TestResumeCommand:
-    def test_with_arg_resolves_and_calls_ui(self):
+    async def test_with_arg_resolves_and_calls_ui(self):
         from EvoScientist.commands.implementation.session import ResumeCommand
 
         ctx, ui = _ctx(
@@ -33,23 +32,23 @@ class TestResumeCommand:
                 metadata={"workspace_dir": "/restored"},
             )
         )
-        _run(ResumeCommand().execute(ctx, ["target-tid"]))
+        await ResumeCommand().execute(ctx, ["target-tid"])
         ui.handle_session_resume.assert_awaited_once_with("target-tid", "/restored")
         # ctx mutations
         assert ctx.thread_id == "target-tid"
         assert ctx.workspace_dir == "/restored"
 
-    def test_no_arg_empty_threads_prints_message(self):
+    async def test_no_arg_empty_threads_prints_message(self):
         from EvoScientist.commands.implementation.session import ResumeCommand
 
         ctx, ui = _ctx()
-        _run(ResumeCommand().execute(ctx, []))
+        await ResumeCommand().execute(ctx, [])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("No sessions to resume" in m for m in msgs)
         ui.wait_for_thread_pick.assert_not_called()
         ui.handle_session_resume.assert_not_called()
 
-    def test_no_arg_calls_picker(self):
+    async def test_no_arg_calls_picker(self):
         from EvoScientist.commands.implementation.session import ResumeCommand
 
         ctx, ui = _ctx()
@@ -60,11 +59,11 @@ class TestResumeCommand:
             resolved_thread_id="picked-tid",
         )
         ctx.graph_gateway = FakeGraphGateway(thread_store=store)
-        _run(ResumeCommand().execute(ctx, []))
+        await ResumeCommand().execute(ctx, [])
         ui.wait_for_thread_pick.assert_awaited_once()
         ui.handle_session_resume.assert_awaited_once()
 
-    def test_picker_cancel_returns(self):
+    async def test_picker_cancel_returns(self):
         from EvoScientist.commands.implementation.session import ResumeCommand
 
         ctx, ui = _ctx()
@@ -72,28 +71,28 @@ class TestResumeCommand:
         threads = [{"thread_id": "t1", "preview": "", "message_count": 0}]
         store = FakeThreadStore(threads=threads)
         ctx.graph_gateway = FakeGraphGateway(thread_store=store)
-        _run(ResumeCommand().execute(ctx, []))
+        await ResumeCommand().execute(ctx, [])
         ui.handle_session_resume.assert_not_called()
 
-    def test_ambiguous_prefix(self):
+    async def test_ambiguous_prefix(self):
         from EvoScientist.commands.implementation.session import ResumeCommand
 
         ctx, ui = _ctx(thread_store=FakeThreadStore(matches=["abc-one", "abc-two"]))
-        _run(ResumeCommand().execute(ctx, ["abc"]))
+        await ResumeCommand().execute(ctx, ["abc"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("Ambiguous" in m for m in msgs)
         ui.handle_session_resume.assert_not_called()
 
-    def test_not_found(self):
+    async def test_not_found(self):
         from EvoScientist.commands.implementation.session import ResumeCommand
 
         ctx, ui = _ctx()
-        _run(ResumeCommand().execute(ctx, ["missing"]))
+        await ResumeCommand().execute(ctx, ["missing"])
         msgs = [c.args[0] for c in ui.append_system.call_args_list]
         assert any("not found" in m for m in msgs)
         ui.handle_session_resume.assert_not_called()
 
-    def test_prefix_resolves_to_unique_match(self):
+    async def test_prefix_resolves_to_unique_match(self):
         from EvoScientist.commands.implementation.session import ResumeCommand
 
         ctx, ui = _ctx(
@@ -102,18 +101,18 @@ class TestResumeCommand:
                 metadata={"workspace_dir": "/ws1"},
             )
         )
-        _run(ResumeCommand().execute(ctx, ["abc"]))
+        await ResumeCommand().execute(ctx, ["abc"])
         ui.handle_session_resume.assert_awaited_once_with("abc-one", "/ws1")
         assert ctx.thread_id == "abc-one"
 
-    def test_empty_workspace_metadata_preserves_ctx_workspace(self):
+    async def test_empty_workspace_metadata_preserves_ctx_workspace(self):
         from EvoScientist.commands.implementation.session import ResumeCommand
 
         ctx, ui = _ctx(
             workspace_dir="/keep",
             thread_store=FakeThreadStore(resolved_thread_id="tid", metadata={}),
         )
-        _run(ResumeCommand().execute(ctx, ["tid"]))
+        await ResumeCommand().execute(ctx, ["tid"])
         # ResumeCommand only overwrites ctx.workspace_dir if metadata has one
         assert ctx.workspace_dir == "/keep"
         # Callback still fires with the metadata value (empty string)

tests/test_rich_command_ui.py

@@ -5,8 +5,6 @@ from unittest.mock import MagicMock
 from rich.console import Console
 from rich.table import Table
 
-from tests.conftest import run_async as _run
-
 
 def _make_ui(**kwargs):
     """Build a RichCLICommandUI backed by a MagicMock console."""
@@ -40,9 +38,9 @@ class TestBasicIO:
         ui.mount_renderable(table)
         console.print.assert_called_once_with(table)
 
-    def test_flush_is_async_noop(self):
+    async def test_flush_is_async_noop(self):
         ui, console = _make_ui()
-        _run(ui.flush())
+        await ui.flush()
         # flush should not print anything
         console.print.assert_not_called()
 
@@ -50,33 +48,29 @@ class TestBasicIO:
 class TestWaitForModelPick:
     """CLI model picker fallback: print table + return None."""
 
-    def test_returns_none(self):
+    async def test_returns_none(self):
         ui, _ = _make_ui()
         entries = [
             ("claude-sonnet-4-6", "anthropic/claude-sonnet", "anthropic"),
             ("gpt-4o", "openai/gpt-4o", "openai"),
         ]
-        result = _run(
-            ui.wait_for_model_pick(
-                entries,
-                current_model="claude-sonnet-4-6",
-                current_provider="anthropic",
-            )
+        result = await ui.wait_for_model_pick(
+            entries,
+            current_model="claude-sonnet-4-6",
+            current_provider="anthropic",
         )
         assert result is None
 
-    def test_prints_table_with_current_model_marker(self):
+    async def test_prints_table_with_current_model_marker(self):
         ui, console = _make_ui()
         entries = [
             ("claude-sonnet-4-6", "anthropic/claude-sonnet", "anthropic"),
             ("gpt-4o", "openai/gpt-4o", "openai"),
         ]
-        _run(
-            ui.wait_for_model_pick(
-                entries,
-                current_model="claude-sonnet-4-6",
-                current_provider="anthropic",
-            )
+        await ui.wait_for_model_pick(
+            entries,
+            current_model="claude-sonnet-4-6",
+            current_provider="anthropic",
         )
         # First call renders the Table (Rich renderable), second prints usage.
         assert console.print.call_count == 2
@@ -87,27 +81,23 @@ class TestWaitForModelPick:
         assert "Usage: /model" in usage_arg
         assert "--save" in usage_arg
 
-    def test_no_current_model_no_marker(self):
+    async def test_no_current_model_no_marker(self):
         ui, console = _make_ui()
         entries = [("claude-sonnet-4-6", "anthropic/claude-sonnet", "anthropic")]
-        _run(
-            ui.wait_for_model_pick(
-                entries,
-                current_model=None,
-                current_provider=None,
-            )
+        await ui.wait_for_model_pick(
+            entries,
+            current_model=None,
+            current_provider=None,
         )
         # Just asserts the coroutine runs without marker-branch issues.
         assert console.print.call_count == 2
 
-    def test_empty_entries_still_prints_header_and_usage(self):
+    async def test_empty_entries_still_prints_header_and_usage(self):
         ui, console = _make_ui()
-        result = _run(
-            ui.wait_for_model_pick(
-                [],
-                current_model=None,
-                current_provider=None,
-            )
+        result = await ui.wait_for_model_pick(
+            [],
+            current_model=None,
+            current_provider=None,
         )
         assert result is None
         # Header table + usage hint should still be printed even with
@@ -213,7 +203,7 @@ class TestWaitForThreadPick:
             },
         ]
 
-    def test_returns_selected_thread_id(self, monkeypatch):
+    async def test_returns_selected_thread_id(self, monkeypatch):
         import EvoScientist.cli.rich_command_ui as mod
 
         ui, _ = _make_ui()
@@ -226,7 +216,7 @@ class TestWaitForThreadPick:
             return prompt
 
         monkeypatch.setattr("questionary.select", fake_select)
-        result = _run(ui.wait_for_thread_pick(self._threads(), "abc123", "pick:"))
+        result = await ui.wait_for_thread_pick(self._threads(), "abc123", "pick:")
         assert result == "abc123"
         assert called["title"] == "pick:"
         # _build_items prepends a workspace header — choices has headers +
@@ -235,14 +225,14 @@ class TestWaitForThreadPick:
         # Table import removed; this test no longer depends on console output.
         assert mod.RichCLICommandUI is not None  # sanity
 
-    def test_cancel_returns_none(self, monkeypatch):
+    async def test_cancel_returns_none(self, monkeypatch):
         ui, _ = _make_ui()
         prompt = self._fake_prompt(None)
         monkeypatch.setattr("questionary.select", lambda *a, **k: prompt)
-        result = _run(ui.wait_for_thread_pick(self._threads(), "abc123", "pick:"))
+        result = await ui.wait_for_thread_pick(self._threads(), "abc123", "pick:")
         assert result is None
 
-    def test_current_thread_marker_in_label(self, monkeypatch):
+    async def test_current_thread_marker_in_label(self, monkeypatch):
         ui, _ = _make_ui()
         prompt = self._fake_prompt(None)
         captured_choices: list = []
@@ -252,7 +242,7 @@ class TestWaitForThreadPick:
             return prompt
 
         monkeypatch.setattr("questionary.select", fake_select)
-        _run(ui.wait_for_thread_pick(self._threads(), "abc123", "pick:"))
+        await ui.wait_for_thread_pick(self._threads(), "abc123", "pick:")
         # At least one Choice title contains "abc123 *" (current marker)
         choice_titles = [getattr(c, "title", "") for c in captured_choices]
         assert any("abc123 *" in t for t in choice_titles)
@@ -282,38 +272,38 @@ class TestCompactIndicator:
 class TestPhaseBMigrated:
     """Session lifecycle callbacks (start/resume) filled in Phase B."""
 
-    def test_start_new_session_fires_callback(self):
+    async def test_start_new_session_fires_callback(self):
         from unittest.mock import AsyncMock
 
         cb = AsyncMock()
         ui, _ = _make_ui(on_start_new_session=cb)
-        _run(ui.start_new_session())
+        await ui.start_new_session()
         cb.assert_awaited_once()
 
-    def test_start_new_session_without_callback_is_noop(self):
+    async def test_start_new_session_without_callback_is_noop(self):
         ui, console = _make_ui()
-        _run(ui.start_new_session())
+        await ui.start_new_session()
         console.print.assert_not_called()
 
-    def test_handle_session_resume_awaits_callback(self):
+    async def test_handle_session_resume_awaits_callback(self):
         from unittest.mock import AsyncMock
 
         cb = AsyncMock()
         ui, _ = _make_ui(on_handle_session_resume=cb)
-        _run(ui.handle_session_resume("tid-x", "/workspace"))
+        await ui.handle_session_resume("tid-x", "/workspace")
         cb.assert_awaited_once_with("tid-x", "/workspace")
 
-    def test_handle_session_resume_without_callback_is_noop(self):
+    async def test_handle_session_resume_without_callback_is_noop(self):
         ui, _ = _make_ui()
         # Should not raise
-        _run(ui.handle_session_resume("tid-x"))
+        await ui.handle_session_resume("tid-x")
 
-    def test_handle_session_resume_workspace_defaults_none(self):
+    async def test_handle_session_resume_workspace_defaults_none(self):
         from unittest.mock import AsyncMock
 
         cb = AsyncMock()
         ui, _ = _make_ui(on_handle_session_resume=cb)
-        _run(ui.handle_session_resume("tid-x"))
+        await ui.handle_session_resume("tid-x")
         cb.assert_awaited_once_with("tid-x", None)
 
 
@@ -321,7 +311,7 @@ class TestPhaseCMigrated:
     """Skill/MCP browse pickers delegate to questionary helpers via
     ``asyncio.to_thread`` since questionary blocks the event loop."""
 
-    def test_skill_browse_delegates_to_picker(self, monkeypatch):
+    async def test_skill_browse_delegates_to_picker(self, monkeypatch):
         from unittest.mock import MagicMock
 
         picker = MagicMock(return_value=["skill-a", "skill-b"])
@@ -330,11 +320,11 @@ class TestPhaseCMigrated:
             picker,
         )
         ui, _ = _make_ui()
-        result = _run(ui.wait_for_skill_browse([{"name": "a"}], {"installed"}, "core"))
+        result = await ui.wait_for_skill_browse([{"name": "a"}], {"installed"}, "core")
         assert result == ["skill-a", "skill-b"]
         picker.assert_called_once_with([{"name": "a"}], {"installed"}, "core")
 
-    def test_skill_browse_cancel_returns_none(self, monkeypatch):
+    async def test_skill_browse_cancel_returns_none(self, monkeypatch):
         from unittest.mock import MagicMock
 
         monkeypatch.setattr(
@@ -342,10 +332,10 @@ class TestPhaseCMigrated:
             MagicMock(return_value=None),
         )
         ui, _ = _make_ui()
-        result = _run(ui.wait_for_skill_browse([], set(), ""))
+        result = await ui.wait_for_skill_browse([], set(), "")
         assert result is None
 
-    def test_mcp_browse_delegates_to_picker(self, monkeypatch):
+    async def test_mcp_browse_delegates_to_picker(self, monkeypatch):
         from unittest.mock import MagicMock
 
         sentinel_entries = [MagicMock(name="entry1"), MagicMock(name="entry2")]
@@ -355,11 +345,11 @@ class TestPhaseCMigrated:
             picker,
         )
         ui, _ = _make_ui()
-        result = _run(ui.wait_for_mcp_browse([MagicMock()], {"configured"}, ""))
+        result = await ui.wait_for_mcp_browse([MagicMock()], {"configured"}, "")
         assert result is sentinel_entries
         picker.assert_called_once()
 
-    def test_mcp_browse_cancel_returns_none(self, monkeypatch):
+    async def test_mcp_browse_cancel_returns_none(self, monkeypatch):
         from unittest.mock import MagicMock
 
         monkeypatch.setattr(
@@ -367,5 +357,5 @@ class TestPhaseCMigrated:
             MagicMock(return_value=None),
         )
         ui, _ = _make_ui()
-        result = _run(ui.wait_for_mcp_browse([], set(), ""))
+        result = await ui.wait_for_mcp_browse([], set(), "")
         assert result is None

tests/test_runtime_integrations.py

@@ -0,0 +1,117 @@
+from __future__ import annotations
+
+import ast
+from pathlib import Path
+
+import pytest
+
+from EvoScientist.runtime_integrations import (
+    RuntimeIntegrationUnavailable,
+    configure_runtime_integrations,
+    get_app_connection,
+    get_image_backend,
+    get_session_connection,
+    get_session_dsn,
+    handle_knowledge_file,
+    record_service_usage,
+    reset_runtime_integrations,
+    resolve_runtime_model,
+)
+
+
+@pytest.fixture(autouse=True)
+def reset_integrations():
+    reset_runtime_integrations()
+    yield
+    reset_runtime_integrations()
+
+
+def test_core_package_does_not_import_gateway():
+    package_root = Path(__file__).resolve().parents[1] / "EvoScientist"
+    violations = []
+    for source_file in package_root.rglob("*.py"):
+        tree = ast.parse(
+            source_file.read_text(encoding="utf-8"), filename=str(source_file)
+        )
+        for node in ast.walk(tree):
+            if isinstance(node, ast.Import):
+                names = [alias.name for alias in node.names]
+            elif isinstance(node, ast.ImportFrom):
+                if node.level:
+                    continue
+                names = [node.module or ""]
+            else:
+                continue
+            if any(name == "gateway" or name.startswith("gateway.") for name in names):
+                violations.append(
+                    f"{source_file.relative_to(package_root)}:{node.lineno}"
+                )
+    assert violations == []
+
+
+@pytest.mark.anyio
+async def test_optional_integrations_are_safe_without_web_runtime(tmp_path):
+    assert get_session_dsn() is None
+    await handle_knowledge_file(tmp_path / "result.md")
+    await record_service_usage("search", "query")
+    with pytest.raises(RuntimeIntegrationUnavailable):
+        await get_app_connection()
+    with pytest.raises(RuntimeIntegrationUnavailable):
+        await get_session_connection()
+    with pytest.raises(RuntimeIntegrationUnavailable):
+        get_image_backend()
+
+
+@pytest.mark.anyio
+async def test_host_can_register_runtime_integrations(tmp_path):
+    app_connection = object()
+    session_connection = object()
+    knowledge_paths = []
+    usage = []
+    image_backend = object()
+
+    async def provide_app_connection():
+        return app_connection
+
+    async def provide_session_connection():
+        return session_connection
+
+    async def handle_file(path):
+        knowledge_paths.append(path)
+
+    async def record_usage(service, action):
+        usage.append((service, action))
+
+    configure_runtime_integrations(
+        app_connection_provider=provide_app_connection,
+        session_connection_provider=provide_session_connection,
+        session_dsn_provider=lambda: "postgresql://example/session",
+        knowledge_file_handler=handle_file,
+        usage_recorder=record_usage,
+        image_backend_factory=lambda: image_backend,
+    )
+
+    path = tmp_path / "result.md"
+    await handle_knowledge_file(path)
+    await record_service_usage("mineru", "parse")
+
+    assert await get_app_connection() is app_connection
+    assert await get_session_connection() is session_connection
+    assert get_session_dsn() == "postgresql://example/session"
+    assert get_image_backend() is image_backend
+    assert knowledge_paths == [path]
+    assert usage == [("mineru", "parse")]
+
+
+def test_host_can_register_model_resolver():
+    resolved = object()
+    calls = []
+
+    def resolve_model(model, provider):
+        calls.append((model, provider))
+        return resolved
+
+    configure_runtime_integrations(model_resolver=resolve_model)
+
+    assert resolve_runtime_model("model-a", "provider-a") is resolved
+    assert calls == [("model-a", "provider-a")]

tests/test_schedule_command.py

@@ -2,8 +2,6 @@
 
 from unittest.mock import MagicMock, patch
 
-from tests.conftest import run_async as _run
-
 
 def _ctx():
     from EvoScientist.commands.base import CommandContext
@@ -12,17 +10,17 @@ def _ctx():
     return CommandContext(agent=None, thread_id="tid", ui=ui), ui
 
 
-def test_list_when_backend_down():
+async def test_list_when_backend_down():
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
     ctx, ui = _ctx()
     with patch("EvoScientist.cron.schedule.is_available", return_value=False):
-        _run(ScheduleCommand().execute(ctx, ["list"]))
+        await ScheduleCommand().execute(ctx, ["list"])
     msgs = [c.args[0] for c in ui.append_system.call_args_list]
     assert any("unavailable" in m.lower() for m in msgs)
 
 
-def test_add_parses_five_field_cron_and_prompt():
+async def test_add_parses_five_field_cron_and_prompt():
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
     ctx, _ui = _ctx()
@@ -33,17 +31,15 @@ def test_add_parses_five_field_cron_and_prompt():
             return_value={"cron_id": "c-9"},
         ) as mk,
     ):
-        _run(
-            ScheduleCommand().execute(
-                ctx, ["add", "*/10", "*", "*", "*", "*", "search", "uk", "weather"]
-            )
+        await ScheduleCommand().execute(
+            ctx, ["add", "*/10", "*", "*", "*", "*", "search", "uk", "weather"]
         )
     kw = mk.call_args.kwargs
     assert kw["schedule"] == "*/10 * * * *"
     assert kw["prompt"] == "search uk weather"
 
 
-def test_list_renders_table():
+async def test_list_renders_table():
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
     ctx, ui = _ctx()
@@ -60,11 +56,11 @@ def test_list_renders_table():
         patch("EvoScientist.cron.schedule.is_available", return_value=True),
         patch("EvoScientist.cron.schedule.list_schedules", return_value=rows),
     ):
-        _run(ScheduleCommand().execute(ctx, ["list"]))
+        await ScheduleCommand().execute(ctx, ["list"])
     ui.mount_renderable.assert_called_once()
 
 
-def test_add_parses_quoted_cron_and_prompt():
+async def test_add_parses_quoted_cron_and_prompt():
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
     ctx, _ui = _ctx()
@@ -75,15 +71,15 @@ def test_add_parses_quoted_cron_and_prompt():
             return_value={"cron_id": "c-9"},
         ) as mk,
     ):
-        _run(
-            ScheduleCommand().execute(ctx, ["add", "*/10 * * * *", "search uk weather"])
+        await ScheduleCommand().execute(
+            ctx, ["add", "*/10 * * * *", "search uk weather"]
         )
     kw = mk.call_args.kwargs
     assert kw["schedule"] == "*/10 * * * *"
     assert kw["prompt"] == "search uk weather"
 
 
-def test_run_with_matching_prefix_fires_matched_prompt():
+async def test_run_with_matching_prefix_fires_matched_prompt():
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
     ctx, _ui = _ctx()
@@ -96,11 +92,11 @@ def test_run_with_matching_prefix_fires_matched_prompt():
             return_value={"run_id": "r-1"},
         ) as rn,
     ):
-        _run(ScheduleCommand().execute(ctx, ["run", "c-123"]))
+        await ScheduleCommand().execute(ctx, ["run", "c-123"])
     rn.assert_called_once_with("do the thing")
 
 
-def test_run_with_no_match_reports():
+async def test_run_with_no_match_reports():
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
     ctx, ui = _ctx()
@@ -109,13 +105,13 @@ def test_run_with_no_match_reports():
         patch("EvoScientist.cron.schedule.list_schedules", return_value=[]),
         patch("EvoScientist.cron.schedule.run_now") as rn,
     ):
-        _run(ScheduleCommand().execute(ctx, ["run", "nope"]))
+        await ScheduleCommand().execute(ctx, ["run", "nope"])
     rn.assert_not_called()
     msgs = [c.args[0] for c in ui.append_system.call_args_list]
     assert any("No schedule matching" in m for m in msgs)
 
 
-def test_pause_resume_set_enabled_with_resolved_id():
+async def test_pause_resume_set_enabled_with_resolved_id():
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
     rows = [{"cron_id": "c-abcdef", "metadata": {"name": "t"}}]
@@ -126,7 +122,7 @@ def test_pause_resume_set_enabled_with_resolved_id():
             patch("EvoScientist.cron.schedule.list_schedules", return_value=rows),
             patch("EvoScientist.cron.schedule.set_enabled") as se,
         ):
-            _run(ScheduleCommand().execute(ctx, [sub, "c-abc"]))
+            await ScheduleCommand().execute(ctx, [sub, "c-abc"])
         se.assert_called_once_with("c-abcdef", expected)
 
 
@@ -135,7 +131,7 @@ def test_pause_resume_set_enabled_with_resolved_id():
 # ---------------------------------------------------------------------------
 
 
-def test_list_error_shows_red_message_no_exception():
+async def test_list_error_shows_red_message_no_exception():
     """B1: list_schedules raising after is_available() shows a red error, not a traceback."""
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
@@ -147,7 +143,7 @@ def test_list_error_shows_red_message_no_exception():
             side_effect=RuntimeError("backend gone"),
         ),
     ):
-        _run(ScheduleCommand().execute(ctx, ["list"]))
+        await ScheduleCommand().execute(ctx, ["list"])
     msgs = [c.args[0] for c in ui.append_system.call_args_list]
     assert any("Error:" in m for m in msgs)
     # Verify no exception escaped (test would have raised above otherwise)
@@ -158,7 +154,7 @@ def test_list_error_shows_red_message_no_exception():
 # ---------------------------------------------------------------------------
 
 
-def test_remove_ambiguous_prefix_aborts_without_deleting():
+async def test_remove_ambiguous_prefix_aborts_without_deleting():
     """B2: two crons sharing a prefix → ambiguity message, delete NOT called."""
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
@@ -172,7 +168,7 @@ def test_remove_ambiguous_prefix_aborts_without_deleting():
         patch("EvoScientist.cron.schedule.list_schedules", return_value=rows),
         patch("EvoScientist.cron.schedule.delete_schedule") as mk,
     ):
-        _run(ScheduleCommand().execute(ctx, ["remove", "abc"]))
+        await ScheduleCommand().execute(ctx, ["remove", "abc"])
     mk.assert_not_called()
     msgs = [c.args[0] for c in ui.append_system.call_args_list]
     assert any("Multiple" in m for m in msgs)
@@ -183,7 +179,7 @@ def test_remove_ambiguous_prefix_aborts_without_deleting():
 # ---------------------------------------------------------------------------
 
 
-def test_remove_backend_error_shows_red_error_not_no_match():
+async def test_remove_backend_error_shows_red_error_not_no_match():
     """FIX 1: list_schedules() crashing in _resolve → red 'Error:' message, not 'No schedule matching'."""
     from EvoScientist.commands.implementation.schedule import ScheduleCommand
 
@@ -196,7 +192,7 @@ def test_remove_backend_error_shows_red_error_not_no_match():
         ),
         patch("EvoScientist.cron.schedule.delete_schedule") as mk,
     ):
-        _run(ScheduleCommand().execute(ctx, ["remove", "abc"]))
+        await ScheduleCommand().execute(ctx, ["remove", "abc"])
     mk.assert_not_called()
     msgs = [c.args[0] for c in ui.append_system.call_args_list]
     assert any("Error:" in m for m in msgs), f"Expected red Error: message, got: {msgs}"
@@ -209,7 +205,7 @@ def test_remove_backend_error_shows_red_error_not_no_match():
     )
 
 
-def test_add_name_sanitized_from_nasty_prompt():
+async def test_add_name_sanitized_from_nasty_prompt():
     """B3: prompt with newline / slashes / special chars → clean kebab-case name."""
     import re
 
@@ -225,7 +221,7 @@ def test_add_name_sanitized_from_nasty_prompt():
             return_value={"cron_id": "c-x"},
         ) as mk,
     ):
-        _run(ScheduleCommand().execute(ctx, ["add", "*/5 * * * *", nasty_prompt]))
+        await ScheduleCommand().execute(ctx, ["add", "*/5 * * * *", nasty_prompt])
     name = mk.call_args.kwargs["name"]
     # Must be non-empty, no spaces, no newlines, no slashes
     assert name

tests/test_serde_default_rich_exception.py

@@ -0,0 +1,242 @@
+"""Regression tests for the helpers ``ErrorNormalizationMiddleware``
+uses to build the SSE error envelope.
+
+- ``_redact_api_keys`` + ``_build_env_key_redaction_re`` — scrubs
+  deployed credentials that the SDK might echo back.
+- ``_extract_status_code`` / ``_extract_provider_code`` /
+  ``_extract_error_type`` — read SDK-specific fields off the raised
+  exception.
+
+Middleware wire behavior + ``_provider_from_model`` live in
+``test_error_normalization_middleware.py``. One end-to-end orjson test
+at the bottom guards that a ``ProviderStreamError`` survives
+langgraph_api's UNPATCHED ``serde.default`` under
+``OPT_SERIALIZE_DATACLASS`` — the whole reason the wrapper exists.
+"""
+
+from __future__ import annotations
+
+import os
+
+import langgraph_api.serde as _serde_mod
+
+from EvoScientist.llm.errors import (
+    _API_KEY_ENV_SUFFIXES,
+    _build_env_key_redaction_re,
+    _extract_error_type,
+    _extract_provider_code,
+    _extract_status_code,
+    _redact_api_keys,
+)
+
+# ---------------------------------------------------------------------------
+# Redaction
+# ---------------------------------------------------------------------------
+
+
+def test_env_deployed_key_redacted_in_message(monkeypatch):
+    """A credential exported via env var is scrubbed by
+    ``_redact_api_keys``. The redaction table is rebuilt per call —
+    ``monkeypatch.setenv`` alone is enough, no attribute reassignment.
+    """
+    key = "sk-proj-aBcDeFgHiJkLmNoPqRsTuVwXyZ1234567890"
+    monkeypatch.setenv("OPENAI_API_KEY", key)
+
+    msg = (
+        f"Invalid API key: {key}. Get a new one at https://platform.openai.com/api-keys"
+    )
+    redacted = _redact_api_keys(msg)
+    assert key not in redacted
+    assert "<redacted>" in redacted
+    assert "Invalid API key" in redacted
+    assert "platform.openai.com" in redacted
+
+
+def test_multiple_env_keys_redacted_independently(monkeypatch):
+    """Each ``*_API_KEY`` / ``*_TOKEN`` / ``*_SECRET`` env var
+    contributes its own prefix to the alternation.
+    """
+    k1 = "sk-or-aBcDeFg012345678901234"
+    k2 = "AIzaABCDEFGHIJ0123456789"
+    k3 = "ghp_p4t70k3n0123456789abcdef"
+    monkeypatch.setenv("OPENROUTER_API_KEY", k1)
+    monkeypatch.setenv("GOOGLE_API_KEY", k2)
+    monkeypatch.setenv("GITHUB_TOKEN", k3)
+
+    msg = _redact_api_keys(f"Failures: {k1}, {k2}, {k3}")
+    assert k1 not in msg
+    assert k2 not in msg
+    assert k3 not in msg
+    assert msg.count("<redacted>") == 3
+
+
+def test_base64_suffix_secret_fully_redacted(monkeypatch):
+    """A base64-style secret (``/`` ``+`` ``=``) must redact end-to-end,
+    not leak its tail past the first padding char.
+    """
+    key = "AbCdEfGh/secret+tail=="
+    monkeypatch.setenv("SOME_SECRET", key)
+
+    msg = _redact_api_keys(f"auth failed with token={key} on retry")
+    assert "secret" not in msg
+    assert "tail" not in msg
+    assert "<redacted>" in msg
+    assert "auth failed" in msg
+    assert "on retry" in msg
+
+
+def test_unknown_shape_not_redacted_without_env(monkeypatch):
+    """Env-only redaction: a key-shaped string not deployed via env is
+    left alone. Tradeoff — we only scrub what we know is a secret.
+    """
+    for k in list(os.environ):
+        if k.endswith(_API_KEY_ENV_SUFFIXES):
+            monkeypatch.delenv(k, raising=False)
+
+    msg = _redact_api_keys("Unknown key seen: sk-or-aBcDeFg012345678901234")
+    assert "sk-or-aBcDeFg012345678901234" in msg
+    assert "<redacted>" not in msg
+
+
+def test_env_key_loaded_after_first_call_is_redacted(monkeypatch):
+    """The pattern rebuilds every call so keys loaded after
+    ``patches.py`` imports (typical ``load_dotenv`` sequence) are
+    still scrubbed on the next call.
+    """
+    for k in list(os.environ):
+        if k.endswith(_API_KEY_ENV_SUFFIXES):
+            monkeypatch.delenv(k, raising=False)
+
+    key = "sk-proj-loaded_after_import_1234567890abcdef"
+    # Pass 1: env empty — key leaks.
+    assert key in _redact_api_keys(f"leak: {key}")
+
+    # Pass 2: after simulated load_dotenv.
+    monkeypatch.setenv("OPENAI_API_KEY", key)
+    redacted = _redact_api_keys(f"leak: {key}")
+    assert key not in redacted
+    assert "<redacted>" in redacted
+
+
+def test_redaction_regex_holds_only_prefix(monkeypatch):
+    """Defense-in-depth: the compiled regex must not embed the full key.
+    A process-memory leak (traceback locals, debugger) exposes at most
+    the first 8 chars — not the secret.
+    """
+    key = "sk-proj-aBcDeFgHiJkLmNoPqRsTuVwXyZ1234567890_secret_suffix"
+    monkeypatch.setenv("OPENAI_API_KEY", key)
+    pattern = _build_env_key_redaction_re()
+    assert pattern is not None
+    assert key not in pattern.pattern
+    assert "aBcDeFgHiJkLmNoPqRs" not in pattern.pattern
+    # Sanity: still matches the full key at runtime via prefix + suffix
+    # greedy.
+    m = pattern.search(f"err: {key}")
+    assert m is not None
+    assert m.group(0) == key
+
+
+# ---------------------------------------------------------------------------
+# Field extractors
+# ---------------------------------------------------------------------------
+
+
+def _fake_exc(**attrs):
+    return type("APIError", (Exception,), attrs)("boom")
+
+
+def test_status_code_read_from_direct_attribute():
+    """openai / anthropic ``APIStatusError`` carries integer
+    ``.status_code`` — the primary path.
+    """
+    assert _extract_status_code(_fake_exc(status_code=429)) == 429
+
+
+def test_status_code_read_via_response_attribute():
+    """Wrappers that don't promote status to top level expose it via
+    ``.response.status_code`` (httpx pattern).
+    """
+
+    class FakeResponse:
+        status_code = 504
+
+    assert _extract_status_code(_fake_exc(response=FakeResponse())) == 504
+
+
+def test_status_code_read_via_integer_code_attribute():
+    """``google.genai.errors.APIError`` stores HTTP status as integer
+    ``.code`` — type-disambiguated from openai/anthropic's string
+    ``.code`` (provider error code).
+    """
+    assert _extract_status_code(_fake_exc(code=400)) == 400
+
+
+def test_provider_code_read_from_string_code_attribute():
+    """Provider error code (``insufficient_quota`` etc.) is a string
+    ``.code`` — higher signal than the integer HTTP status alone.
+    """
+    assert (
+        _extract_provider_code(_fake_exc(code="insufficient_quota"))
+        == "insufficient_quota"
+    )
+
+
+def test_provider_code_ignores_integer_code():
+    """An integer ``.code`` is HTTP status (see above); must not bleed
+    into the provider-code path.
+    """
+    assert _extract_provider_code(_fake_exc(code=429)) is None
+
+
+def test_error_type_read_from_type_attribute():
+    """openai exposes a ``.type`` label (``rate_limit_error``)."""
+    assert _extract_error_type(_fake_exc(type="rate_limit_error")) == "rate_limit_error"
+
+
+def test_extractors_return_none_when_attributes_absent():
+    """A bare exception with no SDK-shape attributes — every extractor
+    returns None so the envelope drops the optional fields.
+    """
+    exc = _fake_exc()
+    assert _extract_status_code(exc) is None
+    assert _extract_provider_code(exc) is None
+    assert _extract_error_type(exc) is None
+
+
+# ---------------------------------------------------------------------------
+# End-to-end: ProviderStreamError survives orjson under
+# OPT_SERIALIZE_DATACLASS via upstream's UNPATCHED serde.default.
+# ---------------------------------------------------------------------------
+
+
+def test_provider_stream_error_survives_orjson_dataclass_option():
+    """Guard: ``ProviderStreamError`` — a plain Exception subclass with
+    a ``model_dump()`` hook — must emerge as the envelope on the wire
+    even under ``OPT_SERIALIZE_DATACLASS``, using ONLY upstream's
+    stock ``serde.default``. Proof that we no longer need to patch
+    the serde module.
+    """
+    import orjson
+
+    from EvoScientist.llm.errors import ProviderStreamError
+
+    err = ProviderStreamError(
+        provider="openrouter",
+        class_qualname="openrouter.errors.foo.UnauthorizedResponseError",
+        message="User not found.",
+        status_code=401,
+    )
+
+    wire = orjson.dumps(
+        err,
+        default=_serde_mod.default,  # upstream, unpatched
+        option=orjson.OPT_SERIALIZE_DATACLASS,
+    )
+    decoded = orjson.loads(wire)
+    assert decoded == {
+        "error": "UnauthorizedResponseError",
+        "class": "openrouter.errors.foo.UnauthorizedResponseError",
+        "message": "User not found.",
+        "provider": "openrouter",
+        "status_code": 401,
+    }

tests/test_serve_agent_holder.py

@@ -27,7 +27,6 @@ from EvoScientist.cli.commands import (
 from EvoScientist.commands.base import ChannelRuntime
 from EvoScientist.config import EvoScientistConfig
 from EvoScientist.gateway import RuntimeGateways, ThreadStore
-from tests.conftest import run_async as _run
 from tests.fakes import FakeGraphGateway, FakeThreadStore
 
 
@@ -71,7 +70,7 @@ def _runtime_state(
     )
 
 
-def test_hook_updates_runtime_state_on_agent_swap():
+async def test_hook_updates_runtime_state_on_agent_swap():
     """``/model`` mutates ``ctx.agent`` to a new handle — the hook must
     push that handle into the shared runtime state so the outer poll loop sees
     it on the next message."""
@@ -86,12 +85,12 @@ def test_hook_updates_runtime_state_on_agent_swap():
     cmd = MagicMock()
     cmd.name = "/model"
 
-    _run(hook(ctx, original_agent, cmd))
+    await hook(ctx, original_agent, cmd)
 
     assert state.agent is new_agent
 
 
-def test_hook_syncs_channel_runtime():
+async def test_hook_syncs_channel_runtime():
     """Other readers (the bus) look at ``ChannelRuntime.agent``; the
     hook keeps the runtime in sync with the runtime state update."""
     original_agent = _agent("original-agent")
@@ -109,13 +108,13 @@ def test_hook_syncs_channel_runtime():
     cmd = MagicMock()
     cmd.name = "/model"
 
-    _run(hook(ctx, original_agent, cmd))
+    await hook(ctx, original_agent, cmd)
 
     assert runtime.agent is new_agent
     assert runtime.thread_id == "t"
 
 
-def test_hook_noop_when_agent_unchanged():
+async def test_hook_noop_when_agent_unchanged():
     """Commands like ``/evoskills`` don't touch ``ctx.agent`` — the
     runtime state must stay put."""
     original_agent = _agent("original-agent")
@@ -128,12 +127,12 @@ def test_hook_noop_when_agent_unchanged():
     cmd = MagicMock()
     cmd.name = "/evoskills"
 
-    _run(hook(ctx, original_agent, cmd))
+    await hook(ctx, original_agent, cmd)
 
     assert state.agent is original_agent
 
 
-def test_hook_noop_when_ctx_agent_is_none():
+async def test_hook_noop_when_ctx_agent_is_none():
     """Guard against commands that reset ``ctx.agent`` to ``None`` —
     we never want to write ``None`` into runtime state."""
     original_agent = _agent("original-agent")
@@ -146,12 +145,12 @@ def test_hook_noop_when_ctx_agent_is_none():
     cmd = MagicMock()
     cmd.name = "/whatever"
 
-    _run(hook(ctx, original_agent, cmd))
+    await hook(ctx, original_agent, cmd)
 
     assert state.agent is original_agent
 
 
-def test_hook_updates_thread_id_on_resume():
+async def test_hook_updates_thread_id_on_resume():
     """``/resume`` mutates ``ctx.thread_id`` — the hook must push the
     new id into runtime state so the outer poll loop runs subsequent
     messages on the resumed thread."""
@@ -166,12 +165,12 @@ def test_hook_updates_thread_id_on_resume():
     cmd = MagicMock()
     cmd.name = "/resume"
 
-    _run(hook(ctx, agent, cmd))
+    await hook(ctx, agent, cmd)
 
     assert state.thread_id == "new-tid"
 
 
-def test_hook_updates_workspace_dir_on_resume():
+async def test_hook_updates_workspace_dir_on_resume():
     """`/resume` can restore a different workspace; serve must reload for it."""
     cfg = _config()
     old_agent = _agent("old-agent")
@@ -201,7 +200,7 @@ def test_hook_updates_workspace_dir_on_resume():
             return_value=reloaded_agent,
         ) as load_agent,
     ):
-        _run(hook(ctx, old_agent, cmd))
+        await hook(ctx, old_agent, cmd)
 
     sync_server.assert_awaited_once_with(cfg, workspace_dir="/restored-ws")
     load_agent.assert_called_once_with(workspace_dir="/restored-ws", config=cfg)
@@ -209,7 +208,7 @@ def test_hook_updates_workspace_dir_on_resume():
     assert state.agent is reloaded_agent
 
 
-def test_hook_syncs_channel_runtime_thread_id():
+async def test_hook_syncs_channel_runtime_thread_id():
     """The bus reads ``ChannelRuntime.thread_id``; hook must sync it
     alongside the runtime state update."""
     agent = _agent("a")
@@ -224,12 +223,12 @@ def test_hook_syncs_channel_runtime_thread_id():
     cmd = MagicMock()
     cmd.name = "/resume"
 
-    _run(hook(ctx, agent, cmd))
+    await hook(ctx, agent, cmd)
 
     assert runtime.thread_id == "new-tid"
 
 
-def test_hook_noop_when_thread_id_unchanged():
+async def test_hook_noop_when_thread_id_unchanged():
     """Most commands don't touch thread_id — runtime state stays put."""
     agent = _agent("a")
     state = _runtime_state(agent=agent, thread_id="same-tid")
@@ -241,12 +240,12 @@ def test_hook_noop_when_thread_id_unchanged():
     cmd = MagicMock()
     cmd.name = "/evoskills"
 
-    _run(hook(ctx, agent, cmd))
+    await hook(ctx, agent, cmd)
 
     assert state.thread_id == "same-tid"
 
 
-def test_hook_skips_resume_warning_when_thread_unchanged():
+async def test_hook_skips_resume_warning_when_thread_unchanged():
     """Bare ``/resume`` with no argument prints usage but leaves
     ``ctx.thread_id`` unchanged — the in-memory-state warning must NOT
     fire because no resume actually happened."""
@@ -261,13 +260,13 @@ def test_hook_skips_resume_warning_when_thread_unchanged():
     cmd = MagicMock()
     cmd.name = "/resume"
 
-    _run(hook(ctx, agent, cmd))
+    await hook(ctx, agent, cmd)
 
     ctx.ui.append_system.assert_not_called()
     ctx.ui.flush.assert_not_called()
 
 
-def test_hook_emits_resume_warning_when_thread_changed():
+async def test_hook_emits_resume_warning_when_thread_changed():
     """``/resume <tid>`` that actually changes thread_id must surface
     the in-memory-state warning via ``ctx.ui``."""
     agent = _agent("a")
@@ -283,7 +282,7 @@ def test_hook_emits_resume_warning_when_thread_changed():
     cmd = MagicMock()
     cmd.name = "/resume"
 
-    _run(hook(ctx, agent, cmd))
+    await hook(ctx, agent, cmd)
 
     ctx.ui.append_system.assert_called_once()
     warn_text, warn_kwargs = (
@@ -296,7 +295,7 @@ def test_hook_emits_resume_warning_when_thread_changed():
     ctx.ui.flush.assert_awaited_once()
 
 
-def test_start_new_session_cb_rotates_thread_id():
+async def test_start_new_session_cb_rotates_thread_id():
     """``/new`` via channel calls this callback — must generate a new
     thread id, push into runtime state, and sync the channel runtime."""
     agent = _agent("a")
@@ -311,13 +310,13 @@ def test_start_new_session_cb_rotates_thread_id():
         state,
         runtime,
     )
-    _run(cb())
+    await cb()
 
     assert state.thread_id == "freshly-generated-tid"
     assert runtime.thread_id == "freshly-generated-tid"
 
 
-def test_start_new_session_cb_leaves_agent_alone():
+async def test_start_new_session_cb_leaves_agent_alone():
     """``/new`` rotates thread only — agent handle must stay put
     (serve's agent is a single pre-loaded instance, not per-thread)."""
     agent = _agent("a")
@@ -328,12 +327,12 @@ def test_start_new_session_cb_leaves_agent_alone():
     )
 
     cb = _make_serve_start_new_session_cb(state)
-    _run(cb())
+    await cb()
 
     assert state.agent is agent
 
 
-def test_serve_resume_callback_syncs_reloads_and_adopts_workspace():
+async def test_serve_resume_callback_syncs_reloads_and_adopts_workspace():
     cfg = _config()
     old_agent = _agent("old-agent")
     reloaded_agent = _agent("reloaded-agent")
@@ -364,7 +363,7 @@ def test_serve_resume_callback_syncs_reloads_and_adopts_workspace():
             side_effect=_load_agent,
         ) as load_agent,
     ):
-        _run(cb("new-tid", "/new-ws"))
+        await cb("new-tid", "/new-ws")
 
     sync_server.assert_awaited_once_with(cfg, workspace_dir="/new-ws")
     load_agent.assert_called_once_with(workspace_dir="/new-ws", config=cfg)
@@ -376,7 +375,7 @@ def test_serve_resume_callback_syncs_reloads_and_adopts_workspace():
     assert runtime.agent is reloaded_agent
 
 
-def test_hook_emits_resume_warning_after_resume_callback_adopts_thread():
+async def test_hook_emits_resume_warning_after_resume_callback_adopts_thread():
     cfg = _config()
     old_agent = _agent("old-agent")
     reloaded_agent = _agent("reloaded-agent")
@@ -399,7 +398,7 @@ def test_hook_emits_resume_warning_after_resume_callback_adopts_thread():
             return_value=reloaded_agent,
         ),
     ):
-        _run(cb("abc12345-resumed-tid", "/new-ws"))
+        await cb("abc12345-resumed-tid", "/new-ws")
 
     hook = _make_serve_cmd_completed_hook(state, runtime, config=cfg)
     ctx = MagicMock()
@@ -410,14 +409,14 @@ def test_hook_emits_resume_warning_after_resume_callback_adopts_thread():
     cmd = MagicMock()
     cmd.name = "/resume"
 
-    _run(hook(ctx, reloaded_agent, cmd))
+    await hook(ctx, reloaded_agent, cmd)
 
     ctx.ui.append_system.assert_called_once()
     assert "in-memory state" in ctx.ui.append_system.call_args.args[0]
     ctx.ui.flush.assert_awaited_once()
 
 
-def test_serve_resume_callback_preserves_state_when_sync_fails():
+async def test_serve_resume_callback_preserves_state_when_sync_fails():
     cfg = _config()
     old_agent = _agent("old-agent")
     loaded_but_not_adopted = _agent("loaded-but-not-adopted")
@@ -442,7 +441,7 @@ def test_serve_resume_callback_preserves_state_when_sync_fails():
         patch("EvoScientist.cli.commands.set_active_workspace") as set_active,
         pytest.raises(RuntimeError, match="workspace conflict"),
     ):
-        _run(cb("new-tid", "/new-ws"))
+        await cb("new-tid", "/new-ws")
 
     load_agent.assert_called_once_with(workspace_dir="/new-ws", config=cfg)
     set_active.assert_called_once_with("/old-ws")
@@ -455,7 +454,7 @@ def test_serve_resume_callback_preserves_state_when_sync_fails():
     assert runtime.thread_id == "old-tid"
 
 
-def test_serve_resume_callback_load_failure_does_not_sync_or_adopt():
+async def test_serve_resume_callback_load_failure_does_not_sync_or_adopt():
     cfg = _config()
     old_agent = _agent("old-agent")
     state = _runtime_state(
@@ -479,7 +478,7 @@ def test_serve_resume_callback_load_failure_does_not_sync_or_adopt():
         ) as sync_server,
         pytest.raises(RuntimeError, match="load failed"),
     ):
-        _run(cb("new-tid", "/new-ws"))
+        await cb("new-tid", "/new-ws")
 
     load_agent.assert_called_once_with(workspace_dir="/new-ws", config=cfg)
     set_active.assert_called_once_with("/old-ws")
@@ -493,7 +492,7 @@ def test_serve_resume_callback_load_failure_does_not_sync_or_adopt():
     assert runtime.thread_id == "old-tid"
 
 
-def test_hook_handles_both_agent_and_thread_swap():
+async def test_hook_handles_both_agent_and_thread_swap():
     """Edge case: a command that changes both (hypothetical). Both
     updates must land in runtime state."""
     old_agent = _agent("old-agent")
@@ -506,7 +505,7 @@ def test_hook_handles_both_agent_and_thread_swap():
     ctx.thread_id = "new-tid"
     cmd = MagicMock()
 
-    _run(hook(ctx, old_agent, cmd))
+    await hook(ctx, old_agent, cmd)
 
     assert state.agent is new_agent
     assert state.thread_id == "new-tid"

tests/test_slack_channel.py

@@ -4,7 +4,6 @@ import pytest
 
 from EvoScientist.channels.base import ChannelError
 from EvoScientist.channels.slack.channel import SlackChannel, SlackConfig
-from tests.conftest import run_async as _run
 
 
 class TestSlackConfig:
@@ -38,24 +37,24 @@ class TestSlackChannel:
         assert channel.config is config
         assert channel._running is False
 
-    def test_start_raises_without_bot_token(self):
+    async def test_start_raises_without_bot_token(self):
         config = SlackConfig(bot_token="", app_token="xapp-test")
         channel = SlackChannel(config)
         with pytest.raises(ChannelError, match="bot token"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_start_raises_without_app_token(self):
+    async def test_start_raises_without_app_token(self):
         config = SlackConfig(bot_token="xoxb-test", app_token="")
         channel = SlackChannel(config)
         with pytest.raises(ChannelError, match="app token"):
-            _run(channel.start())
+            await channel.start()
 
-    def test_stop_when_not_running(self):
+    async def test_stop_when_not_running(self):
         config = SlackConfig(bot_token="xoxb-test", app_token="xapp-test")
         channel = SlackChannel(config)
-        _run(channel.stop())
+        await channel.stop()
 
-    def test_send_returns_false_without_client(self):
+    async def test_send_returns_false_without_client(self):
         from EvoScientist.channels.base import OutboundMessage
 
         config = SlackConfig(bot_token="xoxb-test", app_token="xapp-test")
@@ -66,7 +65,7 @@ class TestSlackChannel:
             content="hello",
             metadata={"chat_id": "C123"},
         )
-        result = _run(channel.send(msg))
+        result = await channel.send(msg)
         assert result is False
 
 

tests/test_status_bar.py

@@ -2,7 +2,6 @@
 
 from __future__ import annotations
 
-import asyncio
 from datetime import datetime, timedelta
 from typing import ClassVar
 
@@ -243,7 +242,7 @@ def test_build_status_text_uses_rich_styles():
     assert text.spans
 
 
-def test_build_session_status_snapshot_uses_fallback_window(monkeypatch):
+async def test_build_session_status_snapshot_uses_fallback_window(monkeypatch):
     class _FakeModel:
         model_name: ClassVar[str] = "provider/demo-model"
         profile: ClassVar[dict[str, object]] = {}
@@ -262,16 +261,12 @@ def test_build_session_status_snapshot_uses_fallback_window(monkeypatch):
         _fake_count,
     )
 
-    snapshot = asyncio.run(
-        build_session_status_snapshot(
-            "thread-1",
-            pending_user_text="pending",
-            graph_gateway=FakeGraphGateway(
-                thread_store=FakeThreadStore(
-                    messages=[HumanMessage(content="existing")]
-                )
-            ),
-        )
+    snapshot = await build_session_status_snapshot(
+        "thread-1",
+        pending_user_text="pending",
+        graph_gateway=FakeGraphGateway(
+            thread_store=FakeThreadStore(messages=[HumanMessage(content="existing")])
+        ),
     )
 
     assert snapshot.model_full == "provider/demo-model"