Commit Graph

2329 Commits

Author SHA1 Message Date
kshitijk4poor 337113c1b8 fix(agent): a parent directory the process cannot stat is not a crash while locating .git (#8751)
_find_git_root walked cwd's parents with Path.exists(); on hosts where a parent is
mode 700 for another user that raised PermissionError out of prompt construction.
Treat an unreadable ancestor as 'no .git here'.
2026-09-05 19:01:22 +05:30
kshitijk4poor 52cf39c908 refactor(auth): reuse _CLEAR_STATUS and the merge's None short-circuit in the reset path
Behaviour-preserving cleanup of the salvaged fix:

- reset_statuses goes back to replace(e, **_CLEAR_STATUS, ...) instead of
  re-spelling the six status fields; every other clear site uses the
  constant, so a seventh field would otherwise drift here. The
  count/new_entries/cleared_ids trio collapses to the original stale-list
  shape, and entry.failure_reason is read through the dataclass's
  __getattr__ like the rest of the module.
- Both bypass sites (write_credential_pool, _update_root_pool_rows) pass
  disk_entry=None to _merge_disk_cooldown_state for a cleared id instead of
  adding a parallel branch; the helper already returns the entry unchanged
  for a missing disk row.
- The cleared-id set in _update_root_pool_rows is built once, not per disk
  row (an Iterable argument would also have been drained on the first row).
- persist_pool_entries passes status_cleared_ids plainly; the kwargs splat
  existed only for one test fake, which now accepts the kwarg.
- Docstrings trimmed to the WHY.

Same 116 targeted tests green; reverting the two source files to main still
fails the two regression tests and passes the guard test.
2026-09-05 17:43:45 +05:30
rodrigogs 8378551311 fix(auth): make hermes auth reset actually clear a binding cooldown
Port onto the split layout: persist_pool_entries now forwards
status_cleared_ids to BOTH writers — write_credential_pool (hermes_cli/
auth.py) and the root-store row merge for single-use-refresh providers
(_update_root_pool_rows), which the original fix predates. Both skip the
disk recency merge for deliberately-cleared entries; the kwarg is only
sent when non-empty so upstream fakes with the old signature keep
working. reset_statuses clears failure_reason (lives in extra, replace()
cannot reach it) alongside the status fields.

tests/agent/test_credential_pool.py: 62 passed; refresh-race suite: 3
passed (mutation: the reset/binding tests fail with the fix stashed).
The dashboard-auth-gate / user-providers failures are pre-existing on
pristine upstream/main (reproduced with changes stashed).
2026-09-05 17:43:45 +05:30
kshitijk4poor 07b161e241 refactor(context): fold simplify findings into oversized-file fallback
- Check text tokens BEFORE building the fenced block: an oversized file no
  longer allocates a second MB-scale string just to discard it, and the token
  estimate runs once instead of twice.
- Drop the companion warning: the fallback block alone carries the message,
  so the model no longer reads "too large to inline safely" twice (Context
  Warnings + Attached Context).
- Extract _on_disk_reference_block + _agent_visible_path shared by the binary
  and oversized blocks (stat/format_bytes/OSError fallback, container path
  translation, nudge text were copy-pasted).
2026-09-05 16:44:23 +05:30
Gille cfa1a90b8c fix(context): fall back for oversized file refs 2026-09-05 16:44:23 +05:30
Teknium d61cff60e3 fix(context): subdirectory AGENTS.md hints keep head+tail and warn when over the ceiling; ceiling 8k -> 32k
The on-demand subdirectory hint loader (agent/subdirectory_hints.py, #5291) tail-chopped anything past
8,000 chars with a bare marker and no log line. apps/desktop/AGENTS.md (11k) has been arriving cut off in
every Desktop-area session since it was written, and nobody could tell. Compared with the field: Codex
caps its whole instruction chain at 32 KiB (project_doc_max_bytes) and documents it; Claude Code and
Cursor apply no cap to nested files; OpenCode has no nested discovery at all. Our on-demand + cache-safe
+ ancestor-walk design is the strongest of the four; only the constant and the silent cut were wrong.

Now: ceiling 32,000 chars (Codex's number, a guard against a stray huge CLAUDE.md in a vendored tree, not
a target), and truncation goes through prompt_builder._truncate_content — head 70% + tail 20%, a marker
naming the file to read_file for the rest, and a WARNING in the log. Area AGENTS.md files should stay
around 8k anyway: the text lands in a tool result on the first touch of the directory.

Tests: oversized hint keeps head+tail, names the path, and logs; a 12k area file (over the old cap, under
the new) arrives intact.
2026-09-04 02:03:08 -07:00
Teknium a0be177aac fix(compat): pointers resolve to the object that MOVED, not a same-named stranger; stdin checker binds stdin= to the splatted definition
Review findings on #102117 (independent reviewer + itsflownium):

* hermes_cli.kanban_db.connect / connect_closing pointed at hermes_cli.projects_db (different DB, no
  board= parameter). The compat generator ranked candidate homes by path proximity when a name is
  defined in several modules. Now it requires shape compatibility with the BASE definition (same
  literal for constants, superset of parameter names for defs) and prefers the facade's own
  <stem>_* sibling. Same class fixed for tools.tts_tool.DEFAULT_XAI_BASE_URL (-> tts_tool_providers),
  and 17 constants/defs that had been pointed at same-named strangers (Matrix MAX_MESSAGE_LENGTH ->
  Signal's 8000, tts MAX_TEXT_LENGTH -> BlueBubbles', honcho/retaindb/supermemory *_SCHEMA -> another
  plugin's schema, ...) are now restored from BASE verbatim instead.
* send_yuanbao_direct (restored-def): body called adapter._outbound.send_direct, which HEAD moved to
  the sender; rewritten to adapter._outbound.sender.send_direct.
* COMPAT_MANIFEST.md states the scope explicitly: public top-level names only; private names and
  test monkeypatch seams are not preserved.
* scripts/check_subprocess_stdin.py: _splat_carries_stdin looked 30 lines ahead in the file text
  and was satisfied by an unrelated later stdin=; it now finds the splatted name's definition via AST
  and requires stdin inside that expression/body.

Tests: tests/test_compat_manifest_targets.py (pointer identity vs the facade's sibling; kanban
connect(board=) opens a Kanban DB, not projects.db; both FAIL on the previous layer),
test_subprocess_stdin_guard gains the false-negative probe, and the MoA -Q quiet-output contract
tests are back (tests/agent/test_moa_quiet_reference_output.py) against build_moa_facade.
2026-09-03 22:00:01 -07:00
Teknium ba030bc0db fix(test-seams): monkeypatch.setattr facade aliases — also patch the defining module (57 files)
Tests did monkeypatch.setattr(<facade module>, name) where name is now defined in a
sibling module and the production path reads the sibling's binding. Where production
reads through BOTH bindings the setattr is duplicated onto the defining module (import
added next to the existing alias import); where only the sibling reads it the target is
repointed. Seams whose production readers go through the facade are left alone.
2026-09-03 19:33:28 -07:00
Teknium deb9f3c1bd simplify(compat): hermes_cli/commands — drop the last PEP 562 lazy re-export hook (25 names), repoint 10 callers + 11 test files to commands_platforms/commands_completion 2026-09-03 17:00:26 -07:00
Teknium eb2e96d657 fix(compat-fallout): module-restore also rebinds pkg.<child> attributes so 'from pkg import child' matches sys.modules after the wipe 2026-09-03 16:20:10 -07:00
Teknium 1bc3af0c05 fix(compat-fallout): sys.modules-wiping agent tests restore the original module objects afterwards (their wipe of hermes_*/tools.* detached sibling files' monkeypatches now that facades no longer re-import the defining modules) 2026-09-03 16:11:41 -07:00
Teknium e6bfbf82fa fix(compat-fallout): repoint 4 test files to defining modules (update_cmd._detect_venv_python_processes, main_install_repair subprocess/_resolve_install_target_python, process_bootstrap.OpenAI, tools.mcp_tool gate import) 2026-09-03 15:54:46 -07:00
Teknium da96762ef8 fix(compat-fallout): repoint 8 test files off dropped facade names (run_agent._DB_PERSISTED_MARKER/_EPHEMERAL_SCAFFOLDING_FLAGS/_is_ephemeral_scaffolding, cli.AIAgent, main._make_tui_argv/_print_curator_recent_run_notice/_format_time_ago, model_switch._collect_authed_provider_slugs, acp has_provider shim); arg_coercion test imports model_tools to populate registry 2026-09-03 15:50:55 -07:00
Teknium 0e78694c72 simplify(compat): hermes_cli.main — drop 198 re-exports/aliases (155 eager + 45 lazy PEP 562 + _warn_stale_dashboard_processes alias + _time/_self/_LAZY_* machinery), repoint 17 source callers + 91 test files 2026-09-03 15:07:39 -07:00
Teknium 4fde117f4b simplify(compat): tests — repoint 638 web_server.<name> references (321 attr, 217 monkeypatch/patch.object, 60 from-imports, 40 patch() strings) across 65 test files to the owning modules 2026-09-03 14:21:52 -07:00
Teknium da7ee6353e simplify(compat): tools/browser_tool tests — repoint 56 test files from tools.browser_tool.<name> to the defining browser_tool_* sibling (patch where the name is looked up) 2026-09-03 14:18:25 -07:00
Teknium 7a33369e81 simplify(compat): interrupt — drop _ThreadAwareEventProxy/_interrupt_event legacy alias, repoint 2 test files
No runtime consumer read the proxy (terminal_tool/environments call is_interrupted()/set_interrupt()
directly); its only users were tests patching tools.interrupt._interrupt_event, which had no effect on
the code under test. tools/terminal_tool.py's own re-export of the name is owned by another worker.
2026-09-03 14:00:59 -07:00
Teknium c93ace77c2 simplify(compat): config/runtime_provider/plugins/commands/secrets_cli/kanban — drop 96 re-exports (incl. PEP 562 facades) + 3 aliases (get_pre_tool_call_directive/_block_message, get_telegram_handler_factories), repoint 56 callers + 50 test files 2026-09-03 14:00:17 -07:00
Teknium 14791b4d4e simplify(compat): approval — drop 43 facade re-exports + _command_detection_variants late-bind seam, repoint 30 callers + 46 test files
tools/approval.py no longer re-exports sibling names (approval_context/prompt/floors/detection/
human_wait/smart/gateway_wait); it imports only what it uses. Siblings reference sibling-defined
names directly (module-attribute reads on tools.approval_context so patching the defining module
still works); only facade-owned state (_lock, _gateway_queues, _permanent_approved, _denied,
_denial_breaker_addendum, _gateway_notify_cb) is still read back through tools.approval.
approval_detection calls its own _command_detection_variants instead of late-binding through the facade.
2026-09-03 13:49:57 -07:00
Teknium 7b8c11bcf7 simplify(compat): models — drop 52 re-exports from hermes_cli.models, repoint 16 callers + 41 test files 2026-09-03 13:48:49 -07:00
Teknium e3ab65fe80 simplify(compat): kanban_db — drop 73 re-exports/aliases, repoint 794 callers 2026-09-03 13:48:14 -07:00
Teknium 53db597201 simplify(compat): hermes_state — drop 81 re-exports + 3 registry aliases + 3 shims, repoint 45 callers + 60 test files
hermes_state.py: delete every '# noqa: F401 (re-exported...)' import block (hermes_state_common/errors/guard/
readpool/sessions/fts/dbfile/wal/repair/registry + agent.context_compressor _DB_PERSISTED_MARKER_KEY); keep
only the names hermes_state.py itself uses, without noqa.
hermes_state_registry.py: drop get_shared_session_db/release_shared_session_db/close_shared_session_dbs
aliases; every caller (gateway/, tools/, tui_gateway/, cron/, mcp_serve, run_agent, tests) now imports
acquire/release/close_all/release_or_close from hermes_state_registry.
hermes_state_titles.py: drop set_auto_title_if_empty shim (title_generator keeps its getattr fallback).
Re-remove shim-only names restored by 34abf954bd: latest_user_message_row_id (tests call
latest_message_row_id(key, role='user'); role-targeting assertions kept) and get_session_activity (tests
build the snapshot via agent.session_activity.build_activity_snapshot over db.get_session(sid)).
hermes_state_wal._log_once resolves its dedupe sets as module globals instead of via hermes_state;
hermes_state_repair helpers call module globals directly (tests patch hermes_state_repair.<name>).
Frozen updater surface untouched (update_cmd_maint imports only SessionDB from hermes_state).
2026-09-03 13:46:50 -07:00
Teknium 491e53ef3c simplify(compat): tool_search/skills_sync_client — drop 22 re-exports, repoint 5 callers + 5 test files; drop stale re-export prose in web_tools_truncate/file_operations_lint 2026-09-03 13:33:47 -07:00
Teknium 707161e77b simplify(compat): file_tools/send_message_tool/cronjob_tools/process_registry — drop 34 re-exports, repoint 5 callers + 27 test files 2026-09-03 13:30:27 -07:00
Teknium f6938b37f3 simplify(compat): terminal/file/environments — drop 42 re-exports/aliases, repoint 20 callers + 41 test files
tools/terminal_tool.py: drop 30 pure re-export names (lifecycle/config/backends/
sudo/guards/result/interrupt/utils/_DockerEnvironment/is_managed_tool_gateway_ready)
and the noqa-F401 comments on the 25 names the facade itself uses. Sibling modules
(terminal_tool_backends/_result/_sudo/_lifecycle, environments/base, process_registry)
that read removed names through the facade now import from the defining module.
tools/environments/base.py: drop 11 re-exports (base_output/base_session_env/
path_utils) and the BaseEnvironment.stop() compat alias (no in-tree caller; the
lifecycle hasattr(env, 'stop') fallback stays for third-party envs).
tools/environments/docker.py: drop 1 re-export + the re-export comment.
Callers/tests repointed to tools.terminal_tool_{lifecycle,backends,sudo,config,
guards,result}, tools.interrupt, tools.environments.{base_output,base_session_env,
path_utils}.
2026-09-03 13:29:55 -07:00
Teknium fcbe4acbef simplify(compat): tools/mcp_tool — repoint 20 non-test callers to the defining mcp_tool_* siblings 2026-09-03 13:29:35 -07:00
Teknium 2a95791992 simplify(compat): run_agent/model_tools/toolsets/acp/providers — drop 42 re-exports/aliases, repoint 15 callers + 99 test files
run_agent.py: delete the `# noqa: F401` re-export block (agent.process_bootstrap
OpenAI/_SafeWriter/_get_proxy_*, model_tools get_tool_definitions/
handle_function_call/check_toolset_requirements, FailoverReason,
_qwen_portal_headers/_routermint_headers, session_persistence names,
estimate_request_tokens_rough, ContextCompressor + friends, jittered_backoff,
prompt_builder names, message_sanitization names, tool_dispatch_helpers
names) — 41 names run_agent never used itself — and the `_STREAM_DIAG_HEADERS`
back-compat class alias (no in-tree reader). run_agent now imports only what
it uses (get_toolset_for_tool, is_local_endpoint, coalesce/uniquify tool-call
ids, cleanup_vm/get_active_env from terminal_tool_lifecycle).

agent/*: `_ra().X` late-binds that only reached a re-export now import the
defining module directly (agent_runtime_helpers -> process_bootstrap.OpenAI,
model_tools.handle_function_call, session_persistence._safe_session_filename_component;
agent_init -> model_tools.get_tool_definitions/check_toolset_requirements,
_lazy_headers("agent.client_lifecycle", ...) for qwen/routermint;
system_prompt -> agent.prompt_builder / model_tools directly, dropping its
own _ra() shim and the `_r` parameter threading). `_ra()` stays for
run_agent-resident names (logger, AIAgent, _hermes_home, _set_interrupt, ...).

toolsets.py: remove resolve_multiple_toolsets (shim-only, restored by
34abf954bd); tests/test_toolsets.py pins the same union behavior via
resolve_toolset over each name.

providers/__init__.py: drop the OMIT_TEMPERATURE re-export (no callers via the
package); ProviderProfile stays because __init__ uses it for annotations —
2 tests repointed to providers.base.

agent/iteration_budget.py: drop the "run_agent re-exports the class"
docstring pointer; 4 tests import IterationBudget from its home.

model_tools.py (arg_coercion names), agent/tool_executor.py, and
hermes_cli/cli_session_mixin.py repoints landed via a sibling commit on this
shared worktree.

Callers repointed: gateway/run.py, hermes_cli/cli_chat_turn_mixin.py,
hermes_cli/cli_tui_mixin.py, tui_gateway/session_workdir.py,
agent/transports/codex.py (one-line imports) + comment pointers in
tools/file_state.py, tools/schema_sanitizer.py, scripts/tool_search_livetest.py.
Tests: patch("run_agent.X") / monkeypatch.setattr(run_agent, "X") /
`from run_agent import X` -> defining module across 99 test files.
2026-09-03 13:28:22 -07:00
Teknium eb8a30cc2f simplify(compat): skills_sync/skill_manager/kanban/computer_use — drop 37 re-exports/aliases, repoint 8 callers + 9 test files 2026-09-03 13:27:01 -07:00
Teknium 8e1a5b9f46 simplify(compat): tools/wake_word — drop 5 re-exports, repoint 1 test 2026-09-03 13:24:36 -07:00
Teknium 98c140bc4b simplify(compat): code_execution_tool/environments.local — drop 36 re-exports, repoint 5 callers + 14 test files 2026-09-03 13:24:04 -07:00
Teknium d179f28307 simplify(compat): anthropic_adapter — drop 30 re-exports + 1 alias, repoint 22 caller files (32 sites), 38 test files (~125 sites) 2026-09-03 13:16:47 -07:00
Teknium 89fbd5d4d3 simplify(compat): conversation_loop — drop 9 re-exports, repoint 7 callers, 35 test sites 2026-09-03 13:12:23 -07:00
Teknium b818085298 simplify(compat): doctor/status — drop 13 re-exports + the doctor_* globals() facade (97 names), repoint 6 callers / 13 tests 2026-09-03 13:10:52 -07:00
Teknium eeb7671e69 simplify(compat): hermes_cli small facades — drop 7 re-exports/aliases (+relay_runtime alias module), repoint 12 callers/tests 2026-09-03 13:05:57 -07:00
Teknium 9e3df66e95 review-fix(locks): keep the lock-free cache read only for manual codex /compress (BASE exception)
BASE's _compress_codex_app_server_session was the single cache-peek site that read
_agent_cache without a lock when _agent_cache_lock was None (tests/gateway/
test_codex_hygiene_compaction.py::_slash_host relies on it). Every other BASE caller returned
None without a lock. Model that as an explicit lockless_fallback=True at that one call site.
2026-09-03 12:42:38 -07:00
Teknium 6e2796b82d review-fix(locks): restore BASE fail-loud lock-slot reads and _is_openai_client_closed truth table
Adolanium review §3 (Medium/Low). BASE shape at every steer/redirect/persist/agent-cache
lock site was: lock = getattr(obj, "_x_lock", None); if lock is not None: with lock: <direct
attribute read>; else: <getattr fallback for object.__new__ test stubs>. The refactor rewrote
several of these as 'with getattr(...) or nullcontext(): getattr(slot, None)', which (a) turned a
missing slot under the lock from a loud AttributeError into silent None and (b) in the gateway
peek helper read the cache without any lock when the lock attribute was absent.

Restored BASE semantics at:
- agent/agent_runtime_helpers.py::_requeue_pending_steer
- agent/interrupt_control.py: steer, redirect, clear_interrupt, _has_pending_redirect,
  _drain_pending_redirect, _drain_pending_steer (new _ic_slot helper: direct read under lock)
- agent/session_persistence.py::_persist_lock (explicit None check + BASE rationale)
- gateway/slash_commands.py::_cached_agent_for (BASE callers read ONLY under the lock; no lock -> None)
- gateway/run_agent_cache.py::_evict_cached_agent (BASE: self._agent_cache direct under lock)
- agent/client_lifecycle.py::_is_openai_client_closed: BASE body + docstring verbatim (outer
  is_closed first; inner _client.is_closed only when _client exists; else False)
- agent/stream_delivery.py::_ensure_stream_writer_state: restore BASE rationale that the lock is
  created unconditionally in agent_init (_STREAM_STATE) and the lazy path is stub-only

A/B (/tmp/rf/rev/ab_lock_fallbacks.py, ab_client_closed.py) is byte-identical BASE vs HEAD on
the reviewer's inputs + edge cases. tests/agent/test_lock_fallback_base_semantics.py pins it
(7 of 25 cases fail on the pre-fix tree).
2026-09-03 12:29:52 -07:00
Teknium c65aebccf0 review-fix(conversation-loop): finalize_turn binds stream-recovered final_response before fallible persist sub-steps (BASE order)
On BASE the #95514 stream-recovery rebound final_response inline, before
the tail-closing / persist-override / micro-compaction / _persist_session
calls inside the guarded persist try-block, so a raise in any of them left
the caller with the streamed text (cleanup_errors reported the failure).
HEAD's _close_transcript_tail returned the recovered value only on normal
completion; an exception in the same block dropped it and the user-visible
answer regressed to the pre-recovery '' (then rewritten by the explainer).

Split the helper into _drop_transcript_scaffolding / _recover_final_from_stream
/ _close_transcript_tail and rebind final_response in the guarded step as soon
as it is computed, restoring BASE ordering. Adds a regression test.
2026-09-03 09:54:04 -07:00
Teknium 8da430d69c review-fix(bedrock): restore call_converse_stream + classify_bedrock_error (public on main) and the 4 deleted streaming tests; pin the live agent stream recovery path
ethernet8023: IAM streaming-denial -> converse fallback and stale-ConnectionClosedError
client eviction lost all coverage. Both public helpers restored byte-identical to BASE
(with THROTTLE/OVERLOAD/CONTEXT_OVERFLOW patterns + is_context_overflow_error), the 4
tests re-added verbatim, and TestAgentBedrockStreamRecovery covers the path the agent
actually uses (chat_completion_helpers._bedrock_converse_call / _BedrockStream._fall_back_to_converse).
2026-09-03 09:39:57 -07:00
Teknium 3a8e3a2e88 review-fix(moa): restore _relay_moa_reference_event/_moa_reference_output_allowed + quiet-reference test
BASE (63279301bc) exported agent.agent_init._moa_reference_output_allowed and
_relay_moa_reference_event (salvaged in 3dfe712384 from #67334, plugin-importable,
zero in-tree callers) and covered them with
tests/agent/test_moa_quiet_reference_output.py. The simplification PR deleted both
the helpers and the test. Restore them byte-identical to BASE.

Note: the live build_moa_facade relay in agent/moa_loop.py is unchanged. A/B
harness (/tmp/rf/rev/moa_quiet_ab.py) shows BASE's facade relay already fired
moa.reference under platform=cli/tool_progress_mode=off — the guard only ever
lived in the orphan helper. -Q is protected on both trees by cli.py nulling
agent.tool_progress_callback (_configure_quiet_agent / BASE cli.py:22308).
2026-09-03 09:35:34 -07:00
Teknium 022785a541 Merge origin/main (63279301bc): reasoning-mandatory 400 recovery folded into turn_recovery/error_classifier/models_reasoning_caps 2026-09-03 05:36:21 -07:00
Teknium f6bd1633f7 fix(reasoning): GLM-5.3 on Nous/OpenRouter no longer 400s when thinking is disabled
Reasoning-mandatory routes answer reasoning: {enabled: false} with HTTP 400
"Reasoning is mandatory for this endpoint and cannot be disabled". Hermes
sends that disable for /reasoning none, agent.reasoning_effort: none, and the
one-shot thinking-exhaustion continuation override (which GLM-5.3-flash
triggers on its own). The Nous profile's catalog guard swallows the disable
only when its per-process capability cache already says mandatory; a gateway
that warmed the cache before the route flipped kept sending it, and the 400
was classified as a non-retryable format_error that aborted the turn.

- error_classifier: new reasoning_mandatory reason (retryable, no fallback,
  no compression), matched before the request-validation branch.
- conversation_loop: one-shot recovery — set agent._reasoning_disable_rejected,
  queue a catalog refresh for the provider, retry.
- chat_completion_helpers: _reasoning_config_for_wire drops every disable
  (configured or ephemeral) once the route has rejected one.
- hermes_cli/models: refresh_reasoning_caps_async(provider) forces a
  background re-fetch of the Nous/OpenRouter catalog.
- openrouter profile: omit a disable when the catalog marks the route
  mandatory (parity with the Nous profile).

Live: z-ai/glm-5.3-flash on the Portal with a poisoned mandatory:false cache.
Before: turn aborted with the 400. After: one retry, thinking stays on, turn
completes.
2026-09-03 05:20:11 -07:00
Teknium 3bff2a581d Merge origin/main (562ee8ab76): fold usage-less empty-response guard, unterminated tool-call strip, skills-first memory guidance into the simplified modules 2026-09-03 04:31:46 -07:00
Teknium 562ee8ab76 fix(prompt): agent routes task-learned knowledge (incl. user preferences/corrections) to skills; memory is the every-session exception
The memory guidance led with 'Save proactively' and the memory tool schema
ranked 'user preferences & corrections' as top priority, while the skills
nudge was a conditional 'offer to save'. In practice that asymmetry made
the agent end sessions writing memory entries (fighting a 2,200-char
budget) and skip updating the skill it had just used, even though the
procedure was the reusable artifact. Both surfaces now state the same
rule with skills first: what you learn doing a task, including the user's
preferences and corrections for that kind of work, goes in the task's
skill; memory is only for facts that apply to every session.
2026-09-03 04:02:35 -07:00
fangliquan 3755dca7d6 fix(agent): handle usage-less empty responses 2026-09-03 03:41:52 -07:00
Teknium 0071ba9965 Merge origin/main (561b053f79) into simp/forwardport: forward-port 220 main commits into the simplified tree 2026-09-03 03:31:03 -07:00
Teknium 561b053f79 perf(agents): run per-child timers on one shared scheduler thread
A fan-out of N in-process subagents used to add one sleeping daemon
thread per delegated child (delegate heartbeat, 30s) and one or two per
active turn (durable turn-lease refresher; turn-liveness watchdog).  A
profiled session with ~130 children was carrying ~1000 threads.  All
of these timers now run on a single process-wide daemon thread.

- agent/periodic_scheduler.py (new): heap-ordered periodic scheduler on
  one Condition-driven daemon thread.  schedule(fn, interval) -> handle;
  handle.cancel(wait=) blocks for an in-flight run like the old join.
  A callback returning False stops itself; a raising callback is logged
  at debug and rescheduled, so one bad timer cannot kill the rest.
- tools/delegate_tool.py: _heartbeat_loop body -> _heartbeat_tick,
  scheduled at _HEARTBEAT_INTERVAL; stale-cycle closure state and
  idle/in-tool thresholds unchanged; cancel(wait=5) in finally where the
  stop-event + join(5) lived.
- run_agent.py: _refresh_durable_turn_lease body scheduled at
  _lease_refresh_interval; lease-lost / refresh-error interrupt paths
  and the stop-event fencing are unchanged; the join(timeout=1.0) is now
  cancel(wait=1.0) so the interrupt clear still runs after any in-flight
  tick.
- agent/turn_liveness.py: TurnLivenessWatchdog.make_thread/start ->
  schedule(); the poll body is _tick(), same sampling state machine.

Bench (evals/fanout_resource_bench.py, 30 children / 10 worktrees,
ok=30/30 both): peak threads 168 -> 132.  At peak the old tree held 30
"Thread-N (_heartbeat_loop)" threads; the new one holds zero plus one
"hermes-periodic-scheduler".
2026-09-03 02:44:24 -07:00
Teknium c3b411dfb7 perf(agents): share one httpx transport pool across every agent's client
A fan-out of 30 delegated children built 183 httpx.HTTPTransport objects
(each with its own httpcore pool + parsed SSL context): 3 per agent x
(primary + aux clients). A profiled session with ~130 children held 107 TLS
sockets to one provider. Peak RSS for the 30-child bench drops 286 -> 195 MB;
live HTTPTransports 183 -> 2, ConnectionPools 183 -> 7.

What is shared: the sync `HTTPTransport` (pool + SSL context) per
(scheme, verify, proxy, happy-eyeballs) identity, in a bounded module dict.
What is NOT shared: the per-agent `httpx.Client` wrapper. Each client mounts
a `_SharedTransport` view whose `close()` marks only that view closed and
never touches the pool, so the #10933 contract (close client A, build client
B, B works) holds unchanged — the pinning tests in
test_create_openai_client_reuse.py / test_sequential_chats_live.py pass as-is.

Safety for cross-thread aborts: `_SharedTransport.handle_request` stamps its
id into `request.extensions`; `_iter_pool_sockets` now only shuts down a
shared pool's in-flight requests carrying the calling client's stamp and
never its idle connections, so interrupting child A cannot sever child B's
stream (#29507 / #72975 walker semantics preserved for unshared pools).

Also:
- `resolve_httpx_verify` caches one SSLContext per CA-bundle path. With
  SSL_CERT_FILE/HERMES_CA_BUNDLE set, every agent used to parse the bundle
  again and — because the share key is context identity — get a private pool.
- The client no longer builds a third, unused default transport; its
  default transport is the https view.
- Mounted transports now actually receive pool limits (Client-level
  `limits=` never reached them, so mounts ran on httpx defaults with a 5 s
  keepalive_expiry). The shared pool uses 50 keepalive / 1000 max so one
  pool covers a whole concurrent fan-out.
- `close_shared_transports()` really closes the pools (tests / shutdown).

Async clients (`async_mode=True`) stay unshared: an httpcore async pool is
bound to the event loop that first uses it. Proxy-backed clients keep
httpx's per-client proxy transport.
2026-09-03 02:35:21 -07:00
Teknium 80fae22bf5 fix(lsp): share one pyright process across git worktrees via workspaceFolders
Multi-root servers (pyright) are keyed by server_id; a file whose resolved
root is new for a running client is attached with
workspace/didChangeWorkspaceFolders instead of spawning another server.
Single-root servers keep the (server_id, workspace_root) key and behavior.
A profiled fan-out across ~30 worktrees ran 30-60 pyright processes
(~8.7 GB); the same fan-out now runs one.
2026-09-03 02:32:15 -07:00
Teknium 5540860ca6 fix(integration): bang_shell builds its env via build_subprocess_env()
Compaction moved run_bang_command's Popen within the env-guard scanner's
proximity window of _bang_env's os.environ.copy(). Route through the single
factory (build_subprocess_env() == _sanitize_subprocess_env(os.environ.copy()))
and allowlist the file for the import-failure fallback copy with justification.
2026-09-03 02:07:33 -07:00
Teknium dc130e54bf fix(integration): repoint auxiliary bridge source-inspection test to generalized field loop
gateway/run.py now bridges MODEL/BASE_URL/API_KEY via one (field, suffix)
table writing AUXILIARY_{_upper}_{_suffix}; the env keys set are unchanged.
2026-09-03 02:06:51 -07:00