Independent review: substring matching lost the selected objective (after a
message offering API or UI work, '/goal ship the API' and '/goal ship the UI'
kicked identically), and the gateway and TUI /goal paths still re-sent the
full goal. The rule now lives once in hermes_cli.goals.goal_kick_prompt:
the pointer replaces the goal only when the goal is >= 400 chars, contained
in the last user message, and >= 80% of it. The CLI reads the last user
message from its history; the gateway and TUI read it from the SessionDB
(fail-safe: goal sent verbatim).
Tests: a fragment that selects one option is kicked verbatim; a long goal
that is a minority of a longer message is kicked verbatim; all three
surfaces call the shared function.
`/goal <text>` queues the goal text as the next user turn to start the loop.
When that text is what the user just pasted (a handoff note, a plan the agent
already has), the kickoff re-sends it verbatim: in the 1,393-agent run
`/goal <2,000-char handoff>` came 16 minutes after the same note was pasted
as a message, and the agent spent 11 API calls / 6 min deciding it was a
replay, with the note duplicated in context from then on.
_goal_kick_prompt() compares the normalized goal against the LAST user
message (string or block content); when contained, the kick is a one-line
'[Goal set] Continue with the goal you were just given; there is no need to
re-read it.' Otherwise the goal text is sent as before. Both /goal <text>
and /goal draft use it.
Tests (2): a just-pasted goal (plain and block content) kicks with the
pointer; a new goal, an unrelated last message, and a goal pasted in an
OLDER turn kick verbatim.
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.
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.
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.
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}.
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.
326 methods lifted by AST (bodies identical; ast.dump-verified) into
hermes_cli/cli_{tui,status_bar,voice,model_switch,session,stream,modal,
terminal,info,loops}_mixin.py. cli.py-internal symbols resolve via lazy
'from cli import ...' inside each method (no import cycle; patch('cli.X')
keeps working). The three 'global' writers (_skill_commands, _cli_wake_owner)
now write the cli module attribute explicitly so the origin's readers still
see them. Dropped imports left unused in cli.py; kept display_hermes_home /
build_welcome_banner as re-exports (mixins + tests resolve them via cli).
Repointed two AST change-detector tests to cli_tui_mixin.py; one test
fixture now keeps 'cli' in sys.modules across its patch.dict scope.
70 nested closures (every prompt_toolkit key handler and handler factory, the
clarify/sudo/secret/approval/model-picker/command-palette/stash display
renderers, hint/placeholder/spinner callables, process_loop, spinner_loop,
_signal_handler, wake-startup) become HermesCLI._tui_* methods; run() only
binds them (kb.add(...)(self._tui_x)). The run() locals they closed over are
published on self at their original binding sites (_tui_multiline_shortcuts,
the four paste list-cells; process_loop reads self._app). Then four zero-
back-ref prologue regions become _tui_print_startup, _tui_init_run_state,
_tui_build_key_bindings -> kb and _tui_build_layout(kb) -> (layout, style).
All moved bodies verified AST-identical modulo cli_ref->self and the free-var
rewrites; unresolved-free-name check on every new method is empty. Source-
inspection tests repointed: getsource(run) -> _tui_process_loop, AST lookup of
handle_enter -> _tui_handle_enter.
process_command resolves the canonical command (hermes_cli/commands.py registry)
to a handler via _slash_handler(): a 44-entry _SLASH_DISPATCH table for the
irregular cases (no-arg handlers, differently named methods, /exit and /update
return-value adapters) and the _handle_<name>_command(cmd) naming convention for
the other 47. Inline branch bodies became small _cmd_<name> methods; the
else-fallthrough (quick_commands -> plugins -> bundles -> skills -> prefix
expansion -> unknown) became _process_unregistered_slash. Pre-dispatch side
effects (pre_command hook, pending-resume disarm) and the False-exits-REPL
contract are unchanged. Also unifies the three random-tip blocks into
_print_random_tip and the repeated arg parsing into _slash_args; removes two
zero-reference helpers (_run_curses_picker, _try_launch_chrome_debug).
Parity guard: tests/cli/test_slash_dispatch_table.py asserts every command of
the old chain resolves and that no other registry command silently gained a
handler.
GatewayRunner.__init__ snapshotted _ephemeral_system_prompt once from the
launch profile's config and _get_system_prompt_for_channel returned that
string for every source, so under multiplex a routed profile's
display.personality / agent.system_prompt never injected (#89161), and
/personality from any chat rewrote the one process-global attribute for
everyone.
Drop the snapshot: _get_system_prompt_for_channel now calls
_load_ephemeral_system_prompt() (env var, then
resolve_ephemeral_system_prompt_from_config(_load_gateway_runtime_config()))
on each call. Its caller run_sync already runs inside
_profile_runtime_scope, so the routed profile's config.yaml is what gets
read; single-profile hot-edits of the personality also take effect on the
next turn instead of requiring a restart. /personality only persists via
persist_personality() (get_hermes_home()/config.yaml = the routed profile)
and no longer touches in-memory state.
Fixes#89161
Co-authored-by: worlldz <101180447+worlldz@users.noreply.github.com>
Adds two bounded fast modes on top of the static /fast toggle, default OFF:
- `auto`: every user turn opens a `agent.fast_auto_seconds` (default 60s)
window; requests inside it carry the provider fast param, later tool-loop
requests fall back to standard pricing.
- `cold`: the same window, but only on the first turn of a session (no prior
user/assistant/tool history).
agent/fast_mode.py holds the whole policy: `begin_turn()` at the
run_conversation ingress arms `agent._fast_until`; `effective_request_overrides()`
is consumed in the ONE place request_overrides feed the transports
(build_api_kwargs), so the fast param is a per-request kwarg only. System
prompt, tools and messages are untouched — the prompt cache is preserved.
resolve_fast_mode_overrides() is now the single gate for static and bounded
modes and accepts provider/base_url: OpenRouter, Nous, Copilot, Azure,
Bedrock and custom base_urls never receive service_tier/speed (#34308's
route gating). Both existing callers (CLI turn route, gateway turn route)
and the TUI config.set path pass the route.
Surfaces: config `agent.service_tier: auto|cold` + `agent.fast_auto_seconds`,
`/fast auto|cold` in CLI, gateway (picker gains both entries), TUI/desktop
config.set; status shows the mode; web dashboard select lists the real
values. Docs: configuration.md Fast Mode section with mode table + cost note,
slash-commands, cli-config.yaml.example, locale strings for the two picker
entries.
Salvages #89991 (bounded fast modes) and #34308 (route gating).
Fixes#64785, #74730.
Co-authored-by: Eva <239388517+100yenadmin@users.noreply.github.com>
Co-authored-by: kbaicai <kbaicai@qq.com>
The speed=fast allowlist still gates on Opus 4.6, but the fast-mode
matrix has changed twice since it was written (verified against the
live docs, platform.claude.com/docs/en/build-with-claude/fast-mode):
- Opus 4.8 and Opus 5 SUPPORT fast mode (research preview, Claude API
only — not Bedrock/Vertex/Foundry).
- Opus 4.6 LOST fast mode on 2026-06-29. The parameter does not error:
requests silently run at standard speed and bill standard rates
(usage.speed: 'standard'). Today's allowlist therefore shows 4.6
users a fast toggle that does nothing, while denying it to the two
models that actually support it.
- Opus 4.7 never had it and hard-400s (unchanged).
- Dedicated '…-fast' model ids (OpenRouter's claude-opus-4.8-fast)
select fast inference via the model field and are explicitly
excluded from the param gate.
Both gates move in lock-step as before: the adapter param gate
(agent.anthropic_adapter._supports_fast_mode) and the CLI toggle gate
(hermes_cli.models._is_anthropic_fast_model). Docstrings now record
the history in both directions so the next matrix change has context.
## How to test
scripts/run_tests.sh tests/agent/test_anthropic_adapter.py tests/cli/test_fast_command.py -- -q
113 tests pass. The updated predicate/matrix tests fail against the
previous allowlist (verified by stashing the source changes). Tested
on Linux (aarch64).
Desktop's cold resume (defer_history + omit_messages, transcript paged over
REST) only ever holds the live tip segment in memory, but session.resume
bounded it against the FULL compression lineage (sessions.max_resume_messages,
default 20000). A Bot Chat with 85 compaction segments / ~29k lineage rows
behind a ~700-row tip was refused at 20001, sent zero model prompts, and sat on
"Waking up default…" forever — the healthiest possible session shape, rejected
by a guard sized for in-memory materialization.
- hermes_state: one `_resume_lineage_ids` definition shared by the resume
readers (get_resume_conversations, get_ancestor_display_prefix) and the
guard (assert_resume_safe / get_resume_message_count). Guard grows
`tip_only=` and names the scope it counted; the branch-aware lineage the
readers already used is now what the guard counts too (a /branch copy was
being counted against its parent's rows).
- tui_gateway session.resume: deferred, omit_messages and lazy resumes are
bounded by the tip; only the full in-memory lineage resume keeps the
lineage-wide bound. Deferred hydration falls back to tip-only history when
the lineage exceeds the limit instead of loading the rows the guard refused.
- CLI mid-setup tip-only path routes through the same guard instead of
borrowing assert_export_safe.
- docs: sessions.max_resume_messages / max_export_messages documented with the
per-surface scope.
Live repro (real SessionDB fixture, 85 segments / 29,226 lineage rows / 666 tip
rows, real tui_gateway.server.handle_request): before — deferred resume ->
4130; after — ok, hydrated history=666 prefix=0; the non-deferred full resume
still returns 4130 on the same fixture.
hermes chat -q wired the interactive prompt_toolkit clarify callback
unconditionally, but a -q turn never builds the prompt_toolkit
application — the modal can never be painted or answered, so the turn
polls its response queue until agent.clarify_timeout expires (default
3600 s, 0 = unlimited). The gateway, cron jobs, the kanban dispatcher
and inter-agent wakeups all deliver work as -q turns. Route the
single-query case to a headless callback at the agent-construction site
that already knows _single_query_mode, mirroring _oneshot_clarify_callback
on the -z path (#94943; third member of the family after #86909 and
#88013).
Follow-up to the salvaged #88097: the same normalization covers the
Shift+letter class reported in #92343 (xterm modifyOtherKeys and both
kitty CSI-u codepoint forms), plus a guard that plain ASCII typing
never triggers the ESC-prefix predicate.
Two growth leaks closed:
1. Pushed-branch tier (the dominant survivor class — 24 of 33 preserved
trees, ~18GB on the reporting box): managed installs fetch with a
single-branch refspec, so pushed PR branches never get refs/remotes/*
entries and read as 'unpushed' forever. When a clean tree's branch head
EXACTLY matches origin (one lazy git ls-remote per sweep), the checkout
is redundant: reap the TREE, keep the BRANCH ref (shielded from the
orphaned-branch pass). Anything diverged/unverifiable stays preserved.
Applied to both the startup pruner and hermes worktree prune/list.
2. Cron-tick maintenance: the pruner only ran on hermes -w launches, so
gateway-driven boxes accumulated trees for days. The scheduler tick now
dispatches the same conservative pruner on a daemon thread, throttled
to once per 6h, against the install checkout + job-workdir repos that
have a .worktrees/ dir.
Follow-up to the cherry-picked #41909/#92696 + #39760 + #97970 cluster:
- single field-key namespace (display.status_bar.fields) instead of the
second tui_statusbar_fields list; cache_hit/latency/tps/stash/battery/
title join the existing key set
- cache-hit % prefers the baseline-delta regime (resets on model switch
and compression) and hides on zero cache reads instead of alarming 0%
- latency/tps segments added to the styled fragment renderer too
- docs updated in website/docs/user-guide/configuration.md
- 7 new tests: rolling latency/t/s, NaN/negative guard, field filtering,
baseline resets, title badge gating
Add a ◎ XX% indicator to the CLI status bar showing the prompt cache
hit rate (cache_read / prompt_tokens). This helps users monitor how
effectively their provider's prompt caching is working.
Features:
- Color-coded: green (≥70%), yellow (40-70%), red (<40%)
- Adaptive precision: integer on narrow terminals, one decimal on wide
- Only shown when cache data is available (provider supports it)
- Compatible with OpenAI, Anthropic, DeepSeek, xiaomi, and other
providers that return prompt_tokens_details.cached_tokens
Tests: 6 new test cases, 43/43 passing
Allow users to control which fields appear in the interactive CLI status
bar via display.status_bar.fields in config.yaml.
Available fields: model, context_pct, context_detail, compressions,
bg_tasks, bg_processes, duration, prompt_elapsed, yolo, total_tokens.
When the list is empty (default), all fields are shown as before.
The field order is fixed (model always first); the config controls
visibility only. Narrow terminals (<76 cols) automatically drop
context_detail regardless of config.
total_tokens is opt-in only (not shown by default) to avoid width
overflow in the prompt_toolkit fragment renderer.
Closes#41909
Live-reproduced on main: /handoff poll-waited a flat 60s for a TERMINAL
state, but the gateway's dispatch is a full synthetic agent turn (whole
transcript replay + delivery) that routinely exceeds 60s on long sessions.
The CLI then printed "Timed out waiting for the gateway. Is `hermes
gateway` running?" (false diagnosis), called fail_handoff() on the RUNNING
row (stomping the gateway's claim), and promised "Your CLI session is
intact" after switch_session had already re-pointed the session. The
watcher later overwrote failed -> completed: split-brain.
- hermes_state.fail_handoff gains only_states CAS; waiters can only fail
rows still pending. Owner (gateway watcher) keeps the unconditional form.
- CLI wait loop is two-phase: 60s for the CLAIM (pending) — a timeout
there really does mean no gateway — then up to 15 min for the claimed
dispatch with 30s heartbeats; a running row is never failed by the CLI.
- Desktop handoff.fail RPC now CAS-fails pending rows only; a running row
returns {failed: false, state: running} instead of stomping the claim.
Repro (real _handoff_watcher, real state.db, CLI as separate process,
75s dispatch): before — CLI timeout @60s + false message + row stomped;
after — pending->running@5s->completed@80s, clean CLI exit.
74a95a3ddf promoted /bg and /btw to independent canonical commands and
retired /background entirely (hermes_cli/commands.py's COMMAND_REGISTRY
has no "background" command or alias). Three places still taught the
old name:
- skills/autonomous-ai-agents/hermes-agent/references/slash-commands.md:
the bundled hermes-agent skill's own slash-command reference — the
skill's SKILL.md explicitly routes the model here for in-session
command questions, so a model following it would emit the dead
`/background <prompt>` and never learn /btw exists.
- ui-tui/README.md: listed /btw as an alias of /background, which is
simply wrong post-split (both are independent, alias-free commands).
- tests/cli/test_cli_background_status_indicator.py: docstring/comments
described the ▶ indicator by the retired command name.
No behavior change; corrects documentation only.
/bg (formerly /background, which is retired) keeps the existing semantics:
spawn a fresh, independent agent session in the background.
/btw is now its own command matching the convention other harnesses use:
ask a quick side question ABOUT the current conversation without
interrupting it. A one-shot auxiliary LLM call (main model by default,
overridable via auxiliary.side_question.* in config.yaml) answers from a
read-only transcript snapshot — the live session's history, role
alternation, and prompt cache are untouched, and the current turn keeps
running.
Surfaces wired: CLI (inline mid-run dispatch), gateway (all messengers,
busy-dispatch table + idle dispatch, i18n across all 17 locales), TUI
(prompt.btw RPC + btw.complete event), Discord native slash, relay
command manifest, desktop exec routing, docs (EN + zh-Hans).
Reuse kitty Unicode placeholders plus after_render write_raw so
prompt_toolkit's screen-diff can host the same crisp sprite as the TUI.
Re-queue the transmit after Ctrl+L / resize so the image comes back.
Co-authored-by: Sam Foreman <saforem2@gmail.com>
The protected agent-instruction gate grants one operation and persists
nothing, but only the TUI/desktop and Runs transports were taught that.
The prompt_toolkit panel, the input() fallback, and the ACP editor menu
still rendered "Allow for session", so a user editing SOUL.md tapped it,
got re-prompted on the next write, and read the gate as broken.
Thread allow_session through prompt_dangerous_approval so a caller that
re-asks every time collapses every surface to once/deny, and cover the
producer-to-transport contract end to end.
text.splitlines() returns [] for empty strings. Accessing msg_lines[0]
then raises IndexError, making session resume crash when the session
contains a message with empty or whitespace-only text (e.g. reasoning-only
turns, tool-only assistant messages).
Guard with `or [""]` in all three branches (user, assistant_last,
regular assistant) so an empty message renders as a blank line.
Fixes#59265
Co-authored-by: AlexFucuson9 <AlexFucuson9@users.noreply.github.com>
tests/tools/test_website_policy.py and tests/cli/test_surrogate_sanitization.py
repeatedly failed under the parallel CI runner (process-teardown timeouts /
async-timeout flakes) across multiple unrelated PRs this cycle, while passing
locally. Removed per maintainer direction to stop the flake taxing every PR.
Builds on @fattchris resolve_turn_limit salvage (#67696): flips the default
from a numeric cap to unlimited across all construction paths (CLI, agent_init,
run_agent subagents), adds inf/infinity/null to the unlimited spellings, and
sets DEFAULT_CONFIG agent.max_turns to null. The turn cap caused more problems
than it solved (silent mid-task truncation).
Teknium