Six days after #23937 (608 fixes) the codebase had accumulated 241 new
PLR6201 violations. Same mechanical `x in (...)` → `x in {...}` fix,
same zero-risk profile: set lookup is O(1) vs O(n) for tuple and the
two are semantically equivalent for hashable scalar membership tests.
All 241 instances fixed via `ruff check --select PLR6201 --fix
--unsafe-fixes`, zero remaining. Every changed value is a hashable
scalar (str/int/None/enum/signal); no risk of unhashable runtime
errors. No behavior change.
Test plan:
- 119 files changed, +244/-244 (net zero) — exactly one-line edits
- `ruff check` clean afterward
- Compile checks pass on the largest touched files (cli.py, run_agent.py,
gateway/run.py, gateway/platforms/discord.py, model_tools.py)
- Subset broad test run on tests/gateway/ tests/hermes_cli/ tests/agent/
tests/tools/: 18187 passed, 59 pre-existing failures (verified against
origin/main with the same shape — identical failure count, identical
category — all xdist test-order flakes unrelated to this change)
Follows the same template as PR #23937 ([tracker: #23972](https://github.com/NousResearch/hermes-agent/issues/23972)).
When Hermes runs on a remote host over SSH, MCP OAuth loopback flows
silently fail: the OAuth provider redirects the user's browser to
http://127.0.0.1:<port>/callback, which reaches the callback server
on the *remote* machine — not the local machine where the browser is
running.
_redirect_handler already detected SSH (via _can_open_browser) and
printed "Headless environment detected — open the URL manually." but
gave no guidance on how to actually reach the callback server. Users
got silent timeouts or "Could not establish connection" errors.
This is the same bug fixed for xAI-oauth and Spotify in #26592, which
added _print_loopback_ssh_hint() in hermes_cli/auth.py. mcp_oauth.py
uses the identical loopback callback pattern (http://127.0.0.1:<port>/callback
via _configure_callback_port / _wait_for_callback) but was missing the hint.
Fix: when SSH_CLIENT or SSH_TTY is set and _oauth_port is available,
print the ssh -N -L port-forward command and the OAuth-over-SSH guide
URL to stderr, consistent with the rest of _redirect_handler's output.
Tests: 4 new cases in TestRedirectHandlerSshHint covering SSH_CLIENT,
SSH_TTY, local session (no hint), and missing _oauth_port (no hint).
_configure_provider() calls _run_post_setup() after collecting env vars
(line 2286). _reconfigure_provider() did not — providers with both
env_vars and post_setup (Browserbase, Browser Use, Firecrawl, Camofox)
skipped the installation step on reconfiguration.
Fix: mirror the _configure_provider() call. post_setup hooks are
idempotent (check before installing), so no behaviour change for users
who already have the dependencies installed.
The x_search toolset is gated on xAI credentials (SuperGrok OAuth or
XAI_API_KEY), but it was staying off-by-default even for users who had
already configured those credentials — they had to also click through
`hermes tools` → X (Twitter) Search to flip it on. The HASS_TOKEN →
homeassistant rule already handles the parallel case cleanly; x_search
needs the same treatment.
Why a separate code path from HASS_TOKEN: `ha_*` tools live inside
the `hermes-cli` composite, so the subset-inference loop picks them
up and the HASS branch just unmasks default_off. `x_search` is its
own one-tool toolset NOT in the composite, so the subset loop never
adds it — it has to be injected directly.
* Add `_xai_credentials_present()` — side-effect-free check for stored
xAI OAuth tokens or XAI_API_KEY (dotenv or env). No network.
* In `_get_platform_tools()` else branch (no explicit user config),
inject `x_search` and carve a parallel hole in default_off.
* Auto-enable does NOT fire when the user has saved an explicit toolset
list via `hermes tools` — that list stays authoritative.
* `agent.disabled_toolsets: [x_search]` still wins (global override).
Tests: 4 new in test_tools_config.py covering OAuth path, API-key path,
no-creds path, and explicit-config-respect. All pass alongside existing
70/70 in that file.
The 5-second startup-grace filter in _on_room_message silently drops
events where event_ts < startup_ts - 5. When the host clock is set
ahead of real time, the comparison flips against every live event and
the bot 'connects but never replies' — exactly the symptom in #12614.
Reporter Schnurzel700 chased this for several weeks before tracing it
to their Debian VM's clock being out of sync. The current /1000.0
millisecond->second conversion is correct (mautrix returns ms); the
failure mode is purely environmental.
Add a one-shot WARNING that fires when:
- we are >30s past startup (initial-sync replay window closed), AND
- 3 consecutive drops share the same skew within 60s (a constant
clock offset, not varied-age backfill from an invited room).
State is reset in connect() so reconnects after fixing NTP rearm the
detector. Includes the NTP fix instruction in the warning message
itself and a new Troubleshooting entry in the Matrix docs.
5 new tests cover the happy path, initial-sync backfill, under-
threshold drops, varied-age backfill, and the reconnect rearm path.
Original commit 75e5d0f6b by hueilau targeted _build_api_kwargs in
pre-refactor run_agent.py. The body now lives in
agent/chat_completion_helpers.build_api_kwargs — re-applied there.
Also: switch the custom_providers forward (from 21078ebce) to use
getattr() — tests build a bare AIAgent via __new__ and would otherwise
hit AttributeError on _custom_providers.
Co-authored-by: hueilau <33933019+hueilau@users.noreply.github.com>
Original commit 8d756a421 by austrian_guy targeted __init__ in
pre-refactor run_agent.py. The body now lives in
agent/agent_init.init_agent — re-applied there.
Co-authored-by: austrian_guy <33156212+ether-btc@users.noreply.github.com>
Original commit 973f27e95 by Teknium targeted _spawn_background_review in
pre-refactor run_agent.py. The body now lives in
agent/background_review._spawn_background_review — re-applied there.
Co-authored-by: Teknium <127238744+teknium1@users.noreply.github.com>
Original commit 21078ebce by PaTTeeL targeted _try_activate_fallback in
pre-refactor run_agent.py. The body now lives in
agent/chat_completion_helpers.try_activate_fallback — re-applied there.
Co-authored-by: PaTTeeL <9150277+PaTTeeL@users.noreply.github.com>
Original commit 33528b428 by konsisumer targeted _restore_primary_runtime
in pre-refactor run_agent.py. The body now lives in
agent/agent_runtime_helpers.restore_primary_runtime — re-applied there.
Fixes#20465
Co-authored-by: konsisumer <der@konsi.org>
Original commit 2b193907d by Teknium added a new module-level
_StreamErrorEvent class and threaded its raise into
_run_codex_create_stream_fallback in pre-refactor run_agent.py.
- _StreamErrorEvent class → run_agent.py (module-level, next to
_qwen_portal_headers; class needs to be top-level for the codex
runtime to import it)
- The fallback event-loop's 'type=error' handler → agent/codex_runtime.py
where run_codex_create_stream_fallback now lives. Imports
_StreamErrorEvent lazily from run_agent to avoid circular import.
Co-authored-by: Teknium <127238744+teknium1@users.noreply.github.com>
Original commit e51d74ab9 by Maxim Esipov targeted _extract_api_error_context
and _recover_with_credential_pool in pre-refactor run_agent.py. Both bodies
now live in agent/agent_runtime_helpers.py — re-applied to that module:
- extract_api_error_context: payload.get('type') added to the reason
fallback chain (Codex error bodies use 'type' instead of 'code'/'error')
- recover_with_credential_pool: usage_limit_reached detection in the
rate_limit branch — skip the retry-once-then-rotate dance and rotate
immediately when the body says the per-account usage limit hit.
Co-authored-by: Maxim Esipov <maksesipov@gmail.com>
Original commits 4ded3ede3 (@konsisumer) + 374dc81c2 (Teknium) added a
413 hint to run_agent.py's agent loop. Final-state version (the sharpened
374dc81c2 wording) ported to agent/conversation_loop.py, where the
payload_too_large branch now lives.
The deprecation detection + _URL_TO_PROVIDER changes from both commits
landed in agent/copilot_acp_client.py and agent/model_metadata.py via
the prior merge.
Closes#10648
Co-authored-by: konsisumer <der@konsi.org>
Co-authored-by: Teknium <127238744+teknium1@users.noreply.github.com>
Original commit 395e9dd9e by Teknium targeted module-level _is_mcp_tool_parallel_safe
and _should_parallelize_tool_batch helpers in pre-refactor run_agent.py. Both
helpers now live in agent/tool_dispatch_helpers.py — re-applied to that
module.
The tools/mcp_tool.py portion (the public is_mcp_tool_parallel_safe API
+ _parallel_safe_servers tracking) merged cleanly from main via the prior
merge commit.
Co-authored-by: Teknium <127238744+teknium1@users.noreply.github.com>
Original commit 9c304a7f5 by helix4u targeted _flatten_exception_chain,
_summarize_api_error, and the _call streaming retry loop in pre-refactor
run_agent.py. Re-applied to:
- New _is_provider_stream_parse_error helper → run_agent.py (next
to _flatten_exception_chain in the AIAgent class)
- _summarize_api_error early-return for the malformed-streaming
ValueError → run_agent.py (kept method body)
- _call streaming retry: _is_stream_parse_err flag wired into
_is_transient AND the post-exhaustion branch + dedicated
malformed-streaming user-status string → agent/chat_completion_helpers.py
(the _call body now lives there)
Co-authored-by: helix4u <4317663+helix4u@users.noreply.github.com>
Original commit 97a32afdc by helix4u targeted _check_compression_model_feasibility
in pre-refactor run_agent.py. The function body now lives in
agent/conversation_compression.py — re-applied the configured-but-unavailable
provider message there.
Co-authored-by: helix4u <4317663+helix4u@users.noreply.github.com>
Collapses the four-commit xAI entitlement-403 chain to its final
on-main state, ported to the post-refactor module layout:
- Added _is_entitlement_failure on AIAgent (run_agent.py) — detects
Grok subscription-shape 403s on (401|403|None) status codes.
- Added entitlement-skip branch to recover_with_credential_pool
(agent/agent_runtime_helpers.py) — breaks the refresh-loop that
Don's 100-iteration trace exposed when a Premium+ user hit a real
entitlement issue.
- Removed _decorate_xai_entitlement_error and unwrapped its two
_summarize_api_error call sites — xAI's own body text already
points users at grok.com/?_s=usage so we surface that verbatim
(dffb602f3 reasoning: X Premium subs DO now work per xAI's
2026-05-16 announcement, so editorialising would misdirect).
- grok-4.3 1M context entry landed in agent/model_metadata.py
via the prior merge — no additional port needed.
Tests already on disk (tests/run_agent/test_codex_xai_oauth_recovery.py)
assert _is_entitlement_failure shape and verbatim body surfacing.
Closes#27110.
Co-authored-by: Teknium <127238744+teknium1@users.noreply.github.com>
The original 068c24f8a (DeepSeek thinking via legacy chat_completions path)
was reverted by cd9470f41 (rewired to DeepSeekProfile.build_api_kwargs_extras).
Both commits' run_agent.py edits cancel out at the extracted-module level.
The active fix lives in plugins/model-providers/deepseek/__init__.py
(merged cleanly from main via the prior merge commit).
Co-authored-by: twebefy <twebefy@gmail.com>
Co-authored-by: teknium1 <127238744+teknium1@users.noreply.github.com>
Original commit 31ba2b0cb by Teknium targeted run_codex_stream() at
its pre-refactor location in run_agent.py. Re-applied:
- Prelude error retry/fallback → agent/codex_runtime.py (in
run_codex_stream where the body now lives)
- _decorate_xai_entitlement_error helper + _summarize_api_error
wrapping → run_agent.py (these methods remained on AIAgent
as @staticmethod's; cherry-pick applied them cleanly)
The xai-oauth provider gate, encrypted_content drop on replay, etc.
landed in agent/codex_responses_adapter.py via the prior merge from main.
Closes#8133, #14634
Co-authored-by: Teknium <127238744+teknium1@users.noreply.github.com>
Original commit 13c3d4b4e by kchantharuan touched __init__ and
_apply_client_headers_for_base_url in pre-refactor run_agent.py. Re-applied to:
- __init__: agent/agent_init.py (3 hunks — NVIDIA branch + _custom_headers
fallback in routed-client and fallback-client paths)
- _apply_client_headers_for_base_url: still in run_agent.py (1 hunk)
build_nvidia_nim_headers was already present in agent/auxiliary_client.py
from the prior merge — no additional port needed.
Co-authored-by: kchantharuan <kchantharuan@nvidia.com>
Original commit b62c99797 by Jaaneek targeted six locations in
pre-refactor run_agent.py. Re-applied to the extracted post-PR locations:
- api_mode dispatch → agent/agent_init.py
- is_xai_responses build_api_kwargs → agent/chat_completion_helpers.py
- codex_auth_retry block + 401 hint → agent/conversation_loop.py
- _try_refresh_codex_client_credentials body → run_agent.py (kept)
The non-run_agent.py portions of the commit (auxiliary_client, codex
transport, hermes_cli/auth, tools/xai_http, tests, docs) merged cleanly
from main via the prior merge commit.
Co-authored-by: Jaaneek <Jaaneek@users.noreply.github.com>
Original commit db84a78e6 by kshitij targeted run_conversation()'s
pre_api_request and post_api_request hooks in pre-refactor run_agent.py.
Re-applied to the extracted location in agent/conversation_loop.py.
Co-authored-by: kshitij <82637225+kshitijk4poor@users.noreply.github.com>
Co-authored-by: xxxigm <tuancanhnguyen706@gmail.com>
Co-authored-by: Brian Conklin <brian@dralth.com>
Original commit 4f8aaf104 by InB4DevOps targeted run_conversation() in
the pre-refactor run_agent.py. Re-applied to the extracted location in
agent/conversation_loop.py.
Co-authored-by: InB4DevOps <tolle.lege+github@gmail.com>
previously only checked provider ID and
base URL. When kimi-k2.6 is served via ollama-cloud (or any third-party
provider), provider is not 'kimi-coding' and base URL is not
api.kimi.com — so reasoning_content pad was never injected. This caused
HTTP 400 from Ollama Cloud's Go backend: 'invalid message content type:
map[string]interface {}'.
Fix: add model-name detection ('kimi' in model.lower()) so any route
serving a kimi model gets the required reasoning_content echo-back.
Refs the 400/401 Telegram errors where kimi-k2.6 via ollama-cloud
consistently failed after tool-call turns.
(cherry picked from commit 9a9f8a6d99)
run_agent.py taken from HEAD (the extracted forwarder structure). The 25
run_agent.py fixes that landed on main during the PR's life need to be
ported into the agent/* extracted modules in follow-up commits.
_LineClient's five aiohttp.ClientSession() calls omit trust_env=True,
silently bypassing HTTP_PROXY / HTTPS_PROXY / ALL_PROXY. Result: every
LINE API call (reply, push, loading, fetch_content, get_bot_user_id)
ignores the system proxy.
Fix: add trust_env=True to all five session constructions. Symmetric
with the wecom and weixin adapters which already set this flag. No
behavior change for users not behind a proxy.
_OPENROUTER_MODEL hardcoded 'google/gemini-3-flash-preview' which
returns 404 on OpenRouter, breaking all vision tasks for users who
rely on the OpenRouter default. Additionally, _try_openrouter()
ignored the user-configured auxiliary.vision.model entirely.
Changes:
- Update _OPENROUTER_MODEL default to google/gemini-2.5-flash (valid)
- Add optional 'model' parameter to _try_openrouter()
- Pass configured model from _resolve_strict_vision_backend() through
to _try_openrouter()
This allows users who set auxiliary.vision.model (e.g. x-ai/grok-4.3)
to have it actually used, while maintaining backward compatibility.
In resolve_provider_client(), the named custom provider code path at
~line 2914 only checked the ``key_env`` field when looking for an
environment-variable-based API key. The documented ``api_key_env``
snake_case alias was silently ignored, causing custom providers
configured with ``api_key_env`` to fall through to the
``no-key-required`` placeholder — which produces a confusing 401
(``****ired`` mask) on auth-required remote endpoints.
This mirrors the same fix already applied to run_agent.py in commit
6ddc48b05 (fix(fallback): resolve api_key_env in fallback chain entries).
Also adds a logger.warning() when the placeholder is reached, so
future alias gaps are easier to debug.
Closes#25091
Refactor the inlined `re.sub(...)[:4000].strip()` cleanup at the
auto-TTS site in `_process_message_background` into an overridable
method `BasePlatformAdapter.prepare_tts_text(text: str) -> str`.
The default implementation is byte-identical to the previous inline
expression — strip `* _ \` # [ ] ( )` and truncate to 4000 chars — so
every existing adapter (Telegram, Discord, Slack, Matrix, IRC, etc.)
gets exactly the same behaviour as before. Zero behaviour change for
any consumer that doesn't override the method.
Why add the hook: voice-first platform adapters need stricter
cleanup than text-bubble platforms. The default strips a handful of
markdown sigils, which is fine when the output goes into a Discord
embed or a Telegram message bubble — but read aloud by a TTS engine,
URLs (`https://example.com/foo`), fenced code blocks, file paths
(`/Users/x/foo.py`), and `MEDIA:` tags turn into long sequences of
unintelligible characters. With this hook an adapter can drop those
spans before TTS while leaving the data-channel transcript intact
for visual rendering.
Without the hook, voice adapters have to either
- duplicate the auto-TTS flow inside their own `handle_response`
pipeline, which means re-implementing the entire `extract_media`,
`extract_images`, `extract_local_files`, attachment routing and
error-handling sequence in `_process_message_background`, or
- live with TTS speaking URLs character-by-character.
Both are worse than a 7-line method addition.
Example consumer:
https://github.com/kortexa-ai/hermes-livekit — LiveKit WebRTC voice
gateway plugin. Its `LiveKitAdapter.prepare_tts_text()` additionally
strips fenced code blocks, inline code, URLs, file paths, and
`MEDIA:` tags before TTS synthesis, while the full response still
reaches connected clients via the data channel. Drop-in installable
via `pip install git+https://github.com/kortexa-ai/hermes-livekit.git`.
Carved out of #3894 (LiveKit WebRTC gateway PR) so the generic hook
can land independently of the LiveKit platform itself.
aiohttp.ClientSession defaults to trust_env=False, which silently ignores
HTTP_PROXY, HTTPS_PROXY, and ALL_PROXY environment variables. Users behind
a corporate or network proxy cannot reach external APIs on any of these
platforms — all outbound requests fail with connection errors.
Symmetric with wecom.py (line 276), weixin.py (lines 1055/1268/1274), and
matrix.py (no-proxy path) which already set this flag. Complements the
open LINE fix (#26635) with the remaining gateway and plugin adapters.
Changed:
- gateway/platforms/sms.py: persistent Twilio session (connect) + fallback
session (send) — both hit https://api.twilio.com
- gateway/platforms/slack.py: ephemeral response_url POST session —
hits https://hooks.slack.com/... callback URLs
- plugins/platforms/teams/adapter.py: standalone send session —
hits login.microsoftonline.com (token) + Bot Framework service URL
- plugins/platforms/google_chat/adapter.py: standalone send session —
hits https://chat.googleapis.com/v1/...
WhatsApp sessions are excluded: they connect to http://127.0.0.1:{port}
(local bridge) and must not be routed through a system proxy.
The check-windows-footguns.py script outputs a checkmark (U+2713) and
cross (U+2717) to report results. Windows terminals default to cp1252,
which cannot encode these characters, so running the script on Windows
threw a UnicodeEncodeError before any results were printed.
This made the tool completely unusable on the exact platform it exists
to help -- a developer on Windows trying to check their code for
Windows-safety issues would just get a crash instead.
Fix: reconfigure stdout and stderr to UTF-8 at the start of main(),
before any output is produced. Verified on Windows 11 Home with
Python 3.13 (terminal defaulting to cp1252).
Tests in TestReadClaudeCodeCredentials were not mocking
_read_claude_code_credentials_from_keychain, which was added after the
tests were written. On macOS machines with real Claude Code credentials
stored in the Keychain, the function returns live credentials instead of
the test fixtures, causing assertions to fail and leaking real tokens in
test output.
Add an autouse fixture that stubs the keychain reader to None so all
tests in the class exercise only the file-based credential path.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replace generator-based result collection with explicit per-future
handling. Each future is now processed independently with a 600s timeout.
Before: _results.extend(f.result() for f in _futures)
- One exception stops the generator, remaining results are lost
- No timeout: one hung job blocks the entire tick
After: as_completed() + per-future try/except
- Each future handled independently
- 600s timeout prevents indefinite blocking
- Failed futures are logged and counted as failures
The goal judge only receives the goal text and the agent's last
response. It has no concept of the current time, making it
impossible to evaluate time-sensitive goals like 'keep working
until 5pm'.
This commit adds 'Current time' to both JUDGE_USER_PROMPT_TEMPLATE
and JUDGE_USER_PROMPT_WITH_SUBGOALS_TEMPLATE, computed from
datetime.now().astimezone() at judge call time.
The Telegram/Discord model picker skipped live model discovery for
custom providers (llama.cpp, Ollama) unless an api_key was configured.
Local providers typically don't require auth on the /models endpoint.
The CLI always probes /models, so this brings the gateway picker into
parity.
Change: `if api_url and api_key:` -> `if api_url:`
Add creationflags=CREATE_NO_WINDOW to every Windows Popen call
across the terminal, process registry, code execution, and kanban
worker subsystems. Prevents visible CMD windows from flashing on
the user's desktop during agent operation.
Also adds the _IS_WINDOWS module constant to kanban_db.py where
it was missing, for consistency with the other patched files.
5 Popen sites across 4 files:
- tools/environments/local.py (terminal foreground spawn)
- tools/process_registry.py (background process spawn)
- tools/code_execution_tool.py (sandbox + interpreter probe)
- hermes_cli/kanban_db.py (kanban worker spawn)
When /goal loop generates synthetic MessageEvents (goal continuations,
status notices), the reply anchor is unavailable (message_id=None). For
Telegram DM topic lanes, the Telegram adapter requires
direct_messages_topic_id to route messages correctly; without it, the
adapter falls back to message_thread_id=None, sending messages to the
root 'All Messages' thread instead of the active topic lane.
The fix includes direct_messages_topic_id in thread metadata for all
non-General Telegram DM topics, ensuring queued/synthetic messages are
delivered to the correct thread even when no reply anchor exists.
_propare_messages_for_non_vision_model() was only called in the legacy
flag path (no provider profile). Providers with registered profiles
(e.g. DeepSeek, Kimi) bypassed the strip, causing HTTP 400 errors when
image_url content blocks reached their non-vision APIs.
This mirrors the existing behavior in the legacy path, ensuring all
non-vision models get image stripping regardless of profile status.
Vision-capable models are unaffected (the function is a no-op for them).
The _normalize_custom_provider_entry() function was dropping the
discover_models field from custom_provider entries because:
1. It was not listed in _KNOWN_KEYS, so it was logged as an
unknown key and ignored.
2. The function builds the normalized dict by explicitly copying
known fields, so even if the warning was suppressed, the value
was not carried through.
This caused downstream model_switch.py to default discover_models
to True, triggering /models HTTP probes on unreachable endpoints.
With 4 unreachable internal endpoints at ~6s timeout each, the
/api/model/options endpoint took ~24s instead of <1s.
