_deliver_result() now returns Optional[str] — None on success, error
message on failure. All failure paths (unknown platform, platform
disabled, config load error, send failure, unresolvable target)
return descriptive error strings.
mark_job_run() gains delivery_error param, tracked as
last_delivery_error on the job — separate from agent execution errors.
A job where the agent succeeded but delivery failed shows
last_status='ok' + last_delivery_error='...'.
The cronjob list tool now surfaces last_delivery_error so agents and
users can see when cron outputs aren't arriving.
Inspired by PR #5863 (oxngon) — reimplemented with proper wiring.
Tests: 3 new mark_job_run tests + 6 new _deliver_result return tests.
Add button-based exec approval to the Feishu adapter, matching the
existing Discord, Telegram, and Slack implementations.
When the agent encounters a dangerous command, Feishu users now see
an interactive card with four buttons instead of text instructions:
- Allow Once (primary)
- Allow Session
- Always Allow
- Deny (danger)
Implementation:
- send_exec_approval() sends an interactive card via the Feishu
message API with buttons carrying hermes_action in their value dict
- _handle_card_action_event() intercepts approval button clicks
before routing them as synthetic commands, directly calling
resolve_gateway_approval() to unblock the agent thread
- _update_approval_card() replaces the orange approval card with a
green (approved) or red (denied) status card showing who acted
- _approval_state dict tracks pending approval_id → session_key
mappings; cleaned up on resolution
The gateway's existing routing in _approval_notify_sync already checks
getattr(type(adapter), 'send_exec_approval', None) and will
automatically use the button-based flow for Feishu.
Tests: 16 new tests covering send, callback resolution, state
management, card updates, and non-interference with existing card
actions.
Salvaged fixes from community PRs:
- fix(model_switch): _read_auth_store → _load_auth_store + fix auth store
key lookup (was checking top-level dict instead of store['providers']).
OAuth providers now correctly detected in /model picker.
Cherry-picked from PR #5911 by Xule Lin (linxule).
- fix(ollama): pass num_ctx to override 2048 default context window.
Ollama defaults to 2048 context regardless of model capabilities. Now
auto-detects from /api/show metadata and injects num_ctx into every
request. Config override via model.ollama_num_ctx. Fixes#2708.
Cherry-picked from PR #5929 by kshitij (kshitijk4poor).
- fix(aux): normalize provider aliases for vision/auxiliary routing.
Adds _normalize_aux_provider() with 17 aliases (google→gemini,
claude→anthropic, glm→zai, etc). Fixes vision routing failure when
provider is set to 'google' instead of 'gemini'.
Cherry-picked from PR #5793 by e11i (Elizabeth1979).
- fix(aux): rewrite MiniMax /anthropic base URLs to /v1 for OpenAI SDK.
MiniMax's inference_base_url ends in /anthropic (Anthropic Messages API),
but auxiliary client uses OpenAI SDK which appends /chat/completions →
404 at /anthropic/chat/completions. Generic _to_openai_base_url() helper
rewrites terminal /anthropic to /v1 for OpenAI-compatible endpoint.
Inspired by PR #5786 by Lempkey.
Added debug logging to silent exception blocks across all fixes.
Co-authored-by: Hermes Agent <hermes@nousresearch.com>
Free-tier Nous Portal users were getting mimo-v2-omni (a multimodal
model) for all auxiliary tasks including compression, session search,
and web extraction. Now routes non-vision tasks to mimo-v2-pro (a
text model) which is better suited for those workloads.
- Added _NOUS_FREE_TIER_AUX_MODEL constant for text auxiliary tasks
- _try_nous() accepts vision=False param to select the right model
- Vision path (_resolve_strict_vision_backend) passes vision=True
- All other callers default to vision=False → mimo-v2-pro
Currently, MCP servers are included on all platforms by default. If a
platform's toolset list does not explicitly name any MCP servers, every
globally enabled MCP server is injected. There is no way to opt a
platform out of MCP servers entirely.
This matters for the API server platform when used as an execution
backend — each spawned agent session gets the full MCP tool schema
injected into its system prompt, dramatically inflating token usage
(e.g. 57K tokens vs 9K without MCP tools) and slowing response times.
Add a "no_mcp" sentinel value for platform_toolsets. When present in a
platform's toolset list, all MCP servers are excluded for that platform.
Other platforms are unaffected.
Usage in config.yaml:
platform_toolsets:
api_server:
- terminal
- file
- web
- no_mcp # exclude all MCP servers
The sentinel is filtered out of the final toolset — it does not appear
as an actual toolset name.
- The MCP SDK Pydantic model uses camelCase (structuredContent), not
snake_case (structured_content). The original getattr was a silent no-op.
- When structuredContent is present, return it AS the result instead of
alongside text — the structured payload is the machine-readable data.
- Move test file to tests/tools/ and fix fake class to use camelCase.
- Patch _run_on_mcp_loop in tests so the handler actually executes.
MCP CallToolResult may include structured_content (a JSON object) alongside
content blocks. The tool handler previously only forwarded concatenated text
from content blocks, silently dropping the structured payload.
This breaks MCP tools that return a minimal human text in content while
putting the actual machine-usable payload in structured_content.
Now, when structured_content is present, it is included in the returned
JSON under the 'structuredContent' key.
FixesNousResearch/hermes-agent#5874
* fix(telegram): replace substring caption check with exact line-by-line match
Captions in photo bursts and media group albums were silently dropped when
a shorter caption happened to be a substring of an existing one (e.g.
"Meeting" lost inside "Meeting agenda"). Extract a shared _merge_caption
static helper that splits on "\n\n" and uses exact match with whitespace
normalisation, then use it in both _enqueue_photo_event and
_queue_media_group_event.
Adds 13 unit tests covering the fixed bug scenarios.
Cherry-picked from PR #2671 by Dilee.
* fix: extend caption substring fix to all platforms
Move _merge_caption helper from TelegramAdapter to BasePlatformAdapter
so all adapters inherit it. Fix the same substring-containment bug in:
- gateway/platforms/base.py (photo burst merging)
- gateway/run.py (priority photo follow-up merging)
- gateway/platforms/feishu.py (media batch merging)
The original fix only covered telegram.py. The same bug existed in base.py
and run.py (pure substring check) and feishu.py (list membership without
whitespace normalization).
* fix(auxiliary): resolve named custom providers and 'main' alias in auxiliary routing
Two bugs caused auxiliary tasks (vision, compression, etc.) to fail when
using named custom providers defined in config.yaml:
1. 'provider: main' was hardcoded to 'custom', which only checks legacy
OPENAI_BASE_URL env vars. Now reads _read_main_provider() to resolve
to the actual provider (e.g., 'custom:beans', 'openrouter', 'deepseek').
2. Named custom provider names (e.g., 'beans') fell through to
PROVIDER_REGISTRY which doesn't know about config.yaml entries.
Now checks _get_named_custom_provider() before the registry fallback.
Fixes both resolve_provider_client() and _normalize_vision_provider()
so the fix covers all auxiliary tasks (vision, compression, web_extract,
session_search, etc.).
Adds 13 unit tests. Reported by Laura via Discord.
---------
Co-authored-by: Dilee <uzmpsk.dilekakbas@gmail.com>
* fix(cli): route error messages through ChatConsole inside patch_stdout
Cherry-pick of PR #5798 by @icn5381.
Replace self.console.print() with ChatConsole().print() for 11 error/status
messages reachable during the interactive session. Inside patch_stdout,
self.console (plain Rich Console) writes raw ANSI escapes that StdoutProxy
mangles into garbled text. ChatConsole uses prompt_toolkit's native
print_formatted_text which renders correctly.
Same class of bug as #2262 — that fix covered agent output but missed
these error paths in _ensure_runtime_credentials, _init_agent, quick
commands, skill loading, and plan mode.
* fix(model-picker): add scrolling viewport to curses provider menu
Cherry-pick of PR #5790 by @Lempkey. Fixes#5755.
_curses_prompt_choice rendered items starting unconditionally from index 0
with no scroll offset. The 'More providers' submenu has 13 entries. On
terminals shorter than ~16 rows, items past the fold were never drawn.
When UP-arrow wrapped cursor from 0 to the last item (Cancel, index 12),
the highlight rendered off-screen — appearing as if only Cancel existed.
Adds scroll_offset tracking that adjusts each frame to keep the cursor
inside the visible window.
* feat(cli): skin-aware compact banner + git state in startup banner
Combined salvage of PR #5922 by @ASRagab and PR #5877 by @xinbenlv.
Compact banner changes (from #5922):
- Read active skin colors and branding instead of hardcoding gold/NOUS HERMES
- Default skin preserves backward-compatible legacy branding
- Non-default skins use their own agent_name and colors
Git state in banner (from #5877):
- New format_banner_version_label() shows upstream/local git hashes
- Full banner title now includes git state (upstream hash, carried commits)
- Compact banner line2 shows the version label with git state
- Widen compact banner max width from 64 to 88 to fit version info
Both the full Rich banner and compact fallback are now skin-aware
and show git state.
- Add 7 unit tests for _profile_arg: default home, named profile,
hash path, nested path, invalid name, systemd integration, launchd integration
- Add stt.local.language to config.yaml (empty = auto-detect)
- Both STT code paths now read config.yaml first, env var fallback,
then default (auto-detect for faster-whisper, 'en' for CLI command)
- HERMES_LOCAL_STT_LANGUAGE env var still works as backward-compat fallback
- gateway/run.py: Strip "(The user sent a message with no text content)"
placeholder when voice transcription succeeds — it was being appended
alongside the transcript, creating duplicate user turns.
- tools/transcription_tools.py: Wire HERMES_LOCAL_STT_LANGUAGE env var
into the faster-whisper backend. It was only used by the CLI fallback
path (_transcribe_local_command), not the primary faster-whisper path.
When a Discord voice message arrives, the adapter sets event.text to
"(The user sent a message with no text content)" since voice messages
have no text content. The transcription enrichment in
_enrich_message_with_transcription() then prepends the transcript but
leaves the placeholder intact, causing the agent to receive both:
[The user sent a voice message~ Here's what they said: "..."]
(The user sent a message with no text content)
The agent sees this as two separate user turns — one transcribed
and one empty — creating confusing duplicate messages.
Fix: when the transcription succeeds and user_text is only the empty
placeholder, return just the transcript without the redundant placeholder.
generate_launchd_plist() and generate_systemd_unit() were missing the
--profile <name> argument in ProgramArguments/ExecStart, causing
hermes gateway start to regenerate plists that fell back to
~/.hermes/active_profile instead of the intended profile.
Fix:
- Add _profile_arg(hermes_home?) helper returning '--profile <name>'
only for ~/.hermes/profiles/<name> paths, empty string otherwise.
- Update generate_launchd_plist() to build ProgramArguments array
dynamically with --profile when applicable.
- Update generate_systemd_unit() both user and system service
branches with {profile_arg} in ExecStart.
This ensures hermes --profile <name> gateway start produces a
service definition that correctly scopes to the named profile.
Follow-up to sroecker's PR #5918 — test mocks were using the old 3-arg
callback signature (name, preview, args) instead of the new
(event_type, name, preview, args, **kwargs).
When /v1/runs receives an OpenAI-style array of messages as input, all
messages except the last user turn are now extracted as conversation_history.
Previously only the last message was kept, silently discarding earlier
context in multi-turn conversations.
Handles multi-part content blocks by flattening text portions. Only fires
when no explicit conversation_history was provided.
Based on PR #5837 by pradeep7127.
Allow clients to pass explicit conversation_history in /v1/responses and
/v1/runs request bodies instead of relying on server-side response chaining
via previous_response_id. Solves problems with stateless deployments where
the in-memory ResponseStore is lost on restart.
Adds input validation (must be array of {role, content} objects) and clear
precedence: explicit conversation_history > previous_response_id.
Based on PR #5805 by VanBladee, with added input validation.
Commit cc2b56b2 changed the tool_progress_callback signature from
(name, preview, args) to (event_type, name, preview, args, **kwargs)
but the API server's chat completion streaming callback was not updated.
This caused tool calls to not display in Open WebUI because the
callback received arguments in wrong positions.
- Update _on_tool_progress to use new 4-arg signature
- Add event_type filter to only show tool.started events
- Add **kwargs for optional duration/is_error parameters
Telegram's set_message_reaction replaces all reactions in one call,
so _remove_reaction was never called (unlike Discord's additive model).
Default reactions to disabled — users opt in via telegram.reactions: true.
Mirror the Discord reaction pattern for Telegram:
- 👀 (eyes) when message processing begins
- ✅ (check) on successful completion
- ❌ (cross) on failure
Controlled via TELEGRAM_REACTIONS env var or telegram.reactions
in config.yaml (enabled by default, like Discord).
Uses python-telegram-bot's Bot.set_message_reaction() API.
Failures are caught and logged at debug level so they never
break message processing.
- ignored_channels: channels where bot never responds (even when mentioned)
- no_thread_channels: channels where bot responds directly without thread
Both support config.yaml and env vars (DISCORD_IGNORED_CHANNELS,
DISCORD_NO_THREAD_CHANNELS), following existing pattern for
free_response_channels.
Fixes#5881
The response validation stage unconditionally marked Codex Responses API
replies as invalid when response.output was empty, triggering unnecessary
retries and fallback chains. However, _normalize_codex_response can
recover from this state by synthesizing output from response.output_text.
Now the validation stage checks for output_text before marking the
response invalid, matching the normalization logic. Also fixes
logging.warning → logger.warning for consistency with the rest of the
file.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace 10 callsites across 6 files that manually opened config.yaml,
called yaml.safe_load(), and handled missing-file/parse-error fallbacks
with the new read_raw_config() helper from hermes_cli/config.py.
Each migrated site previously had 5-8 lines of boilerplate:
config_path = get_hermes_home() / 'config.yaml'
if config_path.exists():
import yaml
with open(config_path) as f:
cfg = yaml.safe_load(f) or {}
Now reduced to:
from hermes_cli.config import read_raw_config
cfg = read_raw_config()
Migrated files:
- tools/browser_tool.py (4 sites): command_timeout, cloud_provider,
allow_private_urls, record_sessions
- tools/env_passthrough.py: terminal.env_passthrough
- tools/credential_files.py: terminal.credential_files
- tools/transcription_tools.py: stt.model
- hermes_cli/commands.py: config-gated command resolution
- hermes_cli/auth.py (2 sites): model config read + provider reset
Skipped (intentionally):
- gateway/run.py: 10+ sites with local aliases, critical path
- hermes_cli/profiles.py: profile-specific config path
- hermes_cli/doctor.py: reads raw then writes fixes back
- agent/model_metadata.py: different file (context_length_cache.yaml)
- tools/website_policy.py: custom config_path param + error types
The webhook adapter stored per-request `deliver`/`deliver_extra` config in
`_delivery_info[chat_id]` during POST handling and consumed it via `.pop()`
inside `send()`. That worked for routes whose agent run produced exactly
one outbound message — the final response — but it broke whenever the
agent emitted any interim status message before the final response.
Status messages flow through the same `send(chat_id, ...)` path as the
final response (see `gateway/run.py::_status_callback_sync` →
`adapter.send(...)`). Common triggers include:
- "🔄 Primary model failed — switching to fallback: ..."
(run_agent.py::_emit_status when `fallback_providers` activates)
- context-pressure / compression notices
- any other lifecycle event routed through `status_callback`
When any of those fired, the first `send()` call popped the entry, so the
subsequent final-response `send()` saw an empty dict and silently
downgraded `deliver_type` from `"telegram"` (or `discord`/`slack`/etc.) to
the default `"log"`. The agent's response was logged to the gateway log
instead of being delivered to the configured cross-platform target — no
warning, no error, just a missing message.
This was easy to hit in practice. Any user with `fallback_providers`
configured saw it the first time their primary provider hiccuped on a
webhook-triggered run. Routes that worked perfectly in dev (where the
primary stays healthy) silently dropped responses in prod.
Fix: read `_delivery_info` with `.get()` so multiple `send()` calls for
the same `chat_id` all see the same delivery config. To keep the dict
bounded without relying on per-send cleanup, add a parallel
`_delivery_info_created` timestamp dict and a `_prune_delivery_info()`
helper that drops entries older than `_idempotency_ttl` (1h, same window
already used by `_seen_deliveries`). Pruning runs on each POST, mirroring
the existing `_seen_deliveries` cleanup pattern.
Worst-case memory footprint is now `rate_limit * TTL = 30/min * 60min =
1800` entries, each ~1KB → under 2 MB. In practice it'll be far smaller
because most webhooks complete in seconds, not the full hour.
Test changes:
- `test_delivery_info_cleaned_after_send` is replaced with
`test_delivery_info_survives_multiple_sends`, which is now the
regression test for this bug — it asserts that two consecutive
`send()` calls both see the delivery config.
- A new `test_delivery_info_pruned_via_ttl` covers the TTL cleanup
behavior.
- The two integration tests that asserted `chat_id not in
adapter._delivery_info` after `send()` now assert the opposite, with
a comment explaining why.
All 40 tests in `tests/gateway/test_webhook_adapter.py` and
`tests/gateway/test_webhook_integration.py` pass. Verified end-to-end
locally against a dynamic `hermes webhook subscribe` route configured
with `--deliver telegram --deliver-chat-id <user>`: with `gpt-5.4` as
the primary (currently flaky) and `claude-opus-4.6` as the fallback,
the fallback notification fires, the agent finishes, and the final
response is delivered to Telegram as expected.
* refactor: re-architect tests to mirror the codebase
* Update tests.yml
* fix: add missing tool_error imports after registry refactor
* fix(tests): replace patch.dict with monkeypatch to prevent env var leaks under xdist
patch.dict(os.environ) can leak TERMINAL_ENV across xdist workers,
causing test_code_execution tests to hit the Modal remote path.
* fix(tests): fix update_check and telegram xdist failures
- test_update_check: replace patch("hermes_cli.banner.os.getenv") with
monkeypatch.setenv("HERMES_HOME") — banner.py no longer imports os
directly, it uses get_hermes_home() from hermes_constants.
- test_telegram_conflict/approval_buttons: provide real exception classes
for telegram.error mock (NetworkError, TimedOut, BadRequest) so the
except clause in connect() doesn't fail with "catching classes that do
not inherit from BaseException" when xdist pollutes sys.modules.
* fix(tests): accept unavailable_models kwarg in _prompt_model_selection mock
Previously, all/new tool progress modes always hard-truncated previews
to 40 chars, ignoring the display.tool_preview_length config. This made
it impossible for gateway users to see meaningful command/path info
without switching to verbose mode (which shows too much detail).
Now all/new modes read tool_preview_length from config:
- tool_preview_length: 0 (default/unset) → 40 chars (no regression)
- tool_preview_length: 120 → 120-char previews in all/new mode
- verbose mode: unchanged (already respected the config)
Users who want longer previews can set:
display:
tool_preview_length: 120
Reported by demontut_ on Discord.
Move _merge_caption helper from TelegramAdapter to BasePlatformAdapter
so all adapters inherit it. Fix the same substring-containment bug in:
- gateway/platforms/base.py (photo burst merging)
- gateway/run.py (priority photo follow-up merging)
- gateway/platforms/feishu.py (media batch merging)
The original fix only covered telegram.py. The same bug existed in base.py
and run.py (pure substring check) and feishu.py (list membership without
whitespace normalization).
Captions in photo bursts and media group albums were silently dropped when
a shorter caption happened to be a substring of an existing one (e.g.
"Meeting" lost inside "Meeting agenda"). Extract a shared _merge_caption
static helper that splits on "\n\n" and uses exact match with whitespace
normalisation, then use it in both _enqueue_photo_event and
_queue_media_group_event.
Adds 13 unit tests covering the fixed bug scenarios.
Cherry-picked from PR #2671 by Dilee.
Based on PR #5413 spec by MaheshtheDev (Mahesh Sanikommu).
Changes:
- Add search_mode config (hybrid/memories/documents) passed to SDK
- Add {identity} template support in container_tag for profile-scoped containers
- Add SUPERMEMORY_CONTAINER_TAG env var override (priority over config)
- Add multi-container mode: enable_custom_container_tags, custom_containers,
custom_container_instructions in supermemory.json
- Dynamic tool schemas when multi-container enabled (optional container_tag param)
- Whitelist validation for custom container tags in tool calls
- Simplify get_config_schema() to only prompt for API key during setup
- Defer container_tag sanitization to initialize() (after template resolution)
- Add custom_id support to documents.add calls
- Update README with multi-container docs, search_mode, identity template,
support links (Discord, email)
- Update memory-providers.md with new features and multi-container example
- Update memory-provider-plugin.md with minimal vs full schema guidance
- Add 12 new tests covering identity template, search_mode, multi-container,
config schema, and env var override
When the model produces structured reasoning (via API fields like .reasoning,
.reasoning_content, .reasoning_details) but no visible text content, append
the assistant message as prefill and continue the loop. The model sees its own
reasoning context on the next turn and produces the text portion.
Inspired by clawdbot's 'incomplete-text' recovery pattern. Up to 2 prefill
attempts before falling through to the existing '(empty)' terminal.
Key design decisions:
- Only triggers for structured reasoning (API fields), NOT inline <think> tags
- Prefill messages are popped on success to maintain strict role alternation
- _thinking_prefill marker stripped from all API message building paths
- Works across all providers: OpenAI (continuation), Anthropic (native prefill)
Verified with E2E tests: simulated thinking-only → real OpenRouter continuation
produces correct content. Also confirmed Qwen models consistently produce
structured-reasoning-only responses under token pressure.
Move _PS_CHECK_IMAGE and _PS_EXTRACT_IMAGE above both the native Windows
and WSL2 sections so both can share them. Removes the duplicate
_WIN_PS_CHECK / _WIN_PS_EXTRACT constants.
Add win32 platform branch to clipboard.py so Ctrl+V image paste
works on native Windows (PowerShell / Windows Terminal), not just
WSL2.
Uses the same .NET System.Windows.Forms.Clipboard approach as the
WSL path but calls PowerShell directly instead of powershell.exe
(the WSL cross-call path). Tries 'powershell' first (Windows
PowerShell 5.1, always available), then 'pwsh' (PowerShell 7+).
PowerShell executable is discovered once and cached for the process
lifetime.
Includes 14 new tests covering:
- Platform dispatch (save_clipboard_image + has_clipboard_image)
- Image detection via PowerShell .NET check
- Base64 PNG extraction and decode
- Edge cases: no PowerShell, empty output, invalid base64, timeout
Adapted from PR #5679 (0xbyt4) to cover edge cases not in the integration tests:
video routing, unknown extension fallback to send_document, multi-file delivery,
and single-failure isolation.
The cron delivery path sent raw 'MEDIA:/path/to/file' text instead
of uploading the file as a native attachment. The standalone path
(via _send_to_platform) already extracted MEDIA tags and forwarded
them as media_files, but the live adapter path passed the unprocessed
delivery_content directly to adapter.send().
Two bugs fixed:
1. Live adapter path now sends cleaned text (MEDIA tags stripped)
instead of raw content — prevents 'MEDIA:/path' from appearing
as literal text in Discord/Telegram/etc.
2. Live adapter path now sends each extracted media file via the
adapter's native method (send_voice for audio, send_image_file
for images, send_video for video, send_document as fallback) —
files are uploaded as proper platform attachments.
The file-type routing mirrors BasePlatformAdapter._process_message_background
to ensure consistent behavior between normal gateway responses and
cron-delivered responses.
Adds 2 tests:
- test_live_adapter_sends_media_as_attachments: verifies Discord
adapter receives send_voice call for .mp3 file
- test_live_adapter_sends_cleaned_text_not_raw: verifies MEDIA tag
stripped from text sent via live adapter
Memory plugins (Mem0, Honcho) used static identifiers ('hermes-user',
config peerName) meaning all gateway users shared the same memory bucket.
Changes:
- AIAgent.__init__: add user_id parameter, store as self._user_id
- run_agent.py: include user_id in _init_kwargs passed to memory providers
- gateway/run.py: pass source.user_id to AIAgent in primary + background paths
- Mem0 plugin: prefer kwargs user_id over config default
- Honcho plugin: override cfg.peer_name with gateway user_id when present
CLI sessions (user_id=None) preserve existing defaults. Only gateway
sessions with a real platform user_id get per-user memory scoping.
Reported by plev333.
When the bot sends a message in a thread, track its ts in _bot_message_ts.
When the bot is @mentioned in a thread, register it in _mentioned_threads.
Both sets enable auto-responding to future messages in those threads
without requiring repeated @mentions — making the bot behave like a
team member that stays engaged once a conversation starts.
Channel message gating now checks 4 signals (in order):
1. @mention in this message
2. Reply in a thread the bot started/participated in (_bot_message_ts)
3. Message in a thread where the bot was previously @mentioned (_mentioned_threads)
4. Existing session for this thread (_has_active_session_for_thread — survives restarts)
Thread context fetching now triggers on ANY first-entry path (not just
@mention), so the agent gets context whether it's entering via a mention,
a bot-thread reply, or a mentioned-thread auto-trigger.
Both tracking sets are bounded (5000 cap with prune-oldest-half) to prevent
unbounded memory growth in long-running deployments.
Salvaged from PR #5754 by @hhhonzik. Preserves our existing approval buttons,
thread context fetching, and session key fix. Does NOT include the
edit_message format_message() removal (that was a regression in the original PR).
Tests: 4 new tests for bot-ts tracking and mentioned-thread bounds.
When the agent waits for dangerous-command approval, the typing
indicator (_keep_typing loop) kept refreshing. On Slack's Assistant
API this is critical: assistant_threads_setStatus disables the
compose box, preventing users from typing /approve or /deny.
- Add _typing_paused set + pause/resume methods to BasePlatformAdapter
- _keep_typing skips send_typing when chat_id is paused
- _approval_notify_sync pauses typing before sending approval prompt
- _handle_approve_command / _handle_deny_command resume typing after
Benefits all platforms — no reason to show 'is thinking...' while
the agent is idle waiting for human input.
Slack:
- Add Block Kit interactive buttons for command approval (Allow Once,
Allow Session, Always Allow, Deny) via send_exec_approval()
- Register @app.action handlers for each approval button
- Add _fetch_thread_context() — fetches thread history via
conversations.replies when bot is first @mentioned mid-thread
- Fix _has_active_session_for_thread() to use build_session_key()
instead of manual key construction (fixes session key mismatch bug
where thread_sessions_per_user flag was ignored, ref PR #5833)
Telegram:
- Add InlineKeyboard approval buttons via send_exec_approval()
- Add ea:* callback handling in _handle_callback_query()
- Uses monotonic counter + _approval_state dict to map button clicks
back to session keys (avoids 64-byte callback_data limit)
Both platforms now auto-detected by the gateway runner's
_approval_notify_sync() — any adapter with send_exec_approval() on
its class gets button-based approval instead of text fallback.
Inspired by community PRs #3898 (LevSky22), #2953 (ygd58), #5833
(heathley). Implemented fresh on current main.
Tests: 24 new tests covering button rendering, action handling,
thread context fetching, session key fix, double-click prevention.
16 callsites across 14 files were re-deriving the hermes home path
via os.environ.get('HERMES_HOME', ...) instead of using the canonical
get_hermes_home() from hermes_constants. This breaks profiles — each
profile has its own HERMES_HOME, and the inline fallback defaults to
~/.hermes regardless.
Fixed by importing and calling get_hermes_home() at each site. For
files already inside the hermes process (agent/, hermes_cli/, tools/,
gateway/, plugins/), this is always safe. Files that run outside the
process context (mcp_serve.py, mcp_oauth.py) already had correct
try/except ImportError fallbacks and were left alone.
Skipped: hermes_constants.py (IS the implementation), env_loader.py
(bootstrap), profiles.py (intentionally manipulates the env var),
standalone scripts (optional-skills/, skills/), and tests.
Comprehensive cleanup across 80 files based on automated (ruff, pyflakes, vulture)
and manual analysis of the entire codebase.
Changes by category:
Unused imports removed (~95 across 55 files):
- Removed genuinely unused imports from all major subsystems
- agent/, hermes_cli/, tools/, gateway/, plugins/, cron/
- Includes imports in try/except blocks that were truly unused
(vs availability checks which were left alone)
Unused variables removed (~25):
- Removed dead variables: connected, inner, channels, last_exc,
source, new_server_names, verify, pconfig, default_terminal,
result, pending_handled, temperature, loop
- Dropped unused argparse subparser assignments in hermes_cli/main.py
(12 instances of add_parser() where result was never used)
Dead code removed:
- run_agent.py: Removed dead ternary (None if False else None) and
surrounding unreachable branch in identity fallback
- run_agent.py: Removed write-only attribute _last_reported_tool
- hermes_cli/providers.py: Removed dead @property decorator on
module-level function (decorator has no effect outside a class)
- gateway/run.py: Removed unused MCP config load before reconnect
- gateway/platforms/slack.py: Removed dead SessionSource construction
Undefined name bugs fixed (would cause NameError at runtime):
- batch_runner.py: Added missing logger = logging.getLogger(__name__)
- tools/environments/daytona.py: Added missing Dict and Path imports
Unnecessary global statements removed (14):
- tools/terminal_tool.py: 5 functions declared global for dicts
they only mutated via .pop()/[key]=value (no rebinding)
- tools/browser_tool.py: cleanup thread loop only reads flag
- tools/rl_training_tool.py: 4 functions only do dict mutations
- tools/mcp_oauth.py: only reads the global
- hermes_time.py: only reads cached values
Inefficient patterns fixed:
- startswith/endswith tuple form: 15 instances of
x.startswith('a') or x.startswith('b') consolidated to
x.startswith(('a', 'b'))
- len(x)==0 / len(x)>0: 13 instances replaced with pythonic
truthiness checks (not x / bool(x))
- in dict.keys(): 5 instances simplified to in dict
- Redefined unused name: removed duplicate _strip_mdv2 import in
send_message_tool.py
Other fixes:
- hermes_cli/doctor.py: Replaced undefined logger.debug() with pass
- hermes_cli/config.py: Consolidated chained .endswith() calls
Test results: 3934 passed, 17 failed (all pre-existing on main),
19 skipped. Zero regressions.
* fix: repair 57 failing CI tests across 14 files
Categories of fixes:
**Test isolation under xdist (-n auto):**
- test_hermes_logging: Strip ALL RotatingFileHandlers before each test
to prevent handlers leaked from other xdist workers from polluting counts
- test_code_execution: Force TERMINAL_ENV=local in setUp — prevents Modal
AuthError when another test leaks TERMINAL_ENV=modal
- test_timezone: Same TERMINAL_ENV fix for execute_code timezone tests
- test_codex_execution_paths: Mock _resolve_turn_agent_config to ensure
model resolution works regardless of xdist worker state
**Matrix adapter tests (nio not installed in CI):**
- Add _make_fake_nio() helper with real response classes for isinstance()
checks in production code
- Replace MagicMock(spec=nio.XxxResponse) with fake_nio instances
- Wrap production method calls with patch.dict('sys.modules', {'nio': ...})
so import nio succeeds in method bodies
- Use try/except instead of pytest.importorskip for nio.crypto imports
(importorskip can be fooled by MagicMock in sys.modules)
- test_matrix_voice: Skip entire file if nio is a mock, not just missing
**Stale test expectations:**
- test_cli_provider_resolution: _prompt_provider_choice now takes **kwargs
(default param added); mock getpass.getpass alongside input
- test_anthropic_oauth_flow: Mock getpass.getpass (code switched from input)
- test_gemini_provider: Mock models.dev + OpenRouter API lookups to test
hardcoded defaults without external API variance
- test_code_execution: Add notify_on_complete to blocked terminal params
- test_setup_openclaw_migration: Mock prompt_choice to select 'Full setup'
(new quick-setup path leads to _require_tty → sys.exit in CI)
- test_skill_manager_tool: Patch get_all_skills_dirs alongside SKILLS_DIR
so _find_skill searches tmp_path, not real ~/.hermes/skills/
**Missing attributes in object.__new__ test runners:**
- test_platform_reconnect: Add session_store to _make_runner()
- test_session_race_guard: Add hooks, _running_agents_ts, session_store,
delivery_router to _make_runner()
**Production bug fix (gateway/run.py):**
- Fix sentinel eviction race: _AGENT_PENDING_SENTINEL was immediately
evicted by the stale-detection logic because sentinels have no
get_activity_summary() method, causing _stale_idle=inf >= timeout.
Guard _should_evict with 'is not _AGENT_PENDING_SENTINEL'.
* fix: address remaining CI failures
- test_setup_openclaw_migration: Also mock _offer_launch_chat (called at
end of both quick and full setup paths)
- test_code_execution: Move TERMINAL_ENV=local to module level to protect
ALL test classes (TestEnvVarFiltering, TestExecuteCodeEdgeCases,
TestInterruptHandling, TestHeadTailTruncation) from xdist env leaks
- test_matrix: Use try/except for nio.crypto imports (importorskip can be
fooled by MagicMock in sys.modules under xdist)
