One-shot local execution built `printf FENCE; <cmd>; __hermes_rc=...`, so a
command ending in a heredoc produced a closing line like `EOF; __hermes_rc=...`,
which is not a valid delimiter. Bash then treated the rest of the wrapper as
heredoc body, leaking it into tool output (e.g. gh issue/PR flows).
Use newline-separated wrapper lines so the delimiter stays alone and the
trailer runs after the heredoc completes.
Co-authored-by: kshitijk4poor <82637225+kshitijk4poor@users.noreply.github.com>
Prevent the agent from accidentally killing its own process with
pkill -f gateway, killall hermes, etc. Adds a dangerous command
pattern that triggers the approval flow.
Co-authored-by: arasovic <arasovic@users.noreply.github.com>
Adds 'hermes webhook' CLI subcommand and a skill — zero new model tools.
CLI commands (require webhook platform to be enabled):
hermes webhook subscribe <name> [--events, --prompt, --deliver, ...]
hermes webhook list
hermes webhook remove <name>
hermes webhook test <name>
All commands gate on webhook platform being enabled in config. If not
configured, prints setup instructions (gateway setup wizard, manual
config.yaml, or env vars).
The agent uses these via terminal tool, guided by the webhook-subscriptions
skill which documents setup, common patterns (GitHub, Stripe, CI/CD,
monitoring), prompt template syntax, security, and troubleshooting.
Adapter enhancement: webhook.py hot-reloads dynamic subscriptions from
~/.hermes/webhook_subscriptions.json on each incoming request (mtime-gated).
Static config.yaml routes always take precedence.
Docs: updated webhooks.md with Dynamic Subscriptions section, added
hermes webhook to cli-commands.md reference.
No new model tools. No toolset changes.
24 new tests for CLI CRUD, persistence, enabled-gate, and adapter
dynamic route loading.
Two fixes for /skills install and /skills uninstall slash commands:
1. input() hangs indefinitely inside prompt_toolkit's TUI event loop,
soft-locking the CLI. The user typing the slash command is already
implicit consent, so confirmation is now always skipped.
2. Cache invalidation was unconditional — installing or uninstalling a
skill mid-session silently broke the prompt cache, increasing costs.
The slash handler now defers cache invalidation by default (skill
takes effect next session). Pass --now to invalidate immediately,
with a message explaining the cost tradeoff. The CLI argparse path
(hermes skills install) is unaffected and still invalidates.
Fixes#3474
Salvaged from PR #3496 by dlkakbs.
hermes update hangs on input() when run from cron, scripts, or piped
contexts. Check both stdin and stdout isatty(), catch EOFError as a
fallback, and print guidance to run 'hermes config migrate' later.
Co-authored-by: phippsbot-byte <phippsbot-byte@users.noreply.github.com>
When all messaging platforms exhaust retries and get queued for background
reconnection, exit with code 1 so systemd Restart=on-failure can restart
the process. Previously the gateway stayed alive as a zombie with no
connected platforms and exit code 0.
Salvaged from PR #3567 by kelsia14. Test updates added.
Co-authored-by: kelsia14 <kelsia14@users.noreply.github.com>
* fix: keep gateway running through telegram proxy failures
- continue gateway startup in degraded mode when Telegram cannot connect yet
- ensure Telegram fallback transport also honors proxy env vars
- support reconnect retries without taking down the whole gateway
* test(telegram): cover proxy env handling in fallback transport
---------
Co-authored-by: kufufu9 <pi@local>
Salvage of PR #2173 (hanai) and PR #3432 (timknip).
Injects PATH, VIRTUAL_ENV, and HERMES_HOME into the macOS launchd plist so gateway subprocesses find user-installed tools (node, ffmpeg, etc.). Matches systemd unit parity with venv/bin, node_modules/.bin, and resolved node dir in PATH. Includes 7 new tests and docs updates across 4 pages.
Co-Authored-By: Han <ihanai1991@gmail.com>
Co-Authored-By: timknip <timknip@users.noreply.github.com>
Ollama reuses index 0 for every tool call in a parallel batch,
distinguishing them only by id. The streaming accumulator now
detects a new non-empty id at an already-active index and redirects
it to a fresh slot, preventing names and arguments from being
concatenated into a single tool call.
No-op for normal providers that use incrementing indices.
Co-authored-by: dmater01 <dmater01@users.noreply.github.com>
* Fixing mattermost configuration parsing bugs
* fix: add homeassistant to skills_config + platform consistency tests
Follow-up for cherry-picked #3512:
- Add homeassistant to skills_config.py PLATFORMS (was in tools_config
but missing from skills_config)
- Add 3 consistency tests that verify all platforms in tools_config have
matching toolset definitions, gateway includes, and skills_config entries
— prevents this class of bug from recurring
---------
Co-authored-by: DaneelV3 <dannel@v3rtical.tech>
Commands sent directly to the bot in groups include @botname suffix
(e.g. /compress@TigerNanoBot). get_command() now strips the @anything
part before lookup, matching how Telegram bot menu generates commands.
Fixes all slash commands silently doing nothing when sent with @mention.
Co-authored-by: MacroAnarchy <MacroAnarchy@users.noreply.github.com>
Add X-Content-Type-Options: nosniff and Referrer-Policy: no-referrer
to all API server responses via a new security_headers_middleware.
Co-authored-by: Oktay Aydin <aydnOktay@users.noreply.github.com>
Adds Access-Control-Max-Age: 600 to CORS preflight responses, telling
browsers to cache the preflight for 10 minutes. Reduces redundant OPTIONS
requests and improves perceived latency for browser-based API clients.
Salvaged from PR #3514 by aydnOktay.
Co-authored-by: aydnOktay <xaydinoktay@gmail.com>
* feat: GPT tool-use steering + strip budget warnings from history
Two changes to improve tool reliability, especially for OpenAI GPT models:
1. GPT tool-use enforcement prompt: Adds GPT_TOOL_USE_GUIDANCE to the
system prompt when the model name contains 'gpt' and tools are loaded.
This addresses a known behavioral pattern where GPT models describe
intended actions ('I will run the tests') instead of actually making
tool calls. Inspired by similar steering in OpenCode (beast.txt) and
Cline (GPT-5.1 variant).
2. Budget warning history stripping: Budget pressure warnings injected by
_get_budget_warning() into tool results are now stripped when
conversation history is replayed via run_conversation(). Previously,
these turn-scoped signals persisted across turns, causing models to
avoid tool calls in all subsequent messages after any turn that hit
the 70-90% iteration threshold.
* fix: replace hardcoded ~/.hermes paths with get_hermes_home() for profile support
Prep for the upcoming profiles feature — each profile is a separate
HERMES_HOME directory, so all paths must respect the env var.
Fixes:
- gateway/platforms/matrix.py: Matrix E2EE store was hardcoded to
~/.hermes/matrix/store, ignoring HERMES_HOME. Now uses
get_hermes_home() so each profile gets its own Matrix state.
- gateway/platforms/telegram.py: Two locations reading config.yaml via
Path.home()/.hermes instead of get_hermes_home(). DM topic thread_id
persistence and hot-reload would read the wrong config in a profile.
- tools/file_tools.py: Security path for hub index blocking was
hardcoded to ~/.hermes, would miss the actual profile's hub cache.
- hermes_cli/gateway.py: Service naming now uses the profile name
(hermes-gateway-coder) instead of a cryptic hash suffix. Extracted
_profile_suffix() helper shared by systemd and launchd.
- hermes_cli/gateway.py: Launchd plist path and Label now scoped per
profile (ai.hermes.gateway-coder.plist). Previously all profiles
would collide on the same plist file on macOS.
- hermes_cli/gateway.py: Launchd plist now includes HERMES_HOME in
EnvironmentVariables — was missing entirely, making custom
HERMES_HOME broken on macOS launchd (pre-existing bug).
- All launchctl commands in gateway.py, main.py, status.py updated
to use get_launchd_label() instead of hardcoded string.
Test fixes: DM topic tests now set HERMES_HOME env var alongside
Path.home() mock. Launchd test uses get_launchd_label() for expected
commands.
PR #3566 intentionally routes suppressed content to stream_delta_callback
when tool calls are present, so reasoning tag extraction can fire during
streaming. The test was still asserting the old behavior where content
after tool calls was fully suppressed from the callback.
Updated the assertion to match: content IS delivered to the callback
(for tag extraction), with display-level suppression handled by the
CLI's _stream_delta.
Add GET /v1/health as an alias to the existing /health endpoint so
OpenAI-compatible health checks work out of the box.
Co-authored-by: Oktay Aydin <aydnOktay@users.noreply.github.com>
The TOOL_USE_ENFORCEMENT_GUIDANCE injection (added in #3528) was
hardcoded to only match gpt/codex model names. This makes it a
config option so users can turn it on for any model family.
New config key: agent.tool_use_enforcement
- "auto" (default): matches gpt/codex (existing behavior)
- true: inject for all models
- false: never inject
- list of strings: custom model-name substrings to match
e.g. ["gpt", "codex", "deepseek", "qwen"]
No version bump needed — deep merge provides the default
automatically for existing installs.
12 new tests covering all config modes.
* fix(gateway): preserve full transcript on /compress instead of overwriting
The /compress command calls _compress_context() which correctly ends the
old session (preserving its full transcript in SQLite) and creates a new
session_id for the continuation. However, it then immediately called
rewrite_transcript() on the OLD session_id, overwriting the preserved
transcript with the compressed version — destroying searchable history.
Auto-compression (triggered by context pressure) does not have this bug
because the gateway already handles the session_id swap via the
agent.session_id != session_id check after _run_agent_sync.
Fix: after _compress_context creates the new session, write the compressed
messages into the NEW session_id and update the session store pointer.
The old session's full transcript stays intact and searchable via
session_search.
Before: /compress destroys original messages, session_search can't find
details from compressed portions.
After: /compress behaves like /new for history — full transcript preserved,
compressed context for the live session.
* fix(gateway): preserve transcript on /compress and hygiene compression
Apply session_id swap after _compress_context in both /compress handler
and hygiene pre-compression. _compress_context creates a new session
(ending the old one), but both paths were calling rewrite_transcript on
the OLD session_id — overwriting the preserved transcript and destroying
searchable history.
Now follows the same pattern as the auto-compression handler (lines
5415-5423): detect the new session_id, update the session store entry,
and write compressed messages to the new session.
Also fix FakeCompressAgent test mock to include session_id attribute
and simulate the session_id change that real _compress_context performs.
Co-authored-by: MacroAnarchy <MacroAnarchy@users.noreply.github.com>
---------
Co-authored-by: MacroAnarchy <MacroAnarchy@users.noreply.github.com>
* fix(matrix): harden e2ee access-token handling
* fix: patch nio mock in e2ee maintenance sync loop test
The sync_loop now imports nio for SyncError checking (from PR #3280),
so the test needs to inject a fake nio module via sys.modules.
---------
Co-authored-by: Cortana <andrew+cortana@chalkley.org>
The minimax-specific auto-correction in runtime_provider.py was
preventing users from overriding to the OpenAI-compatible endpoint
via MINIMAX_BASE_URL. Users in certain regions get nginx 404 on
api.minimax.io/anthropic and need to switch to api.minimax.chat/v1.
The generic URL-suffix detection already handles /anthropic →
anthropic_messages, so the minimax-specific code was redundant for
the default path and harmful for the override path.
Now: default /anthropic URL works via generic detection, user
override to /v1 gets chat_completions mode naturally.
Closes#3546 (different approach — respects user overrides instead
of changing the default endpoint).
Drop the swe-rex dependency for Modal terminal backend and use the
Modal SDK directly (Sandbox.create + Sandbox.exec). This fixes:
- AsyncUsageWarning from synchronous App.lookup() in async context
- DeprecationError from unencrypted_ports / .url on unencrypted tunnels
(deprecated 2026-03-05)
The new implementation:
- Uses modal.App.lookup.aio() for async-safe app creation
- Uses Sandbox.create.aio() with 'sleep infinity' entrypoint
- Uses Sandbox.exec.aio() for direct command execution (no HTTP server
or tunnel needed)
- Keeps all existing features: persistent filesystem snapshots,
configurable resources (CPU/memory/disk), sudo support, interrupt
handling, _AsyncWorker for event loop safety
Consistent with the Docker backend precedent (PR #2804) where we
removed mini-swe-agent in favor of direct docker run.
Files changed:
- tools/environments/modal.py - core rewrite
- tools/terminal_tool.py - health check: modal instead of swerex
- hermes_cli/setup.py - install modal instead of swe-rex[modal]
- pyproject.toml - modal extra: modal>=1.0.0 instead of swe-rex[modal]
- scripts/kill_modal.sh - grep for hermes-agent instead of swe-rex
- tests/ - updated for new implementation
- environments/README.md - updated patches section
- website/docs - updated install command
The plugin system defined six lifecycle hooks but only pre_tool_call and
post_tool_call were invoked. This activates the remaining four so that
external plugins (e.g. memory systems) can hook into the conversation
loop without touching core code.
Hook semantics:
- on_session_start: fires once when a new session is created
- pre_llm_call: fires once per turn before the tool-calling loop;
plugins can return {"context": "..."} to inject into the ephemeral
system prompt (not cached, not persisted)
- post_llm_call: fires once per turn after the loop completes, with
user_message and assistant_response for sync/storage
- on_session_end: fires at the end of every run_conversation call
invoke_hook() now returns a list of non-None callback return values,
enabling pre_llm_call context injection while remaining backward
compatible (existing hooks that return None are unaffected).
Salvaged from PR #2823.
Co-authored-by: Nicolò Boschi <boschi1997@gmail.com>
Browser clients using the Idempotency-Key header for request
deduplication were blocked by CORS preflight because the header
was not listed in Access-Control-Allow-Headers.
Add Idempotency-Key to _CORS_HEADERS and add tests for both the
new header allowance and the existing Vary: Origin behavior.
Co-authored-by: aydnOktay <aydnOktay@users.noreply.github.com>
Co-authored-by: Hermes Agent <hermes@nousresearch.com>
Root cause: Anthropic buffers entire tool call arguments and goes silent
for minutes while thinking (verified: 167s gap with zero SSE events on
direct API). OpenRouter's upstream proxy times out after ~125s of
inactivity and drops the connection with 'Network connection lost'.
Fix: Send the x-anthropic-beta: fine-grained-tool-streaming-2025-05-14
header for Claude models on OpenRouter. This makes Anthropic stream
tool call arguments token-by-token instead of buffering them, keeping
the connection alive through OpenRouter's proxy.
Live-tested: the exact prompt that consistently failed at ~128s now
completes successfully — 2,972 lines written, 49K tokens, 8 minutes.
Additional improvements:
1. Send explicit max_tokens for Claude through OpenRouter. Without it,
OpenRouter defaults to 65,536 (confirmed via echo_upstream_body) —
only half of Opus 4.6's 128K limit.
2. Classify SSE 'Network connection lost' as retryable in the streaming
inner retry loop. The OpenAI SDK raises APIError from SSE error
events, which was bypassing our transient error retry logic.
3. Actionable diagnostic guidance when stream-drop retries exhaust.
Add ~/.local/bin, ~/.cargo/bin, ~/go/bin, ~/.npm-global/bin to the
systemd unit PATH so tools installed via uv/pipx/cargo/go are
discoverable by MCP servers and terminal commands.
Uses a _build_user_local_paths() helper that checks exists() before
adding, and correctly resolves home dir for both user and system
service types.
Co-authored-by: Kal Sze <ksze@users.noreply.github.com>
Cherry-pick of feat/gpt-tool-steering with modifications:
1. Tool-use enforcement prompt (refactored from GPT-specific):
- Renamed GPT_TOOL_USE_GUIDANCE -> TOOL_USE_ENFORCEMENT_GUIDANCE
- Added TOOL_USE_ENFORCEMENT_MODELS tuple: ('gpt', 'codex')
- Injection logic now checks against the tuple instead of hardcoding
'gpt' — adding new model families is a one-line change
- Addresses models describing actions instead of making tool calls
2. Budget warning history stripping:
- _strip_budget_warnings_from_history() strips _budget_warning JSON
keys and [BUDGET WARNING: ...] text from tool results at the start
of run_conversation()
- Prevents old budget warnings from poisoning subsequent turns
Based on PR #3479 by teknium1.
* fix: harden `hermes update` against diverged history, non-main branches, and gateway edge cases
The self-update command (`hermes update` / gateway `/update`) could fail
or silently corrupt state in several scenarios:
1. **Diverged history** — `git pull --ff-only` aborts with a cryptic
subprocess error when upstream has force-pushed or rebased. Now falls
back to `git reset --hard origin/main` since local changes are already
stashed.
2. **User on a feature branch / detached HEAD** — the old code would
either clobber the feature branch HEAD to point at origin/main, or
silently pull against a non-existent remote branch. Now auto-checkouts
main before pulling, with a clear warning.
3. **Fetch failures** — network or auth errors produced raw subprocess
tracebacks. Now shows user-friendly messages ("Network error",
"Authentication failed") with actionable hints.
4. **reset --hard failure** — if the fallback reset itself fails (disk
full, permissions), the old code would still attempt stash restore on
a broken working tree. Now skips restore and tells the user their
changes are safe in stash.
5. **Gateway /update stash conflicts** — non-interactive mode (Telegram
`/update`) called sys.exit(1) when stash restore had conflicts, making
the entire update report as failed even though the code update itself
succeeded. Now treats stash conflicts as non-fatal in non-interactive
mode (returns False instead of exiting).
* fix: restore stash and branch on 'already up to date' early return
The PR moved stash creation before the commit-count check (needed for
the branch-switching feature), but the 'already up to date' early return
didn't restore the stash or switch back to the original branch — leaving
the user stranded on main with changes trapped in a stash.
Now the early-return path restores the stash and checks out the original
branch when applicable.
---------
Co-authored-by: kshitijk4poor <82637225+kshitijk4poor@users.noreply.github.com>
- Change default inference_base_url from dashscope-intl Anthropic-compat
endpoint to coding-intl OpenAI-compat /v1 endpoint. The old Anthropic
endpoint 404'd when used with the OpenAI SDK (which appends
/chat/completions to a /apps/anthropic base URL).
- Update curated model list: remove models unavailable on coding-intl
(qwen3-max, qwen-plus-latest, qwen3.5-flash, qwen-vl-max), add
third-party models available on the platform (glm-5, glm-4.7,
kimi-k2.5, MiniMax-M2.5).
- URL-based api_mode auto-detection still works: overriding
DASHSCOPE_BASE_URL to an /apps/anthropic endpoint automatically
switches to anthropic_messages mode.
- Update provider description and env var descriptions to reflect the
coding-intl multi-provider platform.
- Update tests to match new default URL and test the anthropic override
path instead.
* fix: cap context pressure percentage at 100% in display
The forward-looking token estimate can overshoot the compaction threshold
(e.g. a large tool result pushes it from 70% to 109% in one step). The
progress bar was already capped via min(), but pct_int was not — causing
the user to see '109% to compaction' which is confusing.
Cap pct_int at 100 in both CLI and gateway display functions.
Reported by @JoshExile82.
* refactor: use real API token counts for compression decisions
Replace the rough chars/3 estimation with actual prompt_tokens +
completion_tokens from the API response. The estimation was needed to
predict whether tool results would push context past the threshold, but
the default 50% threshold leaves ample headroom — if tool results push
past it, the next API call reports real usage and triggers compression
then.
This removes all estimation from the compression and context pressure
paths, making both 100% data-driven from provider-reported token counts.
Also removes the dead _msg_count_before_tools variable.
When finish_reason='length' and the response contains only reasoning
(think blocks or empty content), the model exhausted its output token
budget on thinking with nothing left for the actual response.
Previously, this fell into either:
- chat_completions: 3 useless continuation retries (model hits same limit)
- anthropic/codex: generic 'Response truncated' error with rollback
Now: detect the think-only + length condition early and return immediately
with a targeted error message: 'Model used all output tokens on reasoning
with none left for the response. Try lowering reasoning effort or
increasing max_tokens.'
This saves 2 wasted API calls on the chat_completions path and gives
users actionable guidance instead of a cryptic error.
The existing think-only retry logic (finish_reason='stop') is unchanged —
that's a genuine model glitch where retrying can help.
Salvage of #3389 by @binhnt92 with reasoning fallback and retry logic added on top.
All 7 auxiliary LLM call sites now use extract_content_or_reasoning() which mirrors the main agent loop's behavior: extract content, strip think blocks, fall back to structured reasoning fields, retry on empty.
Closes#3389.
Show only agentic models that map to OpenRouter defaults:
Qwen/Qwen3.5-397B-A17B ↔ qwen/qwen3.5-plus
Qwen/Qwen3.5-35B-A3B ↔ qwen/qwen3.5-35b-a3b
deepseek-ai/DeepSeek-V3.2 ↔ deepseek/deepseek-chat
moonshotai/Kimi-K2.5 ↔ moonshotai/kimi-k2.5
MiniMaxAI/MiniMax-M2.5 ↔ minimax/minimax-m2.5
zai-org/GLM-5 ↔ z-ai/glm-5
XiaomiMiMo/MiMo-V2-Flash ↔ xiaomi/mimo-v2-pro
moonshotai/Kimi-K2-Thinking ↔ moonshotai/kimi-k2-thinking
Users can still pick any HF model via Enter custom model name.
Previously, tirith exit code 1 (block) immediately rejected the command
with no approval prompt — users saw 'BLOCKED: Command blocked by
security scan' and the agent moved on. This prevented gateway/CLI users
from approving pipe-to-shell installs like 'curl ... | sh' even when
they understood the risk.
Changes:
- Tirith 'block' and 'warn' now both go through the approval flow.
Users see the full tirith findings (severity, title, description,
safer alternatives) and can choose to approve or deny.
- New _format_tirith_description() builds rich descriptions from tirith
findings JSON so the approval prompt is informative.
- CLI startup now warns when tirith is enabled but not available, so
users know command scanning is degraded to pattern matching only.
The default approval choice is still deny, so the security posture is
unchanged for unattended/timeout scenarios.
Reported via Discord by pistrie — 'curl -fsSL https://mandex.dev/install.sh | sh'
was hard-blocked with no way to approve.
The Anthropic adapter defaulted to max_tokens=16384 when no explicit value
was configured. This severely limits thinking-enabled models where thinking
tokens count toward max_tokens:
- Claude Opus 4.6 supports 128K output but was capped at 16K
- Claude Sonnet 4.6 supports 64K output but was capped at 16K
With extended thinking (adaptive or budget-based), the model could exhaust
the entire 16K on reasoning, leaving zero tokens for the actual response.
This caused two user-visible errors:
- 'Response truncated (finish_reason=length)' — thinking consumed most tokens
- 'Response only contains think block with no content' — thinking consumed all
Fix: add _ANTHROPIC_OUTPUT_LIMITS lookup table (sourced from Anthropic docs
and Cline's model catalog) and use the model's actual output limit as the
default. Unknown future models default to 128K (the current maximum).
Also adds context_length clamping: if the user configured a smaller context
window (e.g. custom endpoint), max_tokens is clamped to context_length - 1
to avoid exceeding the window.
Closes#2706
Models like GLM-5/5.1 can think for 15+ minutes. The previous 900s
(15 min) default for HERMES_API_TIMEOUT killed legitimate requests.
Raised to 1800s (30 min) in both places that read the env var:
- _build_api_kwargs() timeout (non-streaming total timeout)
- _call_chat_completions() write timeout (streaming connection)
The streaming per-chunk read timeout (60s) and stale stream detector
(180-300s) are unchanged — those are appropriate for inter-chunk timing.
Salvage of PR #1747 (original PR #1171 by @davanstrien) onto current main.
Registers Hugging Face Inference Providers (router.huggingface.co/v1) as a named provider:
- hermes chat --provider huggingface (or --provider hf)
- 18 curated open models via hermes model picker
- HF_TOKEN in ~/.hermes/.env
- OpenAI-compatible endpoint with automatic failover (Groq, Together, SambaNova, etc.)
Files: auth.py, models.py, main.py, setup.py, config.py, model_metadata.py, .env.example, 5 docs pages, 17 new tests.
Co-authored-by: Daniel van Strien <davanstrien@gmail.com>
Salvage of #3399 by @binhnt92 with true agent interruption added on top.
When a streaming /v1/chat/completions client disconnects mid-stream, the agent is now interrupted via agent.interrupt() so it stops making LLM API calls, and the asyncio task wrapper is cancelled.
Closes#3399.
Two independent bugs caused the reasoning box to appear three times when
the model produced reasoning + tool_calls:
Bug A: _build_assistant_message() re-fired reasoning_callback with the full
reasoning text even when streaming had already displayed it. The original
guard only checked structured reasoning_content deltas, but reasoning also
arrives via content tag extraction (<REASONING_SCRATCHPAD>/<think> tags
in delta.content), which went through _fire_stream_delta not
_fire_reasoning_delta. Fix: skip the callback entirely when streaming is
active — both paths display reasoning during the stream. Any reasoning not
shown during streaming is caught by the CLI post-response fallback.
Bug B: The post-response reasoning display checked _reasoning_stream_started,
but that flag was reset by _reset_stream_state() during intermediate turn
boundaries (when stream_delta_callback(None) fires between tool calls).
Introduced _reasoning_shown_this_turn flag that persists across the tool
loop and is only reset at the start of each user turn.
Live-tested in PTY: reasoning now shows exactly once per API call, no
duplicates across tool-calling loops.
The OpenAI SDK's AsyncHttpxClientWrapper.__del__ schedules aclose() via
asyncio.get_running_loop().create_task(). When an AsyncOpenAI client is
garbage-collected while prompt_toolkit's event loop is running (the common
CLI idle state), the aclose() task runs on prompt_toolkit's loop but the
underlying TCP transport is bound to a different (dead) worker loop.
The transport's self._loop.call_soon() then raises RuntimeError('Event
loop is closed'), which prompt_toolkit surfaces as the disruptive
'Unhandled exception in event loop ... Press ENTER to continue...' error.
Three-layer fix:
1. neuter_async_httpx_del(): Monkey-patches __del__ to a no-op at CLI
startup before any AsyncOpenAI clients are created. Safe because
cached clients are explicitly cleaned via _force_close_async_httpx,
and uncached clients' TCP connections are cleaned by the OS on exit.
2. Custom asyncio exception handler: Installed on prompt_toolkit's event
loop to silently suppress 'Event loop is closed' RuntimeError.
Defense-in-depth for SDK upgrades that might change the class name.
3. cleanup_stale_async_clients(): Called after each agent turn (when the
agent thread joins) to proactively evict cache entries whose event
loop is closed, preventing stale clients from accumulating.
When a gateway crashes mid-job execution (before mark_job_run can persist
the updated next_run_at), the job would fire again on every restart attempt
within the grace window. For a daily 6:15 AM job with a 2-hour grace,
rapidly restarting the gateway could trigger dozens of duplicate runs.
Fix: call advance_next_run() BEFORE run_job() in tick(). For recurring
jobs (cron/interval), this preemptively advances next_run_at to the next
future occurrence and persists it to disk. If the process then crashes
during execution, the job won't be considered due on restart.
One-shot jobs are left unchanged — they still retry on restart since
there's no future occurrence to advance to.
This changes the scheduler from at-least-once to at-most-once semantics
for recurring jobs, which is the correct tradeoff: missing one daily
message is far better than sending it dozens of times.
python-telegram-bot's BadRequest inherits from NetworkError, so the
send() retry loop was catching 'Message thread not found' as a transient
network error and retrying 3 times before silently failing. This killed
all tool progress messages, streaming responses, and typing indicators
when the incoming message carried an invalid message_thread_id.
Now detect BadRequest inside the NetworkError handler:
- 'thread not found' + thread_id set → clear thread_id and retry once
(message still reaches the chat, just without topic threading)
- Other BadRequest errors → raise immediately (permanent, don't retry)
- True NetworkError → retry as before (transient)
252 silent failures in gateway.log traced to this on 2026-03-26.
5 new tests for thread fallback, non-thread BadRequest, no-thread sends,
network retry, and multi-chunk fallback.
* feat(telegram): auto-discover fallback IPs via DoH when api.telegram.org is unreachable
On some networks (university, corporate), api.telegram.org resolves to a
valid Telegram IP that is unreachable due to routing/firewall rules. A
different IP in the same Telegram-owned 149.154.160.0/20 block works fine.
This adds automatic fallback IP discovery at connect time:
1. Query Google and Cloudflare DNS-over-HTTPS for api.telegram.org A records
2. Exclude the system-DNS IP (the unreachable one), use the rest as fallbacks
3. If DoH is also blocked, fall back to a seed list (149.154.167.220)
4. TelegramFallbackTransport tries primary first, sticks to whichever works
No configuration needed — works automatically. TELEGRAM_FALLBACK_IPS env var
still available as manual override. Zero impact on healthy networks (primary
path succeeds on first attempt, fallback never exercised).
No new dependencies (uses httpx already in deps + stdlib socket).
* fix: share transport instance and downgrade seed fallback log to info
- Use single TelegramFallbackTransport shared between request and
get_updates_request so sticky IP is shared across polling and API calls
- Keep separate HTTPXRequest instances (different timeout settings)
- Downgrade "using seed fallback IPs" from warning to info to avoid
noisy logs on healthy networks
* fix: add telegram.request mock and discovery fixture to remaining test files
The original PR missed test_dm_topics.py and
test_telegram_network_reconnect.py — both need the telegram.request
mock module. The reconnect test also needs _no_auto_discovery since
_handle_polling_network_error calls connect() which now invokes
discover_fallback_ips().
---------
Co-authored-by: Mohan Qiao <Gavin-Qiao@users.noreply.github.com>
dict.get(key, default) returns None — not the default — when the key IS
present but explicitly set to null/~ in YAML. Calling .lower() on that
raises AttributeError.
Use (config.get(key) or fallback) so both missing keys and explicit nulls
coalesce to the intended default.
Files fixed:
- tools/tts_tool.py — _get_provider()
- tools/web_tools.py — _get_backend()
- tools/mcp_tool.py — MCPServerTask auth config
- trajectory_compressor.py — _detect_provider() and config loading
Co-authored-by: dieutx <dangtc94@gmail.com>
V4A patches with only + lines (no context or - lines) were silently
dropped because search_lines was empty and the 'if search_lines:' block
was the only code path. Addition-only hunks are common when the model
generates patches for new functions or blocks.
Adds an else branch that inserts at the context_hint position when
available, or appends at end of file.
Includes 2 regression tests for addition-only hunks with and without
context hints.
Salvaged from PR #3092 by thakoreh.
Co-authored-by: Hiren <hiren.thakore58@gmail.com>
* fix(gateway): add media download retry to Mattermost, Slack, and base cache
Media downloads on Mattermost and Slack fail permanently on transient
errors (timeouts, 429 rate limits, 5xx server errors). Telegram and
WhatsApp already have retry logic, but these platforms had single-attempt
downloads with hardcoded 30s timeouts.
Changes:
- base.py cache_image_from_url: add retry with exponential backoff
(covers Signal and any platform using the shared cache helper)
- mattermost.py _send_media_url: retry on 429/5xx/timeout (3 attempts)
- slack.py _download_slack_file: retry on timeout/5xx (3 attempts)
- slack.py _download_slack_file_bytes: same retry pattern
* test: add tests for media download retry
---------
Co-authored-by: dieutx <dangtc94@gmail.com>
When a new session starts in the gateway (via /new, /reset, or
auto-reset), send the user a summary of the detected configuration:
✨ Session reset! Starting fresh.
◆ Model: qwen3.5:27b-q4_K_M
◆ Provider: custom
◆ Context: 8K tokens (config)
◆ Endpoint: http://localhost:11434/v1
This makes misconfigured context length immediately visible — a user
running a local 8K model that falls to the 128K default will see:
◆ Context: 128K tokens (default — set model.context_length in config to override)
Instead of silently getting no compression and degrading responses.
- _format_session_info() resolves model, provider, context length,
and endpoint from config + runtime, matching the hygiene code's
resolution chain
- Local/custom endpoints shown; cloud endpoints hidden (not useful)
- Context source annotated: config, detected, or default with hint
- Appended to /new and /reset responses, and auto-reset notifications
- 9 tests covering all formatting paths and failure resilience
Addresses the user-facing side of #2708 — instead of trying to fix
every edge case in context detection, surface the values so users
can immediately see when something is wrong.
When user messages have empty content (e.g., Discord @mention-only
messages, unrecognized attachments), the Anthropic API rejects the
request with 'user messages must have non-empty content'.
Changes:
- anthropic_adapter.py: Add empty content validation for user messages
(string and list formats), matching the existing pattern for assistant
and tool messages. Empty content gets '(empty message)' placeholder.
- discord.py: Defense-in-depth check at gateway layer to catch empty
messages before they enter session history.
- Add 4 regression tests covering empty string, whitespace-only,
empty list, and empty text block scenarios.
Fixes#3143
Co-authored-by: Bartok9 <bartok9@users.noreply.github.com>
