async_call_llm (and call_llm) can return non-OpenAI objects from
custom providers or adapter shims, crashing downstream consumers
with misleading AttributeError ('str' has no attribute 'choices').
Add _validate_llm_response() that checks the response has the
expected .choices[0].message shape before returning. Wraps all
return paths in call_llm, async_call_llm, and fallback paths.
Fails fast with a clear RuntimeError identifying the task, response
type, and a preview of the malformed payload.
Closes#7264
`resolve_provider_client()` already drops OpenRouter-format model slugs
(containing "/") when the resolved provider is not OpenRouter (line 1097).
However, `_get_cached_client()` returns `model or cached_default` directly
on cache hits, bypassing this check entirely.
When the main provider is openai-codex, the auto-detection chain (Step 1
of `_resolve_auto`) caches a CodexAuxiliaryClient. Subsequent auxiliary
calls for different tasks (e.g. compression with `summary_model:
google/gemini-3-flash-preview`) hit the cache and pass the OpenRouter-
format model slug straight to the Codex Responses API, which does not
understand it and returns an empty `response.output`.
This causes two user-visible failures:
- "Invalid API response shape" (empty output after 3 retries)
- "Context length exceeded, cannot compress further" (compression itself
fails through the same path)
Add `_compat_model()` helper that mirrors the "/" check from
`resolve_provider_client()` and call it on the cache-hit return path.
Four fixes to auxiliary_client.py:
1. Respect explicit provider as hard constraint (#7559)
When auxiliary.{task}.provider is explicitly set (not 'auto'),
connection/payment errors no longer silently fallback to cloud
providers. Local-only users (Ollama, vLLM) will no longer get
unexpected OpenRouter billing from auxiliary tasks.
2. Eliminate model='default' sentinel (#7512)
_resolve_api_key_provider() no longer sends literal 'default' as
model name to APIs. Providers without a known aux model in
_API_KEY_PROVIDER_AUX_MODELS are skipped instead of producing
model_not_supported errors.
3. Add payment/connection fallback to async_call_llm (#7512)
async_call_llm now mirrors sync call_llm's fallback logic for
payment (402) and connection errors. Previously, async consumers
(session_search, web_tools, vision) got hard failures with no
recovery. Also fixes hardcoded 'openrouter' fallback to use the
full auto-detection chain.
4. Use accurate error reason in fallback logs (#7512)
_try_payment_fallback() now accepts a reason parameter and uses
it in log messages. Connection timeouts are no longer misleadingly
logged as 'payment error'.
Closes#7559Closes#7512
The auxiliary client always calls client.chat.completions.create(),
ignoring the api_mode config flag. This breaks codex-family models
(e.g. gpt-5.3-codex) on direct OpenAI API keys, which need the
/v1/responses endpoint.
Changes:
- Expand _resolve_task_provider_model to return api_mode (5-tuple)
- Read api_mode from auxiliary.{task}.api_mode config and env vars
(AUXILIARY_{TASK}_API_MODE)
- Pass api_mode through _get_cached_client to resolve_provider_client
- Add _needs_codex_wrap/_wrap_if_needed helpers that wrap plain OpenAI
clients in CodexAuxiliaryClient when api_mode=codex_responses or
when auto-detection finds api.openai.com + codex model pattern
- Apply wrapping at all custom endpoint, named custom provider, and
API-key provider return paths
- Update test mocks for the new 5-tuple return format
Users can now set:
auxiliary:
compression:
model: gpt-5.3-codex
base_url: https://api.openai.com/v1
api_mode: codex_responses
Closes#6800
Refactor hardcoded color constants throughout the CLI to resolve from
the active skin engine, so custom themes fully control the visual
appearance.
cli.py:
- Replace _GOLD constant with _ACCENT (_SkinAwareAnsi class) that
lazily resolves response_border from the active skin
- Rename _GOLD_DEFAULT to _ACCENT_ANSI_DEFAULT
- Make _build_compact_banner() read banner_title/accent/dim from skin
- Make session resume notifications use _accent_hex()
- Make status line use skin colors (accent_color, separator_color,
label_color instead of cryptic _dim_c/_dim_c2/_accent_c/_label_c)
- Reset _ACCENT cache on /skin switch
agent/display.py:
- Replace hardcoded diff ANSI escapes with skin-aware functions:
_diff_dim(), _diff_file(), _diff_hunk(), _diff_minus(), _diff_plus()
(renamed from SCREAMING_CASE _ANSI_* to snake_case)
- Add reset_diff_colors() for cache invalidation on skin switch
_discover_bundled_skills() used the directory name to identify skills,
but skills_tool.py and skills_hub.py use the `name:` field from SKILL.md
frontmatter. This mismatch caused 9 builtin skills whose directory name
differs from their SKILL.md name to be written to .bundled_manifest
under the wrong key, so `hermes skills list` showed them as "local"
instead of "builtin".
Read the frontmatter name field (with directory-name fallback) so the
manifest keys match what the rest of the codebase expects.
Closes#6835
Aligns MiniMax provider with official API documentation. Fixes 6 bugs:
transport mismatch (openai_chat -> anthropic_messages), credential leak
in switch_model(), prompt caching sent to non-Anthropic endpoints,
dot-to-hyphen model name corruption, trajectory compressor URL routing,
and stale doctor health check.
Also corrects context window (204,800), thinking support (manual mode),
max output (131,072), and model catalog (M2 family only on /anthropic).
Source: https://platform.minimax.io/docs/api-reference/text-anthropic-api
Co-authored-by: kshitijk4poor <kshitijk4poor@users.noreply.github.com>
Two fixes for the honcho memory plugin: (1) initOnSessionStart — opt-in eager session init in tools mode so sync_turn() works from turn 1 (default false, non-breaking). (2) peerName fix — gateway user_id no longer silently overwrites an explicitly configured peerName. 11 new tests. Contributed by @Kathie-yu.
The pre_llm_call plugin hook receives session_id, user_message,
conversation_history, is_first_turn, model, and platform — but not
the sender's user_id. This means plugins cannot perform per-user
access control (e.g. restricting knowledge base recall to authorized
users).
The gateway already passes source.user_id as user_id to AIAgent,
which stores it in self._user_id. This change forwards it as
sender_id in the pre_llm_call kwargs so plugins can use it for
ACL decisions.
For CLI sessions where no user_id exists, sender_id defaults to
empty string. Plugins can treat empty sender_id as a trusted local
call (the owner is at the terminal) or deny it depending on their
ACL policy.
_is_oauth_token() returned True for any key not starting with 'sk-ant-api',
which means MiniMax and Alibaba API keys were falsely treated as Anthropic
OAuth tokens. This triggered the Claude Code compatibility path:
- All tool names prefixed with mcp_ (e.g. mcp_terminal, mcp_web_search)
- System prompt injected with 'You are Claude Code' identity
- 'Hermes Agent' replaced with 'Claude Code' throughout
Fix: Make _is_oauth_token() positively identify Anthropic OAuth tokens by
their key format instead of using a broad catch-all:
- sk-ant-* (but not sk-ant-api-*) -> setup tokens, managed keys
- eyJ* -> JWTs from Anthropic OAuth flow
- Everything else -> False (MiniMax, Alibaba, etc.)
Reported by stefan171.
Resumed sessions showed raw JSON tool output in content boxes instead
of the compact trail lines seen during live use. The root cause was
two separate rendering paths with no shared code.
Extract buildToolTrailLine() into lib/text.ts as the single source
of truth for formatting tool trail lines. Both the live tool.complete
handler and toTranscriptMessages now call it.
Server-side, reconstruct tool name and args from the assistant
message's tool_calls field (tool_name column is unpopulated) and
pass them through _tool_ctx/build_tool_preview — the same path
the live tool.start callback uses.
Generate project ideas through creative constraints. Constraint + direction
= creativity.
Core skill (SKILL.md, 147 lines):
- 15 curated constraints across 3 categories: developers, makers, anyone
- Developer-focused prompts: 'solve your own itch', 'the CLI tool that
should exist', 'automate the annoying thing', 'nothing new except glue'
- Matching table: maps user mood/intent to appropriate constraints
- Complete worked example with 3 concrete project ideas
- Output format for consistent, actionable idea presentation
Extended library (references/full-prompt-library.md, 110 lines):
- 30+ additional constraints: communication, screens, philosophy,
transformation, identity, scale, starting points
Constraint approach inspired by wttdotm.com/prompts.html. Adapted for
software development and general-purpose ideation.
session.resume was building conversation history with only role and
content, stripping tool_call_id, tool_calls, and tool_name. The API
requires tool messages to reference their parent tool_call, so resumed
sessions with tool history would fail with HTTP 500.
Use get_messages_as_conversation() which already preserves the full
message structure including tool metadata and reasoning fields.
- Add agent.close() call to _finalize_shutdown_agents() to prevent
zombie processes (terminal sandboxes, browser daemons, httpx clients)
- Global cleanup (process_registry, environments, browsers) preserved
in _stop_impl() during conflict resolution
- Move /restart CommandDef from 'Info' to 'Session' category to match
/stop and /status
* fix: circuit breaker stops CPU-burning restart loops on persistent errors
When a gateway session hits a non-retryable error (e.g. invalid model
ID → HTTP 400), the agent fails and returns. But if the session keeps
receiving messages (or something periodically recreates agents), each
attempt spawns a new AIAgent — reinitializing MCP server connections,
burning CPU — only to hit the same 400 error again. On a 4-core server,
this pegs an entire core per stuck session and accumulates 300+ minutes
of CPU time over hours.
Fix: add a per-session consecutive failure counter in the gateway runner.
- Track consecutive non-retryable failures per session key
- After 3 consecutive failures (_MAX_CONSECUTIVE_FAILURES), block
further agent creation for that session and notify the user:
'⚠️ This session has failed N times in a row with a non-retryable
error. Use /reset to start a new session.'
- Evict the cached agent when the circuit breaker engages to prevent
stale state from accumulating
- Reset the counter on successful agent runs
- Clear the counter on /reset and /new so users can recover
- Uses getattr() pattern so bare GatewayRunner instances (common in
tests using object.__new__) don't crash
Tests:
- 8 new tests in test_circuit_breaker.py covering counter behavior,
threshold, reset, session isolation, and bare-runner safety
Addresses #7130.
* Revert "fix: circuit breaker stops CPU-burning restart loops on persistent errors"
This reverts commit d848ea7109.
* fix: don't evict cached agent on failed runs — prevents MCP restart loop
When a run fails (e.g. invalid model ID → 400) and fallback activated,
the gateway was evicting the cached agent to 'retry primary next time.'
But evicting a failed agent forces a full AIAgent recreation on the next
message — reinitializing MCP server connections, spawning stdio
processes — only to hit the same 400 again. This created a CPU-burning
loop (91%+ for hours, #7130).
The fix: add `and not _run_failed` to the fallback-eviction check.
Failed runs keep the cached agent. The next message reuses it (no MCP
reinit), hits the same error, returns it to the user quickly. The user
can /reset or /model to fix their config.
Successful fallback runs still evict as before so the next message
retries the primary model.
Addresses #7130.
- Remove unreachable `if not content_sample` branch inside the truthy
`if content_sample` block in `_is_likely_binary()` (dead code that
could never execute).
- Replace `linter_cmd.format(file=...)` with `linter_cmd.replace("{file}", ...)`
in `_check_lint()` so file paths containing curly braces (e.g.
`src/{test}.py`) no longer raise KeyError/ValueError.
- Add 16 unit tests covering both fixes and edge cases.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
GPT-5+ models (except gpt-5-mini) are only accessible via the Responses
API on Copilot. When these models were configured as the compression
summary_model (or any auxiliary task), the plain OpenAI client sent them
to /chat/completions which returned a 400 error:
model "gpt-5.4-mini" is not accessible via the /chat/completions endpoint
resolve_provider_client() now checks _should_use_copilot_responses_api()
for the copilot provider and wraps the client in CodexAuxiliaryClient
when needed, routing calls through responses.stream() transparently.
Adds tests for both the wrapping (gpt-5.4-mini) and non-wrapping
(gpt-4.1-mini) paths.
Add delegation.reasoning_effort config key so subagents can run at a
different thinking level than the parent agent. When set, overrides
the parent's reasoning_config; when empty, inherits as before.
Valid values: xhigh, high, medium, low, minimal, none (disables thinking).
Config path: delegation.reasoning_effort in config.yaml
Files changed:
- tools/delegate_tool.py: resolve override in _build_child_agent
- hermes_cli/config.py: add reasoning_effort to DEFAULT_CONFIG
- tests/tools/test_delegate.py: 4 new tests covering all cases
Follow-up fixes for the matrix-nio → mautrix migration:
1. Module-level mautrix.types import now wrapped in try/except with
proper stub classes. Without this, importing gateway.platforms.matrix
crashes the entire gateway when mautrix isn't installed — even for
users who don't use Matrix. The stubs mirror mautrix's real attribute
names so tests that exercise adapter methods (send, reactions, etc.)
work without the real SDK.
2. Removed _ensure_mautrix_mock() from test_matrix_mention.py — it
permanently installed MagicMock modules in sys.modules via setdefault(),
polluting later tests in the suite. No longer needed since the module
imports cleanly without mautrix.
3. Fixed thread persistence tests to use direct class reference in
monkeypatch.setattr() instead of string-based paths, which broke
when the module was reimported by other tests.
4. Moved the module-importability test to a subprocess to prevent it
from polluting sys.modules (reimporting creates a second module object
with different __dict__, breaking patch.object in subsequent tests).
The old nio code only handled RoomMessageText (m.text). The mautrix
rewrite dispatched both m.text and m.notice, which would cause infinite
loops between bots since m.notice is the conventional msgtype for bot
responses in the Matrix ecosystem.
- Add api.session.close() on E2EE dep check and E2EE setup failure
paths (two missing cleanup points from the mautrix migration)
- Replace raw pickle.load/dump with HMAC-SHA256 signed payloads to
prevent arbitrary code execution from a tampered store file
- Extract _resolve_message_context() to deduplicate ~40 lines of
mention/thread/DM gating logic between text and media handlers
- Move mautrix.types imports to module level (16 scattered local
imports consolidated)
- Parse mention/thread env vars once in __init__ instead of per-message
- Cache _is_bot_mentioned() result instead of calling 3x per event
- Consolidate send_emote/send_notice into shared _send_simple_message()
- Use _is_dm_room() in get_chat_info() instead of inline duplication
- Add _CRYPTO_PICKLE_PATH constant (was duplicated in 2 locations)
- Fix fragile event_ts extraction (double getattr, None safety)
- Clean up leaked aiohttp session on auth failure paths
- Remove redundant trailing _track_thread() calls
Address two bugs found by code review:
1. MemoryCryptoStore loses all E2EE keys on restart — now pickle the
store to disk on disconnect and restore on connect, preserving
Megolm sessions across restarts.
2. Encrypted events buffered for retry were silently dropped after
decryption because _on_encrypted_event registered the event ID
in the dedup set, then _on_room_message rejected it as a
duplicate. Now clear the dedup entry before routing decrypted
events.
Translate all nio SDK calls to mautrix equivalents while preserving the
adapter structure, business logic, and all features (E2EE, reactions,
threading, mention gating, text batching, media caching, voice MSC3245).
Key changes:
- nio.AsyncClient -> mautrix.client.Client + HTTPAPI + MemoryStateStore
- Manual E2EE key management -> OlmMachine with auto key lifecycle
- isinstance(resp, nio.XxxResponse) -> mautrix returns values directly
- add_event_callback per type -> single ROOM_MESSAGE handler with
msgtype dispatch
- Room state (member_count, display_name) via async state store lookups
- Upload/download return ContentURI/bytes directly (no wrapper objects)
matrix-nio pulls in peewee -> atomicwrites (sdist-only, archived,
missing build-system metadata) which breaks nix flake builds.
mautrix-python publishes wheels, has a leaner dep tree, and its
[encryption] extra uses the same python-olm without the problematic
transitive chain.
- Add shared is_wsl() to hermes_constants (like is_termux)
- Update supports_systemd_services() to verify systemd is actually
running on WSL before returning True
- Add WSL-specific guidance in gateway install/start/setup/status
for both cases: WSL+systemd and WSL without systemd
- Improve help strings: 'run' now says recommended for WSL/Docker,
'start'/'install' now mention systemd/launchd explicitly
- Add WSL gateway FAQ section with tmux/nohup/Task Scheduler tips
- Update CLI commands docs with WSL tip
- Deduplicate _is_wsl() from clipboard.py to shared hermes_constants
- Fix clipboard tests to reset hermes_constants cache
- 20 new WSL-specific tests covering detection, systemd check,
supports_systemd_services integration, and command output
Motivated by user feedback: took 1 hour to figure out run vs start
on WSL, Telegram bot kept disconnecting due to flaky WSL systemd.
Cover the three key behaviors:
- bulk_upload_fn is called instead of per-file upload_fn
- Fallback to upload_fn when bulk_upload_fn is None
- Rollback on bulk upload failure retries all files
warnings.warn() is suppressed/invisible when running as a gateway
or agent. Switch to logger.warning() so the disk cap message
actually appears in logs.
Fixes#7362 (item 3).
Add terminal.container_cpu, container_memory, container_disk, and
container_persistent to the _config_to_env_sync dict so that
`hermes config set terminal.container_memory 8192` correctly
writes TERMINAL_CONTAINER_MEMORY=8192 to ~/.hermes/.env.
Previously these YAML keys had no effect because terminal_tool.py
reads only env vars and the bridge was missing these mappings.
Fixes#7362 (item 2).
FileSyncManager now accepts an optional bulk_upload_fn callback.
When provided, all changed files are uploaded in one call instead
of iterating one-by-one with individual HTTP POSTs.
DaytonaEnvironment wires this to sandbox.fs.upload_files() which
batches everything into a single multipart POST — ~580 files goes
from ~5 min to <2s on init.
Parent directories are pre-created in one mkdir -p call.
Fixes#7362 (item 1).
- Remove auto-activation: when context.engine is 'compressor' (default),
plugin-registered engines are NOT used. Users must explicitly set
context.engine to a plugin name to activate it.
- Add curses_radiolist() to curses_ui.py: single-select radio picker
with keyboard nav + text fallback, matching curses_checklist pattern.
- Rewrite cmd_toggle() as composite plugins UI:
Top section: general plugins with checkboxes (existing behavior)
Bottom section: provider plugin categories (Memory Provider, Context Engine)
with current selection shown inline. ENTER/SPACE on a category opens
a radiolist sub-screen for single-select configuration.
- Add provider discovery helpers: _discover_memory_providers(),
_discover_context_engines(), config read/save for memory.provider
and context.engine.
- Add tests: radiolist non-TTY fallback, provider config save/load,
discovery error handling, auto-activation removal verification.
Follow-up fixes for the context engine plugin slot (PR #5700):
- Enhance ContextEngine ABC: add threshold_percent, protect_first_n,
protect_last_n as class attributes; complete update_model() default
with threshold recalculation; clarify on_session_end() lifecycle docs
- Add ContextCompressor.update_model() override for model/provider/
base_url/api_key updates
- Replace all direct compressor internal access in run_agent.py with
ABC interface: switch_model(), fallback restore, context probing
all use update_model() now; _context_probed guarded with getattr/
hasattr for plugin engine compatibility
- Create plugins/context_engine/ directory with discovery module
(mirrors plugins/memory/ pattern) — discover_context_engines(),
load_context_engine()
- Add context.engine config key to DEFAULT_CONFIG (default: compressor)
- Config-driven engine selection in run_agent.__init__: checks config,
then plugins/context_engine/<name>/, then general plugin system,
falls back to built-in ContextCompressor
- Wire on_session_end() in shutdown_memory_provider() at real session
boundaries (CLI exit, /reset, gateway expiry)
- PluginContext.register_context_engine() lets plugins replace the
built-in ContextCompressor with a custom ContextEngine implementation
- PluginManager stores the registered engine; only one allowed
- run_agent.py checks for a plugin engine at init before falling back
to the default ContextCompressor
- reset_session_state() now calls engine.on_session_reset() instead of
poking internal attributes directly
- ContextCompressor.on_session_reset() handles its own internals
(_context_probed, _previous_summary, etc.)
- 19 new tests covering ABC contract, defaults, plugin slot registration,
rejection of duplicates/non-engines, and compressor reset behavior
- All 34 existing compressor tests pass unchanged
Introduces agent/context_engine.py — an abstract base class that defines
the pluggable context engine interface. ContextCompressor now inherits
from ContextEngine as the default implementation.
No behavior change. All 34 existing compressor tests pass.
This is the foundation for a context engine plugin slot, enabling
third-party engines like LCM (Lossless Context Management) to replace
the built-in compressor via the plugin system.
