hermes-agent/tools
Teknium 70768665a4
fix(mcp): consolidate OAuth handling, pick up external token refreshes (#11383)
* feat(mcp-oauth): scaffold MCPOAuthManager

Central manager for per-server MCP OAuth state. Provides
get_or_build_provider (cached), remove (evicts cache + deletes
disk), invalidate_if_disk_changed (mtime watch, core fix for
external-refresh workflow), and handle_401 (dedup'd recovery).

No behavior change yet — existing call sites still use
build_oauth_auth directly. Task 1 of 8 in the MCP OAuth
consolidation (fixes Cthulhu's BetterStack reliability issues).

* feat(mcp-oauth): add HermesMCPOAuthProvider with pre-flow disk watch

Subclasses the MCP SDK's OAuthClientProvider to inject a disk
mtime check before every async_auth_flow, via the central
manager. When a subclass instance is used, external token
refreshes (cron, another CLI instance) are picked up before
the next API call.

Still dead code: the manager's _build_provider still delegates
to build_oauth_auth and returns the plain OAuthClientProvider.
Task 4 wires this subclass in. Task 2 of 8.

* refactor(mcp-oauth): extract build_oauth_auth helpers

Decomposes build_oauth_auth into _configure_callback_port,
_build_client_metadata, _maybe_preregister_client, and
_parse_base_url. Public API preserved. These helpers let
MCPOAuthManager._build_provider reuse the same logic in Task 4
instead of duplicating the construction dance.

Also updates the SDK version hint in the warning from 1.10.0 to
1.26.0 (which is what we actually require for the OAuth types
used here). Task 3 of 8.

* feat(mcp-oauth): manager now builds HermesMCPOAuthProvider directly

_build_provider constructs the disk-watching subclass using the
helpers from Task 3, instead of delegating to the plain
build_oauth_auth factory. Any consumer using the manager now gets
pre-flow disk-freshness checks automatically.

build_oauth_auth is preserved as the public API for backwards
compatibility. The code path is now:

    MCPOAuthManager.get_or_build_provider  ->
      _build_provider  ->
        _configure_callback_port
        _build_client_metadata
        _maybe_preregister_client
        _parse_base_url
        HermesMCPOAuthProvider(...)

Task 4 of 8.

* feat(mcp): wire OAuth manager + add _reconnect_event

MCPServerTask gains _reconnect_event alongside _shutdown_event.
When set, _run_http / _run_stdio exit their async-with blocks
cleanly (no exception), and the outer run() loop re-enters the
transport to rebuild the MCP session with fresh credentials.
This is the recovery path for OAuth failures that the SDK's
in-place httpx.Auth cannot handle (e.g. cron externally consumed
the refresh_token, or server-side session invalidation).

_run_http now asks MCPOAuthManager for the OAuth provider
instead of calling build_oauth_auth directly. Config-time,
runtime, and reconnect paths all share one provider instance
with pre-flow disk-watch active.

shutdown() defensively sets both events so there is no race
between reconnect and shutdown signalling.

Task 5 of 8.

* feat(mcp): detect auth failures in tool handlers, trigger reconnect

All 5 MCP tool handlers (tool call, list_resources, read_resource,
list_prompts, get_prompt) now detect auth failures and route
through MCPOAuthManager.handle_401:

  1. If the manager says recovery is viable (disk has fresh tokens,
     or SDK can refresh in-place), signal MCPServerTask._reconnect_event
     to tear down and rebuild the MCP session with fresh credentials,
     then retry the tool call once.

  2. If no recovery path exists, return a structured needs_reauth
     JSON error so the model stops hallucinating manual refresh
     attempts (the 'let me curl the token endpoint' loop Cthulhu
     pasted from Discord).

_is_auth_error catches OAuthFlowError, OAuthTokenError,
OAuthNonInteractiveError, and httpx.HTTPStatusError(401). Non-auth
exceptions still surface via the generic error path unchanged.

Task 6 of 8.

* feat(mcp-cli): route add/remove through manager, add 'hermes mcp login'

cmd_mcp_add and cmd_mcp_remove now go through MCPOAuthManager
instead of calling build_oauth_auth / remove_oauth_tokens
directly. This means CLI config-time state and runtime MCP
session state are backed by the same provider cache — removing
a server evicts the live provider, adding a server populates
the same cache the MCP session will read from.

New 'hermes mcp login <name>' command:
  - Wipes both the on-disk tokens file and the in-memory
    MCPOAuthManager cache
  - Triggers a fresh OAuth browser flow via the existing probe
    path
  - Intended target for the needs_reauth error Task 6 returns
    to the model

Task 7 of 8.

* test(mcp-oauth): end-to-end integration tests

Five new tests exercising the full consolidation with real file
I/O and real imports (no transport mocks):

  1. external_refresh_picked_up_without_restart — Cthulhu's cron
     workflow. External process writes fresh tokens to disk;
     on the next auth flow the manager's mtime-watch flips
     _initialized and the SDK re-reads from storage.

  2. handle_401_deduplicates_concurrent_callers — 10 concurrent
     handlers for the same failed token fire exactly ONE recovery
     attempt (thundering-herd protection).

  3. handle_401_returns_false_when_no_provider — defensive path
     for unknown servers.

  4. invalidate_if_disk_changed_handles_missing_file — pre-auth
     state returns False cleanly.

  5. provider_is_reused_across_reconnects — cache stickiness so
     reconnects preserve the disk-watch baseline mtime.

Task 8 of 8 — consolidation complete.
2026-04-16 21:57:10 -07:00
..
browser_providers feat: ungate Tool Gateway — subscription-based access with per-tool opt-in 2026-04-16 12:36:49 -07:00
environments fix: harden sync_back — PID-suffix temp path, size cap, lifecycle guards 2026-04-16 19:39:21 -07:00
neutts_samples refactor(tts): replace NeuTTS optional skill with built-in provider + setup flow 2026-03-17 02:33:12 -07:00
__init__.py Merge branch 'main' into rewbs/tool-use-charge-to-subscription 2026-03-31 08:48:54 +09:00
ansi_strip.py fix: strip ANSI at the source — clean terminal output before it reaches the model 2026-03-23 07:43:12 -07:00
approval.py fix(approval): heartbeat activity during gateway approval wait (#11245) 2026-04-16 14:48:50 -07:00
binary_extensions.py fix(tools): address PR review — remove _extract_raw_output, BudgetConfig everywhere, read_file hardening 2026-04-08 02:24:32 -07:00
browser_camofox.py fix: /browser connect CDP override now takes priority over Camofox (#10523) 2026-04-15 14:11:18 -07:00
browser_camofox_state.py feat(browser): add persistent Camofox sessions and VNC URL discovery (salvage #4400) (#4419) 2026-04-01 04:18:50 -07:00
browser_tool.py fix(browser): runtime fallback to local Chromium when cloud provider fails 2026-04-16 04:19:34 -07:00
budget_config.py fix: preserve existing thresholds, remove pre-read byte guard 2026-04-08 02:24:32 -07:00
checkpoint_manager.py fix(checkpoints): isolate shadow git repo from user's global config (#11261) 2026-04-16 16:06:49 -07:00
clarify_tool.py refactor: add tool_error/tool_result helpers + read_raw_config, migrate 129 callsites 2026-04-07 13:36:38 -07:00
code_execution_tool.py fix: follow-up for salvaged PR #10854 2026-04-16 06:42:45 -07:00
credential_files.py refactor: extract shared helpers to deduplicate repeated code patterns (#7917) 2026-04-11 13:59:52 -07:00
cronjob_tools.py fix: replace hardcoded ~/.hermes with display_hermes_home() in agent-facing text (#10285) 2026-04-15 04:57:55 -07:00
debug_helpers.py refactor: codebase-wide lint cleanup — unused imports, dead code, and inefficient patterns (#5821) 2026-04-07 10:25:31 -07:00
delegate_tool.py fix: UnboundLocalError on 'entry' in parallel subagent polling loop (#11050) 2026-04-16 06:53:44 -07:00
env_passthrough.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
file_operations.py feat: improve file search UX — fuzzy @ completions, mtime sorting, better suggestions (#9467) 2026-04-13 23:54:45 -07:00
file_tools.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
fuzzy_match.py fix(patch): harden V4A patch parser and fuzzy match — 9 correctness bugs 2026-04-10 16:47:44 -07:00
homeassistant_tool.py fix: clean up description escaping, add string-data tests 2026-04-13 04:45:07 -07:00
image_generation_tool.py feat(image_gen): multi-model FAL support with picker in hermes tools (#11265) 2026-04-16 20:19:53 -07:00
interrupt.py fix: scope tool interrupt signal per-thread to prevent cross-session leaks (#7930) 2026-04-11 14:02:58 -07:00
managed_tool_gateway.py fix(tools): add debug logging for token refresh and tighten domain check 2026-04-02 12:40:03 +11:00
mcp_oauth.py fix(mcp): consolidate OAuth handling, pick up external token refreshes (#11383) 2026-04-16 21:57:10 -07:00
mcp_oauth_manager.py fix(mcp): consolidate OAuth handling, pick up external token refreshes (#11383) 2026-04-16 21:57:10 -07:00
mcp_tool.py fix(mcp): consolidate OAuth handling, pick up external token refreshes (#11383) 2026-04-16 21:57:10 -07:00
memory_tool.py fix: nest msvcrt import inside fcntl except block 2026-04-14 10:18:05 -07:00
mixture_of_agents_tool.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
neutts_synth.py fix(tts): document NeuTTS provider and align install guidance (#1903) 2026-03-18 02:55:30 -07:00
openrouter_client.py refactor: route ad-hoc LLM consumers through centralized provider router 2026-03-11 20:02:36 -07:00
osv_check.py feat: OSV malware check for MCP extension packages (#5305) 2026-04-05 12:46:07 -07:00
patch_parser.py fix(patch): harden V4A patch parser and fuzzy match — 9 correctness bugs 2026-04-10 16:47:44 -07:00
path_security.py refactor: extract shared helpers to deduplicate repeated code patterns (#7917) 2026-04-11 13:59:52 -07:00
process_registry.py fix: prevent agent hang when backgrounding processes via terminal tool 2026-04-15 17:26:31 -07:00
registry.py fix: tighten AST check to module-level only 2026-04-14 21:12:29 -07:00
rl_training_tool.py refactor: codebase-wide lint cleanup — unused imports, dead code, and inefficient patterns (#5821) 2026-04-07 10:25:31 -07:00
send_message_tool.py retry transient telegram send failures 2026-04-16 03:47:00 -07:00
session_search_tool.py fix(session_search): coerce limit to int to prevent TypeError with non-int values (#10522) 2026-04-15 14:11:05 -07:00
skill_manager_tool.py fix: five HERMES_HOME profile-isolation leaks (#10570) 2026-04-15 17:09:41 -07:00
skills_guard.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
skills_hub.py feat(skills): centralized skills index — eliminate GitHub API calls for search/install 2026-04-12 16:39:04 -07:00
skills_sync.py fix(skills): read name from SKILL.md frontmatter in skills_sync 2026-04-11 01:21:20 -07:00
skills_tool.py fix: use absolute skill_dir for external skills (#10313) (#10587) 2026-04-15 17:22:55 -07:00
terminal_tool.py feat: ungate Tool Gateway — subscription-based access with per-tool opt-in 2026-04-16 12:36:49 -07:00
tirith_security.py fix: handle cross-device shutil.move failure in tirith auto-install (#10127) (#10524) 2026-04-15 14:50:07 -07:00
todo_tool.py fix(tools): enforce ID uniqueness in TODO store during replace operations 2026-04-11 16:22:50 -07:00
tool_backend_helpers.py feat: ungate Tool Gateway — subscription-based access with per-tool opt-in 2026-04-16 12:36:49 -07:00
tool_result_storage.py fix(tools): neutralize shell injection in _write_to_sandbox via path quoting (#7940) 2026-04-11 14:26:11 -07:00
transcription_tools.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
tts_tool.py feat(tts): add Google Gemini TTS provider (#11229) 2026-04-16 14:23:16 -07:00
url_safety.py fix: make safe_url_for_log public, add SSRF redirect guards to base.py cache helpers 2026-04-10 05:04:28 -07:00
vision_tools.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
voice_mode.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
web_tools.py feat: ungate Tool Gateway — subscription-based access with per-tool opt-in 2026-04-16 12:36:49 -07:00
website_policy.py refactor: codebase-wide lint cleanup — unused imports, dead code, and inefficient patterns (#5821) 2026-04-07 10:25:31 -07:00
xai_http.py feat(xai): upgrade to Responses API, add TTS provider 2026-04-16 02:24:08 -07:00