hermes-agent/tests/hermes_cli
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
..
__init__.py test: reorganize test structure and add missing unit tests 2026-02-26 03:20:08 +03:00
test_anthropic_oauth_flow.py refactor(tests): re-architect tests + fix CI failures (#5946) 2026-04-07 17:19:07 -07:00
test_anthropic_provider_persistence.py refactor(tests): re-architect tests + fix CI failures (#5946) 2026-04-07 17:19:07 -07:00
test_api_key_providers.py chore: rename AI Gateway → Vercel AI Gateway, move Xiaomi to #5 (#9326) 2026-04-13 19:51:54 -07:00
test_arcee_provider.py feat(providers): add Arcee AI as direct API provider 2026-04-13 18:40:06 -07:00
test_argparse_flag_propagation.py fix: --yolo and other flags silently dropped when placed before 'chat' subcommand (#5145) 2026-04-04 16:55:13 -07:00
test_atomic_json_write.py refactor(tests): re-architect tests + fix CI failures (#5946) 2026-04-07 17:19:07 -07:00
test_atomic_yaml_write.py refactor(tests): re-architect tests + fix CI failures (#5946) 2026-04-07 17:19:07 -07:00
test_auth_codex_provider.py fix: write refreshed Codex tokens back to ~/.codex/auth.json (#8277) 2026-04-12 02:05:20 -07:00
test_auth_commands.py fix: resolve remaining 4 CI test failures (#9543) 2026-04-14 02:18:38 -07:00
test_auth_nous_provider.py fix: dashboard shows Nous Portal as 'not connected' despite active auth (#9261) 2026-04-13 16:32:11 -07:00
test_auth_provider_gate.py fix: resolve CI test failures — add missing functions, fix stale tests (#9483) 2026-04-14 01:43:45 -07:00
test_auth_qwen_provider.py feat(qwen): add Qwen OAuth provider with portal request support 2026-04-08 13:46:30 -07:00
test_backup.py feat: fix SQLite safety in hermes backup + add --quick snapshots + /snapshot command (#8971) 2026-04-13 04:46:13 -07:00
test_banner.py fix(banner): normalize toolset labels and use skin colors 2026-03-18 03:22:58 -07:00
test_banner_git_state.py fix: CLI/UX batch — ChatConsole errors, curses scroll, skin-aware banner, git state banner (#5974) 2026-04-07 17:59:42 -07:00
test_banner_skills.py fix: disabled skills respected across banner, system prompt, slash commands, and skill_view (#1897) 2026-03-18 03:17:37 -07:00
test_chat_skills_flag.py fix(termux): add local image chat route 2026-04-09 16:24:53 -07:00
test_claw.py fix: unify OpenClaw detection, add isatty guard, fix print_warning import 2026-04-12 16:40:37 -07:00
test_clear_stale_base_url.py fix: warn and clear stale OPENAI_BASE_URL on provider switch (#5161) 2026-04-11 01:52:58 -07:00
test_cmd_update.py fix(update): skip config migration prompts in non-interactive sessions (#3584) 2026-03-28 14:26:32 -07:00
test_coalesce_session_args.py fix(cli): handle unquoted multi-word session names in -c/--continue and -r/--resume 2026-03-09 21:36:29 -07:00
test_codex_cli_model_picker.py fix: openai-codex and anthropic not appearing in /model picker for external credentials (#8224) 2026-04-12 00:33:42 -07:00
test_codex_models.py fix(model): normalize native provider-prefixed model ids 2026-04-10 05:52:45 -07:00
test_commands.py fix: remove 'q' alias from /quit so /queue's 'q' alias works (#10467) (#10538) 2026-04-15 15:04:01 -07:00
test_completion.py fix: preserve profile name completion in dynamic shell completion 2026-04-14 10:45:42 -07:00
test_config.py feat(discord): add channel_prompts config 2026-04-15 16:31:28 -07:00
test_config_env_expansion.py refactor(tests): re-architect tests + fix CI failures (#5946) 2026-04-07 17:19:07 -07:00
test_config_validation.py feat: config structure validation — detect malformed YAML at startup (#5426) 2026-04-05 23:31:20 -07:00
test_container_aware_cli.py fix(gateway): harden Docker/container gateway pathway 2026-04-12 16:36:11 -07:00
test_copilot_auth.py fix: remove 115 verified dead code symbols across 46 production files 2026-04-10 03:44:43 -07:00
test_cron.py feat: add multi-skill cron editing and docs 2026-03-14 19:18:10 -07:00
test_custom_provider_model_switch.py fix: three provider-related bugs (#8161, #8181, #8147) (#8243) 2026-04-12 01:44:18 -07:00
test_debug.py fix: /debug privacy — auto-delete pastes after 1 hour, add privacy notices (#10510) 2026-04-15 13:40:27 -07:00
test_deprecated_cwd_warning.py fix: enforce config.yaml as sole CWD source + deprecate .env CWD vars + add hermes memory reset (#11029) 2026-04-16 06:48:33 -07:00
test_doctor.py fix(doctor): skip health check for OpenCode Go (no shared /models endpoint) 2026-04-15 15:05:32 -07:00
test_doctor_command_install.py feat(doctor): add Command Installation check for hermes bin symlink 2026-04-14 23:13:11 -07:00
test_env_loader.py fix(config): reload .env over stale shell overrides 2026-03-15 06:46:28 -07:00
test_env_sanitize_on_load.py fix: follow-up for salvaged PR #8939 2026-04-13 04:35:37 -07:00
test_gateway.py fix: add all_profiles param + narrow exception handling 2026-04-11 14:44:29 -07:00
test_gateway_linger.py fix(termux): disable gateway service flows on android 2026-04-09 16:24:53 -07:00
test_gateway_runtime_health.py fix(gateway): harden Telegram polling conflict handling 2026-03-14 12:11:23 -07:00
test_gateway_service.py fix(tests): resolve CI test failures — pool auto-seeding, stale assertions, mock isolation 2026-04-15 22:05:21 -07:00
test_gateway_wsl.py feat(gateway): WSL-aware gateway with smart systemd detection (#7510) 2026-04-10 21:15:47 -07:00
test_gemini_provider.py refactor(tests): re-architect tests + fix CI failures (#5946) 2026-04-07 17:19:07 -07:00
test_launcher.py fix: use argparse entrypoint in top-level launcher (#3874) 2026-03-29 21:54:36 -07:00
test_logs.py feat: component-separated logging with session context and filtering (#7991) 2026-04-11 17:23:36 -07:00
test_managed_installs.py chore: prepare Hermes for Homebrew packaging (#4099) 2026-03-30 17:34:43 -07:00
test_mcp_config.py fix(mcp): consolidate OAuth handling, pick up external token refreshes (#11383) 2026-04-16 21:57:10 -07:00
test_mcp_tools_config.py feat: interactive MCP tool configuration in hermes tools (#1694) 2026-03-17 03:48:44 -07:00
test_memory_reset.py fix: enforce config.yaml as sole CWD source + deprecate .env CWD vars + add hermes memory reset (#11029) 2026-04-16 06:48:33 -07:00
test_model_normalize.py fix: preserve dots in model names for OpenCode Zen and ZAI providers (#8794) 2026-04-12 21:22:59 -07:00
test_model_provider_persistence.py fix(setup): validate base URL input in hermes model flow (#8264) 2026-04-12 01:51:57 -07:00
test_model_switch_copilot_api_mode.py fix: recompute Copilot api_mode after model switch 2026-04-16 01:16:14 -07:00
test_model_switch_custom_providers.py test: add regression tests for custom_providers multi-model dedup and grouping 2026-04-13 16:41:30 -07:00
test_model_switch_opencode_anthropic.py fix(opencode): strip /v1 from base_url on mid-session /model switch to Anthropic-routed models (#11286) 2026-04-16 19:41:41 -07:00
test_model_switch_variant_tags.py fix(models): preserve OpenRouter variant tags (:free, :extended, :fast) during model switch (#6383) 2026-04-08 19:58:16 -07:00
test_model_validation.py fix(models): add glm-5.1 to opencode-go catalogs 2026-04-16 16:49:22 -07:00
test_models.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
test_non_ascii_credential.py fix: detect and strip non-ASCII characters from API keys (#6843) 2026-04-14 20:20:31 -07:00
test_nous_hermes_non_agentic.py fix(cli): narrow Nous Hermes non-agentic warning to actual hermes-3/-4 models 2026-04-13 04:33:52 -07:00
test_nous_subscription.py feat: ungate Tool Gateway — subscription-based access with per-tool opt-in 2026-04-16 12:36:49 -07:00
test_ollama_cloud_auth.py refactor(tests): re-architect tests + fix CI failures (#5946) 2026-04-07 17:19:07 -07:00
test_ollama_cloud_provider.py fix: wire up Ollama Cloud dynamic model discovery in /model TUI picker 2026-04-16 07:17:45 -07:00
test_opencode_go_in_model_list.py fix(models): add glm-5.1 to opencode-go catalogs 2026-04-16 16:49:22 -07:00
test_overlay_slug_resolution.py fix: resolve overlay provider slug mismatch in /model picker (#7373) 2026-04-10 14:46:57 -07:00
test_path_completion.py feat(cli): add file path autocomplete in the input prompt (#1545) 2026-03-16 06:07:45 -07:00
test_placeholder_usage.py fix: cover remaining config placeholder help text 2026-03-14 10:35:14 -07:00
test_plugin_cli_registration.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
test_plugins.py feat(plugins): add dispatch_tool() to PluginContext (#10763) 2026-04-15 22:23:01 -07:00
test_plugins_cmd.py fix: no auto-activation + unified hermes plugins UI with provider categories 2026-04-10 19:15:50 -07:00
test_profile_export_credentials.py fix: also exclude .env from default profile exports 2026-04-01 11:20:33 -07:00
test_profiles.py fix: improve profile creation UX — seed SOUL.md + credential warning (#8553) 2026-04-12 12:22:34 -07:00
test_reasoning_effort_menu.py fix: normalize reasoning effort ordering in UI 2026-04-09 14:20:16 -07:00
test_runtime_provider_resolution.py fix(config): restore custom providers after v11→v12 migration 2026-04-13 10:50:52 -07:00
test_session_browse.py feat: interactive session browser with search filtering (#718) 2026-03-08 17:42:50 -07:00
test_sessions_delete.py fix(cli): handle EOFError in sessions delete/prune confirmation prompts (#3101) 2026-03-25 18:06:04 -07:00
test_set_config_value.py fix(cli): allow empty strings and falsy values in config set 2026-03-31 11:41:12 -07:00
test_setup.py feat: ungate Tool Gateway — subscription-based access with per-tool opt-in 2026-04-16 12:36:49 -07:00
test_setup_hermes_script.py fix(termux): make setup-hermes use android path 2026-04-09 16:24:53 -07:00
test_setup_matrix_e2ee.py docs(matrix): update all references from matrix-nio to mautrix 2026-04-10 21:15:59 -07:00
test_setup_model_provider.py fix: remove 115 verified dead code symbols across 46 production files 2026-04-10 03:44:43 -07:00
test_setup_noninteractive.py Harden setup provider flows 2026-04-10 02:57:39 -07:00
test_setup_openclaw_migration.py fix: OpenClaw migration now shows dry-run preview before executing (#6769) 2026-04-09 12:15:06 -07:00
test_setup_prompt_menus.py fix(ci): resolve 4 pre-existing main failures (docs lint + 3 stale tests) (#11373) 2026-04-16 20:43:41 -07:00
test_skills_config.py fix: respect per-platform disabled skills in Telegram menu and gateway dispatch (#4799) 2026-04-03 10:10:53 -07:00
test_skills_hub.py fix(cli): add ChatConsole.status for /skills search 2026-04-11 15:38:43 -07:00
test_skills_install_flags.py fix: add --yes flag to bypass confirmation in /skills install and uninstall (#1647) 2026-03-17 01:59:07 -07:00
test_skills_skip_confirm.py fix(skills): cache-aware /skills install and uninstall in TUI (#3586) 2026-03-28 14:32:23 -07:00
test_skills_subparser.py fix(cli): resolve duplicate 'skills' subparser crash on Python 3.11+ 2026-03-11 00:50:39 -07:00
test_skin_engine.py fix(cli): handle null/non-dict display config in skin initialization 2026-04-16 06:35:31 -07:00
test_status.py fix(termux): improve status and install UX 2026-04-09 16:24:53 -07:00
test_status_model_provider.py feat: ungate Tool Gateway — subscription-based access with per-tool opt-in 2026-04-16 12:36:49 -07:00
test_subparser_routing_fallback.py fix(cli): defensive subparser routing for argparse bpo-9338 (#10113) 2026-04-14 23:13:02 -07:00
test_subprocess_timeouts.py fix(cli): add missing subprocess.run() timeouts in doctor and status (#4009) 2026-03-30 11:17:15 -07:00
test_terminal_menu_fallbacks.py Harden setup provider flows 2026-04-10 02:57:39 -07:00
test_tips.py refactor: remove dead code — 1,784 lines across 77 files (#9180) 2026-04-13 16:32:04 -07:00
test_tool_token_estimation.py fix(tests): resolve 10 CI failures across hooks, tiktoken, plugins (#3848) 2026-03-29 20:05:59 -07:00
test_tools_config.py feat(image_gen): multi-model FAL support with picker in hermes tools (#11265) 2026-04-16 20:19:53 -07:00
test_tools_disable_enable.py fix: MCP toolset resolution for runtime and config (#3252) 2026-03-26 13:39:41 -07:00
test_update_autostash.py fix(update): always reset on stash conflict — never leave conflict markers (#7010) 2026-04-10 00:32:20 -07:00
test_update_check.py fix: profile paths broken in Docker — profiles go to /root/.hermes instead of mounted volume (#7170) 2026-04-10 05:53:10 -07:00
test_update_gateway_restart.py fix: write update exit code before gateway restart (cgroup kill race) (#8288) 2026-04-12 02:33:21 -07:00
test_user_providers_model_switch.py fix(model): Support providers: dict for custom endpoints in /model 2026-04-13 05:16:21 -07:00
test_web_server.py dashboard: show GATEWAY_HEALTH_URL instead of PID for remote gateways 2026-04-16 16:48:14 -07:00
test_webhook_cli.py feat(webhook): hermes webhook CLI + skill for event-driven subscriptions (#3578) 2026-03-28 14:33:35 -07:00
test_xiaomi_provider.py docs: add Xiaomi MiMo to all provider docs + fix MiMo-V2-Flash ctx len 2026-04-11 11:17:52 -07:00