Review gate on #7720 (two regressions reproduced against master with the
installed Agent):
1. Deleting a profile's last MCP server left ADOPTED tools callable after
/reload-mcp. The scoped teardown only stops connections the profile owns,
and the Agent's discover_mcp_tools() returns before its reconcile step
when the profile's config has no server, so an adopted shared connection
was never judged against the config. /reload-mcp now calls
register_mcp_servers({}) under the profile's scope before rediscovery:
the Agent's overlay heal detaches only this profile's overlay for servers
its config no longer has; the owner's connection keeps running (master's
wildcard shutdown stopped the owner too).
2. Portable plugin MCP servers are merged by the Agent at runtime and never
appear in the profile's raw ``mcp_servers``; the registry fallback
required a config entry and dropped every plugin tool from
/api/mcp/tools and the Notes inventory. The profile's own registry slot
(registry_tool_owned_by_view) is the isolation check; the config is not
an allowlist any more.
Tests (red before, green after): agent-free unit tests for both (the fake
agent now models the overlay heal and discovery's empty-config early
return), and a real-Agent integration test for the adopted-then-deleted
case (owner connection survives, adopter slot emptied, "Removed" reported).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Two profiles configuring a same-named MCP server shared one connection and
the MCP settings panel showed another profile's runtime ("Active, N tools"
next to an empty inventory); /reload-mcp stopped every profile's servers and
rediscovered from the process-default config.
- api.profiles._set_hermes_home() is the single writer of the process-profile
home: it records get_process_profile_home() and hands it to Hermes Agent's
hermes_constants.pin_process_hermes_home when available, at startup and on
switch_profile(process_wide=True), so the per-turn HERMES_HOME mirror no
longer makes every turn look like the process profile (bare-name,
cross-profile MCP keys) and the anchor cannot drift from HERMES_HOME.
- Run get_mcp_status/registry reads and /reload-mcp inside
api.mcp_runtime.mcp_runtime_scope(): request profile's Hermes-home override
and secret scope (root included), no os.environ mutation, restored on exit.
- One ledger predicate, ledger_key_serves_view() (owned via
_server_scope_keys/key scope, or adopted via _server_tool_scopes), is shared
by status, inventory and the reload summary. The Agent's launch-profile view
is process-wide and name-indexed, so filter_runtime_status_to_view()
rebuilds connected/tools/status/error from the entries serving the profile:
the root profile never shows a routed profile's connection (nor lets its own
parked failed-spawn task borrow a routed profile's live same-named one).
- Inventory lists only tools in the profile's own registry slot for servers it
configures; responses carry runtime_scope.
- Fail closed when the agent's routing does not match the resolved profile:
hide runtime data (runtime_scope "unavailable", shown as a notice in the MCP
panel, 14 locales) and refuse /reload-mcp.
- /reload-mcp shuts down only the profile's own ledger names (parked
included), drops its own connect cooldown/error entries so a failed spawn is
retried (the wildcard cleared every cooldown; a scoped shutdown only clears
the live keys it tears down), rediscovers under the profile's home and
counts live connections only. The bare wildcard remains for agents without
profile-scoped MCP.
- Signature introspection is centralized (accepts_keywords) and falls back to
the bare call instead of blanking every status; ledger key helpers resolve
to the Agent's tools.mcp_tool_scope when present; tools.mcp_tool is
resolved by full name so stubbed agents still work.
- ARCHITECTURE.md §4.10 and docs/troubleshooting.md document the boundary.
- Tests: agent-free contract (models adoption, parked tasks, cooldowns and the
process-wide root view) and real-Agent integration (fake stdio MCP server,
skips without hermes-agent + mcp SDK).
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Hermes Agent moved several names the WebUI uses into sibling modules
(tools.approval_context, tools.mcp_tool_discovery/_lifecycle,
hermes_cli.kanban_db_connect/_dispatch, hermes_cli.models_local,
agent.anthropic_credentials). The old paths only resolved through
PLUGIN-COMPAT __getattr__ pointers that warn on every use and are removed
on schedule by reverting one commit.
With the pointers removed, every affected WebUI call site is wrapped in a
broad except, so nothing crashes; features silently degrade instead:
per-turn approval session-key binding is skipped, MCP discovery/status
return nothing, /reload-mcp reports "MCP runtime unavailable", Claude Code
credential linking returns None, every kanban connection raises
AttributeError, and the dispatcher reports "unavailable".
Add api.agent_compat.agent_attr, which resolves a moved name from the
original module's own namespace first (pre-split Agents and existing test
stubs), then from its new home (split Agents, with or without pointers),
then plain attribute access (non-module doubles). Importing only the new
modules would have broken pre-split Agents and bypassed the sys.modules
stubs the test suite installs on the original modules.
Names still native to their original module (_servers, _lock,
has_blocking_approval, is_session_yolo_enabled, register/unregister
gateway notify, the rest of kanban_db) are unchanged.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
A hermes-agent build that returns a non-dict (e.g. None) for an unknown
preset without raising would make resolved.update(selected) raise TypeError,
which bypasses the routes.py except RuntimeError guard and surfaces as an
unhandled 500 on /api/chat/start. Coerce a non-dict result to {} so preset
resolution degrades cleanly. Adds a regression test.
- Keep resolve_moa_preset optional so older hermes-agent installs still degrade gracefully.
- Surface MoA presets in the WebUI model picker as a virtual provider.
- Keep configured Copilot per-model settings from collapsing the built-in catalog.
- Refresh the Copilot static fallback used when the live catalog probe misses.
Verification:
./scripts/test.sh tests/test_issue5057_moa_webui_route.py tests/test_copilot_provider_model_settings_not_allowlist.py tests/test_moa_model_picker_provider.py -q
11 passed, 1 skipped
Adds a dedicated GET /api/commands/moa/resolve endpoint + resolve_moa_config()
that reads the MoA preset/usage server-side. The /moa send-path passes only a
moa_config:true boolean to /api/chat/start; the server re-resolves the real
config itself (never trusts a browser-supplied dict) and threads it per-turn into
the live agent run. Fails closed (409) on gateway-backed sessions.
Replaces the originally-proposed broad 43-command subprocess approach (bounced)
with this narrow, security-reviewed /moa route. Maintainer fixes on top of the
contributor's converged design: moa_config is added to run_conversation only
when not None (older-agent TypeError safety) at the main call AND both auth
self-heal retry sites, and forwarded through the legacy-journal runtime adapter.
Gate: Codex SAFE (after the maintainer fixes) + full suite 10784 pass + the
activity-stream regression gate clean.
Co-authored-by: rodboev <rodboev@users.noreply.github.com>
## Release v0.51.263 — Release IE (stage-r13)
Batch 1 (fresh) — trimmed to the clean pair after the gate held two.
### Fixed
| Issue/PR | Author | Fix |
|----------|--------|-----|
| #3621 | @luanxu-dev | `/codex-runtime` + `/codex_runtime` now run as a WebUI slash command (routed through the executor reusing the agent's `codex_runtime_switch`) instead of being sent to the model as a chat message. |
### Tests
| Issue/PR | Author | Change |
|----------|--------|--------|
| #3595 | @rodboev | Regression coverage for the already-shipped `activity_feed_expanded_default` setting. |
### Held back from this batch (Codex regression gate)
- **#3624** (passkey-challenge cap, security) — the cap **raises** `PasskeyRateLimitError` when full instead of **evicting oldest**, so an attacker (or 8 abandoned legit attempts per context) can lock out genuine registration/login until TTL — the protection becomes a lockout DoS. Held with the oldest-first-eviction fix.
- **#3618** (prefer server-side STT) — forcing MediaRecorder by default breaks browser `SpeechRecognition` dictation on installs with **no** server STT configured (`_transcribeBlob` only toasts on failure, never falls back). Held with the graceful-fallback fix.
### Gate
- Full pytest suite: **7714 passed, 0 failed**
- ESLint: CLEAN · ruff: CLEAN
- Codex (regression): SHIP-ONLY-WITH-FIXES (#3624 DoS-lockout + #3618 STT-no-fallback) → both dropped/held → **SAFE TO SHIP** (verified no passkeys.py/boot.js remnants, codex-runtime dispatch reaches the allowlist).
Co-authored-by: luanxu-dev <luanxu-dev@users.noreply.github.com>
Co-authored-by: rodboev <rodboev@users.noreply.github.com>
Opus stage-339 review SHOULD-FIX items:
1. server.py: drop 'unsafe-eval' from CSP report-only policy.
Verified by grepping all production JS — zero matches for eval(),
new Function(), or string-form setTimeout/setInterval. Keeping it
was a gratuitous privilege.
2. server.py: add https://cdn.jsdelivr.net to script-src + style-src.
index.html loads Prism/xterm/katex from this CDN with SRI hashes —
without the allowance every page load fires known-good CSP violations
that drown out real signal once a collector is wired.
3. api/commands.py: sanitize plugin command error. Previously returned
f'Plugin command error: {exc}' which would leak paths/env from
FileNotFoundError('/etc/something/secret.key') etc. Now returns only
the exception type name; full traceback goes to server log.
Test asserts updated to match the new policy shape.
Co-authored-by: Opus advisor <opus-advisor@hermes.local>
