Commit graph

3 commits

Author SHA1 Message Date
Till JS
8acf35eecf chore(dev): finish --watch → --hot sweep across remaining Bun services
Some checks failed
CD Mac Mini / Detect Changes (push) Failing after 11s
CI / Detect Changes (push) Successful in 7s
CI / Validate (push) Has been skipped
CI / Build mana-auth (push) Waiting to run
CI / Build mana-search (push) Waiting to run
CI / Build mana-sync (push) Waiting to run
CI / Build mana-notify (push) Waiting to run
CI / Build mana-api-gateway (push) Waiting to run
CI / Build mana-crawler (push) Waiting to run
CI / Build mana-media (push) Waiting to run
CI / Build mana-credits (push) Waiting to run
CI / Auth flow integration test (push) Has been skipped
Docker Validate / Validate Dockerfiles (push) Failing after 1m35s
Docker Validate / Build calendar-web (push) Has been skipped
Docker Validate / Build quotes-web (push) Has been skipped
Docker Validate / Build todo-backend (push) Has been skipped
Docker Validate / Build todo-web (push) Has been skipped
Docker Validate / Build mana-auth (push) Has been skipped
Docker Validate / Build mana-sync (push) Has been skipped
Docker Validate / Build mana-media (push) Has been skipped
Mirror to Forgejo / Push to Forgejo (push) Failing after 1s
CD Mac Mini / Deploy (push) Has been cancelled
Catches the service-level package.json files that the previous
sweep (4cca25ed0) missed — they don't appear in any dev:*:full
orchestrator but get invoked when someone runs `pnpm --filter
@mana/<service> dev` directly.

Touched: mana-geocoding, mana-mail, mana-subscriptions, mana-mcp,
news-ingester, mana-persona-runner, mana-research, mana-user,
plus apps/memoro (server + audio-server).

mana-ai stays on --watch on purpose: its entry uses an explicit
`Bun.serve({...})` call instead of `export default { port,
fetch }`, plus a SIGTERM/SIGINT handler that calls
`server.stop()`. --hot would replace the module without releasing
the old server reference and produce exactly the EADDRINUSE we're
trying to avoid. If mana-ai gets refactored to the standard
default-export shape, flip its dev script too.
2026-05-08 14:33:27 +02:00
Till JS
c94ab01c69 feat(mana-mcp): Prometheus metrics for policy gate + tool invocations
Replaces the stub /metrics endpoint with a real prom-client registry
(mana_mcp_ prefix, {service="mana-mcp"} default label). Default
process metrics come along for free.

Policy-gate telemetry is the whole point — without it we can't soak
POLICY_MODE=log-only safely or decide when to flip to enforce. New
counter mana_mcp_policy_decisions_total{decision, reason, mode} buckets
every evaluatePolicy() call:

  decision ∈ {allow, deny, flagged}
  reason   ∈ {admin-scope-not-invokable, destructive-not-allowed,
              rate-limit-exceeded, injection-marker, clean, unknown}
  mode     ∈ {log-only, enforce}

So the rate of "would have been denied" during soak is visible directly
as policy_decisions_total{decision="deny", mode="log-only"}.

Also:
  - mana_mcp_tool_invocations_total{tool, outcome} — success |
    handler-error | input-invalid. Policy denies are NOT counted here
    (they're in policy_decisions_total above); this counter only counts
    calls that actually reached the handler or tripped zod validation.
  - mana_mcp_tool_duration_seconds histogram per tool/outcome.

Dep: prom-client ^15.1.3 (same version mana-ai pins).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-23 14:23:08 +02:00
Till JS
16c8818338 feat(mcp): M1+M1.5 MCP gateway + tool-registry + shared-crypto
Foundation for autonomous Claude-driven testing. Plan:
docs/plans/mana-mcp-and-personas.md.

New packages
- @mana/tool-registry — schema-first ToolSpec<InputSchema, OutputSchema>
  with zod generics, scope ('user-space' | 'admin') and policyHint
  ('read' | 'write' | 'destructive'). sync-client helpers speak the
  mana-sync push/pull protocol directly so RLS and field-level LWW are
  preserved. MasterKeyClient fetches per-user MKs via the existing
  mana-auth GET /api/v1/me/encryption-vault/key endpoint (JWT-gated,
  ZK-aware, already audited) — no new service-key endpoint built.
  ZeroKnowledgeUserError surfaced as a typed throw.
- @mana/shared-crypto — AES-GCM-256 primitives extracted from the web
  app's $lib/data/crypto/aes.ts so the server-side tool handlers and the
  browser produce byte-for-byte identical wire format
  (enc:1:{b64(iv)}.{b64(ct)}). Web app aes.ts now re-exports from
  shared-crypto — 5 existing importers unchanged, svelte-check stays
  green.

New service
- services/mana-mcp (:3069, Bun/Hono) — MCP Streamable HTTP gateway.
  JWKS auth against mana-auth, per-user session isolation (session-id
  belongs to the user who opened it — cross-user access returns 403),
  admin-scoped tools filtered out before registration. MasterKeyClient
  cached per process with a 5-minute TTL.

11 tools registered
- habits.{create,list,update,archive}, spaces.list (plaintext, M1)
- todo.{create,list,complete}, notes.{create,search}, journal.add
  (encrypted — field lists match
  apps/mana/apps/web/src/lib/data/crypto/registry.ts verbatim)

Infra
- Port 3069 added to docs/PORT_SCHEMA.md
- services/mana-mcp/CLAUDE.md with architecture, auth model,
  tool-authoring recipe, local smoke-test steps
- Root CLAUDE.md services list updated

Type-check green across shared-crypto, mana-tool-registry, mana-mcp.
svelte-check on apps/mana/apps/web stays at 0 errors / 0 warnings.
Boot smoke verified: /health returns registry.loaded=true, unauthed
/mcp → 401, invalid-JWT /mcp → 401 with descriptive message.

Decisions locked in for later milestones (per plan D1–D10):
- Personas will be real mana-auth users (users.kind='persona'), no
  service-key bypass (D1, D2)
- Tool-registry is the SSOT; mana-ai and the legacy
  apps/api/src/mcp/server.ts get merged into it in M4 (three current
  parallel tool catalogs collapse to one)
- Persona-runner (:3070) will be a separate service using the Claude
  Agent SDK + MCP client (D5)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-23 13:18:35 +02:00