Initial commit: SIC harness (backend, web, pi-adapter, configs, docs)

- pnpm monorepo: apps/api (Fastify + SQLite + SSE), apps/web (React+Vite), packages/shared, packages/pi-adapter
- Local auth (admin/webhook-runner roles) + Keycloak JWT ready
- Multi-session chat with reliable history (user persisted before LLM, assistant persisted after stream)
- Markdown knowledge base with /api/docs/search + /api/docs/:id
- YAML webhook catalog with backend-only execution, retry/backoff, audit (webhook_runs), and per-user rate limit
- Skills config (sre-on-call, blameless-postmortem, security-incident) injected into LLM system prompt
- LLM provider failover chain (config/models.yml fallback + LLM_FALLBACK_CHAIN override)
- Context-aware webhooks panel + backend id-mention safety net
- Per-message stats (time/duration/tokens/model), Markdown+GFM render, code & table copy/download buttons
- Vitest suite, end-to-end smoke test (scripts/smoke.mjs), per-session system prompt override
- /metrics Prometheus endpoint + /api/metrics JSON, request-id correlation
- dotenv with explicit repo-root path; envString/envNumber helpers (handles empty-string env)
- Runbooks + SOPs under knowledge/ in English; README, docs, and INDEX.md in English

config/mcp.yml

@@ -0,0 +1,75 @@
+# MCP (Model Context Protocol) tool catalog.
+#
+# Phase 1: this is a DECLARATIVE catalog of available tools. The SIC backend
+# does NOT call any MCP server directly. The LLM sees these tools in its
+# context (so it can reason about what's possible) and the right panel
+# renders them as recommendations that the user can inspect. Execution is
+# reserved for the future MCP runtime, which will run a per-request MCP
+# client; for now every tool here is "inspect only".
+#
+# Each tool:
+#   id            stable identifier
+#   name          human-readable label
+#   description   what the tool does (sent to the LLM verbatim)
+#   server        optional reference to a configured MCP server (Phase 2+)
+#   parameters    JSON Schema for the tool arguments
+#   tags          free-form tags for filtering
+#   enabled       whether the tool is exposed at all
+mcp_servers: []
+
+mcp_tools:
+  - id: kb.search
+    name: Search knowledge base
+    description: |
+      Search the internal knowledge base (runbooks, SOPs, architecture docs) for
+      documents relevant to the query. Returns matching document ids and
+      relevance scores. Does NOT execute anything.
+    server: null
+    parameters:
+      type: object
+      required: [query]
+      properties:
+        query:
+          type: string
+          description: Natural-language search query.
+        limit:
+          type: integer
+          minimum: 1
+          maximum: 20
+          description: Maximum results to return. Defaults to 5.
+    tags: [search, knowledge, docs]
+    enabled: true
+  - id: sessions.list
+    name: List recent chat sessions
+    description: |
+      List the calling user's most recent chat sessions, ordered by updated_at
+      descending. Does NOT execute anything; returns metadata only.
+    server: null
+    parameters:
+      type: object
+      required: []
+      properties:
+        limit:
+          type: integer
+          minimum: 1
+          maximum: 50
+          description: Maximum sessions to return. Defaults to 10.
+    tags: [sessions, listing]
+    enabled: true
+  - id: webhooks.usage
+    name: Webhook usage stats
+    description: |
+      Return per-webhook usage stats for the calling user over a recent time
+      window (defaults to 7 days). Read-only.
+    server: null
+    parameters:
+      type: object
+      required: []
+      properties:
+        days:
+          type: integer
+          minimum: 1
+          maximum: 90
+          description: Window in days. Defaults to 7.
+    tags: [webhooks, audit, read-only]
+    enabled: true