feat(batch-agent): extract Batch Agent Service (Step 3)

- agent_runner: local directory + cloud agent orchestration via Redis
- 5 domain agents: filesystem, task, note, project, timeline
- integrations: Gmail, MS Graph (Outlook + Teams)
- journey: guided chatbot conversation to build prompt_template
- routes: REST endpoints (catalog, can-create, trigger)
- redis_consumer: subscribes to batch:request:* pattern
- ws_context: Redis-based execute_on_client for tool round-trip
- Dockerfile with 300s timeout for long-running batch jobs

.env.example

@@ -4,9 +4,17 @@ ENV=dev
 # ── Database ──────────────────────────────────────────────────────────────────
 DATABASE_URL=postgresql+asyncpg://postgres:postgres@localhost:5432/adiuva
 
-# ── Auth ──────────────────────────────────────────────────────────────────────
+# ── Redis ─────────────────────────────────────────────────────────────────────
-JWT_SECRET=replace-with-a-long-random-secret
+REDIS_URL=redis://localhost:6379/0
-JWT_ALGORITHM=HS256
+
+# ── Auth (JWT RS256) ──────────────────────────────────────────────────────────
+# Public key for optional local JWT verification (Traefik ForwardAuth handles
+# this in production — services trust X-User-* headers from Traefik).
+# Generate keypair:
+#   openssl genpkey -algorithm RSA -out private.pem -pkeyopt rsa_keygen_bits:2048
+#   openssl rsa -in private.pem -pubout -out public.pem
+# Paste PEM content with literal \n for newlines.
+JWT_PUBLIC_KEY=
 JWT_ACCESS_TOKEN_EXPIRE_MINUTES=30
 JWT_REFRESH_TOKEN_EXPIRE_DAYS=30
 
@@ -17,7 +25,6 @@ OPENAI_API_KEY=
 ANTHROPIC_API_KEY=
 GOOGLE_API_KEY=
 LLM_MODEL=gpt-4o
-LLM_ROUTER_MODEL=gpt-4o-mini
 
 # ── Stripe (leave empty to stub billing) ──────────────────────────────────────
 STRIPE_SECRET_KEY=
@@ -42,3 +49,8 @@ QDRANT_API_KEY=
 # ── CORS ──────────────────────────────────────────────────────────────────────
 # Comma-separated list parsed by Settings (override default if needed)
 # CORS_ORIGINS=["app://.","http://localhost:3000"]
+
+# ── Langfuse (observability) ─────────────────────────────────────────────────
+LANGFUSE_SECRET_KEY=sk-lf-...
+LANGFUSE_PUBLIC_KEY=pk-lf-...
+LANGFUSE_HOST=https://cloud.langfuse.com  # or self-hosted URL

.gitignore

@@ -13,6 +13,9 @@ env/
 # Environment variables
 .env
 
+# Cryptographic keys
+*.pem
+
 # IDE
 .vscode/
 .idea/
@@ -31,3 +34,4 @@ Thumbs.db
 
 # Claude Code
 .claude/
+logs/

BACKEND_PLAN.md

@@ -1,533 +0,0 @@
-# Backend Plan — Adiuva Cloud API
-
-> **Separate repository.** This document defines the FastAPI backend that the Electron app communicates with.
->
-> The backend owns: orchestration logic, chat agent intelligence, prompt IP, auth, billing, E2E backup blob storage, cloud storage (encrypted blobs), cloud vector store, and plugin marketplace.
-> The backend NEVER persists user data in plaintext. Cloud storage blobs are E2E encrypted before upload — the backend only verifies integrity, never decrypts.
-
----
-
-## Project Structure
-
-```
-adiuva-api/
-├── app/
-│   ├── __init__.py
-│   ├── main.py                    # FastAPI entry + CORS + lifespan + router includes
-│   ├── core/
-│   │   ├── __init__.py
-│   │   ├── agent_registry.py      # Base classes + singleton registry
-│   │   ├── orchestrator.py        # LLM-based intent router
-│   │   ├── execution_plan.py      # Plan builder + cache
-│   │   └── plugin_loader.py       # Dynamic agent loading
-│   ├── agents/                    # Chat agents (proprietary logic + prompts)
-│   │   ├── __init__.py            # Auto-registers all agents
-│   │   ├── task_agent.py
-│   │   ├── calendar_agent.py
-│   │   ├── email_agent.py
-│   │   └── analytics_agent.py
-│   ├── api/
-│   │   ├── __init__.py
-│   │   ├── routes/
-│   │   │   ├── __init__.py
-│   │   │   ├── chat.py            # POST /chat + WS /chat/stream
-│   │   │   ├── plans.py           # GET /plans/playbook
-│   │   │   ├── storage.py         # CRUD cloud storage (E2E encrypted blobs)
-│   │   │   ├── vectors.py         # Upsert/search cloud vector store
-│   │   │   ├── backup.py          # PUT/GET /backup
-│   │   │   ├── plugins.py         # Plugin marketplace
-│   │   │   ├── auth.py            # Register/login/refresh
-│   │   │   └── billing.py         # Checkout/webhook/subscription
-│   │   └── middleware/
-│   │       ├── __init__.py
-│   │       ├── auth.py            # JWT validation
-│   │       ├── rate_limit.py      # Tier-aware rate limiting
-│   │       └── sanitizer.py       # Strip prompt metadata from responses
-│   ├── storage/
-│   │   ├── __init__.py
-│   │   ├── blob_store.py          # S3 for E2E encrypted blobs
-│   │   ├── vector_store.py        # Cloud vector store (Pinecone/Qdrant)
-│   │   └── encryption.py          # Integrity verification only — NO decryption
-│   ├── marketplace/
-│   │   ├── __init__.py
-│   │   ├── plugin_registry.py     # Plugin catalog (metadata, versions, ratings)
-│   │   ├── plugin_review.py       # Review queue + approval workflow
-│   │   └── revenue_share.py       # 70/30 split tracking with Stripe Connect
-│   ├── billing/
-│   │   ├── __init__.py
-│   │   ├── stripe_service.py      # Stripe checkout + webhooks
-│   │   └── tier_manager.py        # Feature matrix per tier
-│   └── config/
-│       ├── __init__.py
-│       └── settings.py            # Pydantic BaseSettings (env-based)
-├── tests/
-│   ├── __init__.py
-│   ├── conftest.py                # Fixtures: test client, mock agents, mock LLM
-│   ├── test_orchestrator.py
-│   ├── test_agents.py
-│   ├── test_auth.py
-│   ├── test_backup.py
-│   ├── test_storage.py
-│   └── test_plugins.py
-├── alembic/                       # DB migrations (auth/billing/marketplace tables only)
-│   ├── alembic.ini
-│   └── versions/
-├── requirements.txt
-├── Dockerfile
-├── docker-compose.yml             # App + PostgreSQL + Redis (dev)
-├── .env.example
-└── README.md
-```
-
----
-
-## Step-by-Step Implementation
-
-### Step 1 — Project scaffolding ✅
-- [x] Initialize repo with the directory structure above
-- [x] Write `requirements.txt`:
-  ```
-  fastapi>=0.115.0
-  uvicorn[standard]>=0.34.0
-  langchain>=0.3.0
-  langchain-openai>=0.3.0
-  pydantic>=2.10.0
-  python-jose[cryptography]>=3.3.0
-  stripe>=11.0.0
-  boto3>=1.35.0
-  slowapi>=0.1.9
-  sqlalchemy>=2.0.0
-  asyncpg>=0.30.0
-  alembic>=1.14.0
-  bcrypt>=4.2.0
-  python-dotenv>=1.0.0
-  httpx>=0.28.0
-  websockets>=14.0
-  pytest>=8.0.0
-  pytest-asyncio>=0.24.0
-  ```
-- [x] Write `app/main.py`: FastAPI app with CORS (allow `app://`, `http://localhost:*`), lifespan (init DB pool, init agent registry), include all routers under `/api/v1`
-- [x] Write `app/config/settings.py`: `Settings(BaseSettings)` with fields: `DATABASE_URL`, `JWT_SECRET`, `JWT_ALGORITHM` (default HS256), `STRIPE_SECRET_KEY`, `STRIPE_WEBHOOK_SECRET`, `S3_BUCKET`, `S3_REGION`, `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`, `OPENAI_API_KEY`, `CORS_ORIGINS`, `ENV` (dev/prod), `PINECONE_API_KEY`, `PINECONE_INDEX`, `QDRANT_URL`, `QDRANT_API_KEY`
-- [x] Write `Dockerfile`: Python 3.12 slim, multi-stage (builder + runtime), non-root user
-- [x] Write `docker-compose.yml`: app, postgres:16, optional redis
-- [x] Write `.env.example`
-- **Outcome:** Runnable FastAPI skeleton (returns 404 on all routes).
-
-### Step 2 — Pydantic schemas (API contracts) ✅
-- [x] Create `app/schemas.py` (mirrors `src/shared/api-types.ts` from Electron repo):
-  - `ChatRequest`: `message: str`, `context: ChatContext`, `execution_mode: Literal['direct', 'plan']`
-  - `ChatContext`: `user_profile: dict`, `relevant_documents: list[str]`, `recent_tasks: list[dict]`, `conversation_history: list[dict]`
-  - `ChatResponse`: `response: str`, `actions: list[PlanAction]`
-  - `PlanAction`: `type: Literal['create_record', 'update_record', 'delete_record', 'index_document', 'send_notification', 'call_agent']`, `table: str | None`, `data: dict | None`, `agent: str | None`
-  - `ExecutionPlan`: `agent: str`, `steps: list[PlanStep]`
-  - `PlanStep`: `action: str`, `prompt_template: str | None`, `variables: dict | None`, `data_from_step: int | None`
-  - `BackupMetadata`: `version: int`, `timestamp: int`, `checksum: str`, `chunk_count: int`
-  - `BillingTier`: `Literal['free', 'pro', 'power', 'team']`
-  - `AuthTokens`: `access_token: str`, `refresh_token: str`, `expires_at: int`
-  - `UserProfile`: `id: str`, `email: str`, `tier: BillingTier`
-  - `StorageRecord`: `id: str`, `user_id: str`, `table: str`, `blob: bytes`, `checksum: str`, `created_at: int`, `updated_at: int` — blob is always E2E encrypted by client
-  - `StorageRecordCreate`: `table: str`, `blob: bytes`, `checksum: str`
-  - `StorageRecordUpdate`: `blob: bytes`, `checksum: str`
-  - `VectorUpsertRequest`: `vectors: list[VectorItem]`
-  - `VectorItem`: `id: str`, `blob: bytes`, `checksum: str` — vector + metadata encrypted by client
-  - `VectorSearchRequest`: `query_blob: bytes`, `top_k: int = 10`
-  - `VectorSearchResponse`: `results: list[VectorSearchResult]`
-  - `VectorSearchResult`: `id: str`, `score: float`, `blob: bytes`
-  - `PluginManifest`: `id: str`, `name: str`, `description: str`, `version: str`, `author: str`, `permissions: list[str]`, `category: str`, `price_cents: int = 0`
-  - `PluginListResponse`: `plugins: list[PluginManifest]`, `total: int`, `page: int`
-  - `PluginInstallRequest`: `plugin_id: str`
-- **Outcome:** All request/response models defined and validated.
-
-### Step 3 — Agent Registry + base classes ✅
-- [x] `app/core/agent_registry.py`:
-  - `BaseAgent(ABC)`:
-    - `user_id: str`, `shared_memory: dict`, `vector_store_context: list[str]`, `skills: list[str]`
-    - Abstract `get_name() -> str`, `get_description() -> str`
-  - `ChatAgent(BaseAgent)`:
-    - Abstract `async handle(query: str, context: dict) -> str`
-    - Abstract `get_tools() -> list` (LangChain tool definitions)
-    - Concrete `_tool_loop(llm, messages, tools, max_iter=5) -> str` — shared tool-calling loop
-  - `AgentRegistry` (singleton):
-    - `_agents: dict[str, ChatAgent]`
-    - `register(agent_class)` — decorator pattern
-    - `get(name) -> ChatAgent`
-    - `list_agents() -> list[dict]` — returns `[{name, description}]` for orchestrator prompt
-    - `async call_agent(name, query, context) -> str` — for inter-agent calls
-- [x] Unit tests: register, get, list, call_agent with mock
-- **Outcome:** Pluggable agent framework.
-
-### Step 4 — Orchestrator ✅
-- [x] `app/core/orchestrator.py`:
-  - `async classify_intent(message, context, registry) -> str`:
-    - System prompt: "You are an intent classifier. Given the user message and context, decide which agent to route to. Available agents: {registry.list_agents()}. Respond with just the agent name."
-    - Uses gpt-4o-mini via LangChain for low latency
-    - Falls back to `task_agent` if no clear match
-  - `async route_single(agent_name, message, context) -> ChatResponse`:
-    - Instantiates agent from registry
-    - Calls `agent.handle(message, context)`
-    - Returns response + any actions the agent produced
-  - `async route_pipeline(agent_names, message, context) -> ChatResponse`:
-    - Executes agents in sequence
-    - Each agent receives `{...context, previous_results: [...]}`
-    - Final synthesis via LLM: "Summarize these agent results into a coherent response"
-  - `async orchestrate(request: ChatRequest) -> ChatResponse | ExecutionPlan`:
-    - Main entry point
-    - Context is transparent to orchestrator — data may originate from local or cloud storage on the client side
-    - Classifies intent
-    - If `execution_mode == 'direct'`: route + return response
-    - If `execution_mode == 'plan'`: route + return execution plan with template IDs
-  - `async orchestrate_stream(request: ChatRequest) -> AsyncGenerator[str, None]`:
-    - Same as orchestrate but yields tokens for WebSocket streaming
-- [x] Integration tests with mocked LLM and mocked agents
-- **Outcome:** Intelligent routing with single-agent and pipeline modes.
-
-### Step 5 — Execution Plan generator ✅
-- [x] `app/core/execution_plan.py`:
-  - `PromptTemplateRegistry`: dict of `template_id -> prompt_text`. Templates are server-side only — client receives IDs.
-  - `ExecutionPlanBuilder`:
-    - `add_step(action, params) -> self`
-    - `add_llm_step(template_id, variables) -> self`
-    - `add_data_step(action, data_from_step) -> self`
-    - `build() -> ExecutionPlan` — validates step references
-  - `PlanCache`:
-    - In-memory LRU (maxsize=1000)
-    - `cache_plan(key, plan)`, `get_plan(key)`, `get_all_playbooks() -> list[ExecutionPlan]`
-    - Playbooks are pre-built plans for common operations (e.g., "create task from email", "generate weekly report")
-- **Outcome:** Plans are cacheable as playbooks. Prompt IP never leaves the server.
-
-### Step 6 — Chat Agents ✅
-- [x] `app/agents/task_agent.py` — `@registry.register`:
-  - Description: "Manages tasks and comments: list, create, update, delete, due-today, comments"
-  - Tools (8): `list_tasks(project_id, status, search, order_by)`, `create_task(title, description, status, priority, assignees, due_date, project_id, is_ai_suggested, is_approved)`, `update_task(task_id, ...)`, `delete_task(task_id)`, `list_tasks_due_today()`, `list_task_comments(task_id)`, `add_task_comment(task_id, author, content)`, `delete_task_comment(comment_id)`
-  - status: `todo|in_progress|done`; priority: `high|medium|low`; assignees: JSON-encoded string; due_date: ms timestamp
-  - Accepts flexible context; sentinel `-1` for optional integer update fields
-- [x] `app/agents/checkpoint_agent.py` — `@registry.register`:
-  - Description: "Manages project checkpoints (milestones): list, create, update, delete"
-  - Tools (4): `list_checkpoints(project_id)`, `create_checkpoint(project_id, title, date, is_ai_suggested, is_approved)`, `update_checkpoint(checkpoint_id, ...)`, `delete_checkpoint(checkpoint_id)`
-  - `project_id` is required for create; date is a ms timestamp; supports AI-suggestion + approval workflow
-- [x] `app/agents/project_agent.py` — `@registry.register`:
-  - Description: "Manages projects: list, get, create, update, archive, delete"
-  - Tools (6): `list_projects(client_id, include_archived)`, `list_all_projects()`, `get_project(project_id)`, `create_project(name, client_id)`, `update_project(project_id, ...)`, `delete_project(project_id)`
-  - status: `active|archived`; prefers archive over deletion (docstring guard on delete)
-- [x] `app/agents/note_agent.py` — `@registry.register`:
-  - Description: "Manages notes: list, get, create, update, delete"
-  - Tools (5): `list_notes(project_id)`, `get_note(note_id)`, `create_note(title, content, project_id)`, `update_note(note_id, ...)`, `delete_note(note_id)`
-  - content is Markdown; `get_note` should be called before update to preserve existing content
-- [x] `app/agents/__init__.py`: imports all four agent modules to trigger `@registry.register` decorators
-- [x] Unit tests per agent with mocked LLM (registration, names, tool counts, handle(), direct tool invocation)
-- **Outcome:** Four domain-specific agents matching the UI data model (Tasks, Checkpoints, Projects, Notes), all registered and tested.
-
-### Step 7 — Storage Layer ✅
-- [x] `app/storage/blob_store.py`:
-  - `BlobStore`: `async upload`, `async download`, `async delete` (idempotent), `async list_keys`
-  - Keys: `{user_id}/{table}/{record_id}` — backend never inspects blob content
-  - boto3 S3 with SSE-S3 at-rest encryption; client checksum stored in S3 object metadata
-- [x] `app/storage/vector_store.py`:
-  - `VectorStore`: `async upsert`, `async search`, `async delete`
-  - Pinecone (default, `namespace=user_id`) or Qdrant (`user_id` payload filter) — runtime-configurable
-  - 32-dim SHA-256-derived float vector; blob stored as base64 in metadata/payload
-  - ANN on encrypted data: known accuracy trade-off, documented
-- [x] `app/storage/encryption.py`:
-  - `verify_checksum(blob, checksum) -> bool` — SHA-256 + `hmac.compare_digest` (constant-time)
-  - `reject_if_tampered(blob, checksum)` — raises `HTTP 400` on mismatch
-  - Backend NEVER holds decryption keys
-- [x] `app/schemas.py`: added `StorageRecord*`, `VectorItem`, `VectorUpsertRequest`, `VectorSearch*`, `Plugin*` schemas
-- [x] `app/config/settings.py`: added `PINECONE_API_KEY`, `PINECONE_INDEX`, `QDRANT_URL`, `QDRANT_API_KEY`
-- [x] `requirements.txt`: added `moto[s3]`, `pinecone`, `qdrant-client`
-- [x] 37 unit tests covering encryption, BlobStore (moto), VectorStore Pinecone, VectorStore Qdrant
-- **Outcome:** Cloud storage layer that handles E2E encrypted blobs without ever accessing plaintext.
-
-### Step 8 — API Routes ✅
-
-#### 8a — Chat endpoint
-- [x] `app/api/routes/chat.py`:
-  - `POST /api/v1/chat`:
-    - Request: `ChatRequest`
-    - Calls `orchestrate(request)` or `orchestrate()` + `build_plan()`
-    - Response: `ChatResponse` or `ExecutionPlan`
-  - `WebSocket /api/v1/chat/stream`:
-    - Client sends `ChatRequest` as first JSON frame
-    - Server yields token strings via `orchestrate_stream()`
-    - Final frame: JSON `ChatResponse` with `{"done": true, "response": "...", "actions": [...]}`
-    - Heartbeat ping every 30s to keep connection alive
-
-#### 8b — Plans endpoint
-- [x] `app/api/routes/plans.py`:
-  - `GET /api/v1/plans/playbook`: Returns all playbooks available for the user's tier
-  - `GET /api/v1/plans/playbook/{plan_id}`: Returns a specific plan
-
-#### 8c — Storage endpoint (cloud records)
-- [x] `app/api/routes/storage.py`:
-  - `POST /api/v1/storage/records`: Create encrypted record
-    - Request: `StorageRecordCreate`
-    - Verifies checksum, stores blob in S3, inserts metadata row in PostgreSQL
-    - Response: `{id: str, created_at: int}`
-  - `GET /api/v1/storage/records`: List record metadata (no blobs)
-    - Query params: `table: str`, `page: int`, `limit: int`
-    - Response: `list[{id, table, checksum, created_at, updated_at}]`
-  - `GET /api/v1/storage/records/{id}`: Download encrypted blob
-    - Response: blob bytes + `X-Checksum` header
-  - `PUT /api/v1/storage/records/{id}`: Update encrypted blob
-    - Request: `StorageRecordUpdate`
-  - `DELETE /api/v1/storage/records/{id}`: Delete record + S3 blob
-  - All routes enforce tier cloud_storage_gb quota via `TierManager.check_quota(user_id)`
-
-#### 8d — Vectors endpoint (cloud vector store)
-- [x] `app/api/routes/vectors.py`:
-  - `POST /api/v1/storage/vectors/upsert`:
-    - Request: `VectorUpsertRequest`
-    - Verifies checksums, delegates to `VectorStore.upsert()`
-    - Response: `{upserted: int}`
-  - `POST /api/v1/storage/vectors/search`:
-    - Request: `VectorSearchRequest`
-    - Delegates to `VectorStore.search()`
-    - Response: `VectorSearchResponse`
-  - `DELETE /api/v1/storage/vectors`:
-    - Request: `{ids: list[str]}`
-
-#### 8e — Backup endpoint
-- [x] `app/api/routes/backup.py`:
-  - `PUT /api/v1/backup`: Accepts binary blob + metadata headers (`X-Backup-Version`, `X-Backup-Timestamp`, `X-Backup-Checksum`). Stores in S3 keyed by `{user_id}/{timestamp}`. Enforces tier limits:
-    - Free: 0 (no backup)
-    - Pro: 5 GB
-    - Power: 25 GB
-    - Team: unlimited
-  - `GET /api/v1/backup`: Returns latest blob for authenticated user. Supports `If-Modified-Since`.
-  - `GET /api/v1/backup/history`: Returns list of `BackupMetadata` (no blobs).
-  - `DELETE /api/v1/backup/{backup_id}`: Delete specific backup.
-
-#### 8f — Plugins endpoint
-- [x] `app/api/routes/plugins.py`:
-  - `GET /api/v1/plugins`:
-    - Query params: `category: str | None`, `q: str | None`, `page: int`, `sort: Literal['rating', 'installs', 'newest']`
-    - Response: `PluginListResponse`
-    - Available from Power tier and above
-  - `GET /api/v1/plugins/{id}`:
-    - Response: `PluginManifest` + ratings + install count
-  - `POST /api/v1/plugins/{id}/install`:
-    - Request: `PluginInstallRequest`
-    - Records installation for the user (billing tracking, analytics)
-    - If plugin is paid: triggers Stripe Connect charge + revenue split (70% developer, 30% platform)
-    - Response: `{ok: true, download_url: str}` — signed S3 URL for plugin package
-  - `DELETE /api/v1/plugins/{id}/install`:
-    - Unregisters installation
-
-#### 8g — Auth endpoint
-- [x] `app/api/routes/auth.py`:
-  - `POST /api/v1/auth/register`: `{email, password}` → bcrypt hash → insert user → return `AuthTokens`
-  - `POST /api/v1/auth/login`: Validate credentials → return `AuthTokens`
-  - `POST /api/v1/auth/refresh`: Rotate refresh token → return new `AuthTokens`
-  - `GET /api/v1/auth/me`: Return `UserProfile` for current JWT
-
-#### 8h — Billing endpoint
-- [x] `app/api/routes/billing.py`:
-  - `POST /api/v1/billing/checkout`: Creates Stripe checkout session → returns URL
-  - `POST /api/v1/billing/webhook`: Handles Stripe webhooks (subscription lifecycle)
-  - `GET /api/v1/billing/subscription`: Returns current subscription info
-  - `DELETE /api/v1/billing/subscription`: Cancels subscription
-
-- **Outcome:** Complete REST + WebSocket API covering orchestration, storage, vectors, backup, marketplace.
-
-### Step 9 — Middleware
-
-#### 9a — Auth middleware
-- [x] `app/api/middleware/auth.py`:
-  - FastAPI dependency: `get_current_user(token: str = Depends(oauth2_scheme)) -> UserProfile`
-  - Validates JWT signature, expiry, extracts `user_id` and `tier`
-  - Raises `401` on invalid/expired token
-  - Exempt routes: `/api/v1/auth/register`, `/api/v1/auth/login`, `/api/v1/billing/webhook`
-
-#### 9b — Rate limiter
-- [x] `app/api/middleware/rate_limit.py`:
-  - Uses `slowapi` with `Limiter(key_func=get_user_id_from_jwt)`
-  - Tier-based limits:
-    - Free: 20 req/min
-    - Pro: 60 req/min
-    - Power: 120 req/min
-    - Team: 200 req/seat/min
-  - Custom 429 response with `Retry-After` header
-
-#### 9c — Sanitizer
-- [x] `app/api/middleware/sanitizer.py`:
-  - Response middleware that scans response bodies
-  - Strips: system prompt fragments, agent internal reasoning, tool schemas, routing metadata
-  - Pattern-based detection + exact match against known prompt fingerprints
-  - Logs sanitization events for monitoring
-
-- **Outcome:** Secure, rate-limited API with prompt IP protection.
-
-### Step 10 — Plugin Marketplace ✅
-- [x] `app/marketplace/plugin_registry.py`:
-  - `PluginRegistry`:
-    - `async list_plugins(category, query, page, sort) -> PluginListResponse`
-    - `async get_plugin(plugin_id) -> PluginManifest | None`
-    - `async submit_plugin(manifest: PluginManifest, package_s3_key: str) -> str` — returns plugin_id, sets status = 'pending_review'
-    - `async approve_plugin(plugin_id) -> None` — admin only, sets status = 'approved'
-    - `async reject_plugin(plugin_id, reason: str) -> None`
-- [x] `app/marketplace/plugin_review.py`:
-  - `ReviewQueue`:
-    - `async get_pending() -> list[dict]`
-    - `async submit_review(plugin_id, reviewer_id, decision, notes) -> None`
-  - Security checklist enforced before approval: manifest schema valid, permissions are from allowed set, no binary blobs in manifest
-- [x] `app/marketplace/revenue_share.py`:
-  - `RevenueShare`:
-    - `async record_install(plugin_id, user_id, amount_cents) -> None`
-    - `async payout_developer(plugin_id, period) -> None` — Stripe Connect transfer: 70% to developer
-    - `async get_earnings(developer_id, period) -> dict`
-- **Outcome:** Plugin marketplace with catalog, review workflow, and revenue split.
-
-### Step 11 — Billing & Tier management ✅
-- [x] `app/billing/stripe_service.py`:
-  - `create_checkout_session(user_id, tier) -> str`
-  - `handle_webhook(payload, sig_header) -> None`: processes `checkout.session.completed`, `customer.subscription.updated`, `customer.subscription.deleted`, `invoice.payment_failed`
-  - `get_subscription(user_id) -> dict | None`
-  - `cancel_subscription(user_id) -> None`
-- [x] `app/billing/tier_manager.py`:
-  - `TierManager`:
-    - Feature matrix:
-      ```python
-      FEATURES = {
-          'free':  {
-              'agents': 3,
-              'batch_active': 2,
-              'cloud_storage_gb': 0,
-              'backup_gb': 0,
-              'providers': 1,
-              'batch_builder': False,
-              'plugin_marketplace': False,
-              'sso': False,
-          },
-          'pro':   {
-              'agents': -1,          # unlimited
-              'batch_active': 10,
-              'cloud_storage_gb': 5,
-              'backup_gb': 5,
-              'providers': -1,
-              'batch_builder': False,
-              'plugin_marketplace': False,
-              'sso': False,
-          },
-          'power': {
-              'agents': -1,
-              'batch_active': -1,    # unlimited
-              'cloud_storage_gb': 25,
-              'backup_gb': 25,
-              'providers': -1,
-              'batch_builder': True,
-              'plugin_marketplace': True,
-              'sso': False,
-          },
-          'team':  {
-              'agents': -1,
-              'batch_active': -1,
-              'cloud_storage_gb': -1,
-              'backup_gb': -1,
-              'providers': -1,
-              'batch_builder': True,
-              'plugin_marketplace': True,
-              'sso': True,
-          },
-      }
-      ```
-    - `get_tier(user_id) -> BillingTier`
-    - `check_feature(user_id, feature) -> bool`
-    - `get_rate_limit(tier) -> int`
-    - `check_quota(user_id) -> bool` — checks cloud_storage_gb current usage vs limit
-- [x] `app/billing/__init__.py`: exports `stripe_service` and `tier_manager` singletons
-- [x] `app/api/routes/billing.py`: refactored to delegate to `StripeService`
-- [x] `app/api/routes/storage.py` and `backup.py`: `_check_quota` now delegates to `tier_manager.enforce_quota` / `enforce_backup_quota`
-- **Outcome:** Stripe integration with tier-based feature gating matching Free/Pro(15€)/Power(29€)/Team(49€/seat).
-
-### Step 12 — Database (auth/billing/marketplace only)
-- [x] PostgreSQL schema via Alembic:
-  - `users`: `id UUID PK`, `email UNIQUE`, `password_hash`, `tier` (default 'free'), `stripe_customer_id`, `created_at`, `updated_at`
-  - `refresh_tokens`: `id UUID PK`, `user_id FK`, `token_hash`, `expires_at`, `created_at`
-  - `subscriptions`: `id UUID PK`, `user_id FK`, `stripe_subscription_id`, `tier`, `status`, `current_period_end`, `created_at`
-  - `backup_metadata`: `id UUID PK`, `user_id FK`, `s3_key`, `version`, `timestamp`, `checksum`, `size_bytes`, `created_at`
-  - `storage_records`: `id UUID PK`, `user_id FK`, `table_name VARCHAR`, `s3_key`, `checksum`, `size_bytes`, `created_at`, `updated_at` — metadata only, no plaintext
-  - `plugins`: `id UUID PK`, `name`, `description`, `version`, `author_id FK`, `category`, `status` (pending_review/approved/rejected), `price_cents`, `s3_package_key`, `install_count`, `avg_rating`, `created_at`
-  - `plugin_installations`: `id UUID PK`, `plugin_id FK`, `user_id FK`, `installed_at`
-  - `plugin_reviews`: `id UUID PK`, `plugin_id FK`, `reviewer_id FK`, `decision`, `notes`, `reviewed_at`
-  - `revenue_events`: `id UUID PK`, `plugin_id FK`, `user_id FK`, `amount_cents`, `developer_share_cents`, `stripe_transfer_id`, `created_at`
-- [x] Initial Alembic migration
-- [x] SQLAlchemy models in `app/models.py`
-- **Outcome:** Auth, billing, storage metadata, and marketplace persistence. Zero user data in plaintext.
-
-### Step 13 — Testing & deployment ✅
-- [x] `tests/conftest.py`: TestClient fixture, mock LLM fixture (`AsyncMock` returning canned responses), mock agent fixture, test DB (SQLite in-memory for speed), mock S3 (moto), mock Pinecone
-- [x] `tests/test_orchestrator.py`: classify_intent routing, single agent, pipeline, plan mode
-- [x] `tests/test_agents.py`: each agent with mocked tools
-- [x] `tests/test_auth.py`: register → login → access protected → refresh → expired token
-- [x] `tests/test_backup.py`: upload → download → history → delete, tier limit enforcement
-- [x] `tests/test_storage.py`: create record → list → download → update → delete, checksum rejection, quota enforcement
-- [x] `tests/test_plugins.py`: list plugins, install, uninstall, revenue event creation, tier gate (free user blocked)
-- [x] `Dockerfile` optimized for production (gunicorn + uvicorn workers)
-- [x] GitHub Actions CI: lint (ruff), test (pytest), build Docker image
-- **Outcome:** Fully tested, deployable backend.
-
----
-
-## API Contract Summary
-
-| Method | Endpoint | Auth | Request | Response |
-|--------|----------|------|---------|----------|
-| POST | `/api/v1/auth/register` | No | `{email, password}` | `AuthTokens` |
-| POST | `/api/v1/auth/login` | No | `{email, password}` | `AuthTokens` |
-| POST | `/api/v1/auth/refresh` | No | `{refresh_token}` | `AuthTokens` |
-| GET | `/api/v1/auth/me` | JWT | — | `UserProfile` |
-| POST | `/api/v1/chat` | JWT | `ChatRequest` | `ChatResponse \| ExecutionPlan` |
-| WS | `/api/v1/chat/stream` | JWT | `ChatRequest` (first frame) | Token stream + final JSON |
-| GET | `/api/v1/plans/playbook` | JWT | — | `ExecutionPlan[]` |
-| GET | `/api/v1/plans/playbook/:id` | JWT | — | `ExecutionPlan` |
-| POST | `/api/v1/storage/records` | JWT | `StorageRecordCreate` | `{id, created_at}` |
-| GET | `/api/v1/storage/records` | JWT | `?table&page&limit` | `RecordMeta[]` |
-| GET | `/api/v1/storage/records/:id` | JWT | — | Binary blob |
-| PUT | `/api/v1/storage/records/:id` | JWT | `StorageRecordUpdate` | `{ok: true}` |
-| DELETE | `/api/v1/storage/records/:id` | JWT | — | `{ok: true}` |
-| POST | `/api/v1/storage/vectors/upsert` | JWT | `VectorUpsertRequest` | `{upserted: int}` |
-| POST | `/api/v1/storage/vectors/search` | JWT | `VectorSearchRequest` | `VectorSearchResponse` |
-| DELETE | `/api/v1/storage/vectors` | JWT | `{ids: list[str]}` | `{ok: true}` |
-| PUT | `/api/v1/backup` | JWT | Binary blob + headers | `{ok: true}` |
-| GET | `/api/v1/backup` | JWT | — | Binary blob |
-| GET | `/api/v1/backup/history` | JWT | — | `BackupMetadata[]` |
-| DELETE | `/api/v1/backup/:id` | JWT | — | `{ok: true}` |
-| GET | `/api/v1/plugins` | JWT | `?category&q&page&sort` | `PluginListResponse` |
-| GET | `/api/v1/plugins/:id` | JWT | — | `PluginManifest` + stats |
-| POST | `/api/v1/plugins/:id/install` | JWT | `PluginInstallRequest` | `{ok, download_url}` |
-| DELETE | `/api/v1/plugins/:id/install` | JWT | — | `{ok: true}` |
-| POST | `/api/v1/billing/checkout` | JWT | `{tier}` | `{checkout_url}` |
-| POST | `/api/v1/billing/webhook` | Stripe sig | Stripe event | `{ok: true}` |
-| GET | `/api/v1/billing/subscription` | JWT | — | Subscription info |
-| DELETE | `/api/v1/billing/subscription` | JWT | — | `{ok: true}` |
-| GET | `/api/v1/health` | No | — | `{status, version}` |
-
----
-
-## Stack
-
-| Layer | Technology |
-|-------|-----------|
-| Framework | FastAPI + Uvicorn |
-| LLM | LangChain + langchain-openai |
-| Auth | PyJWT + bcrypt + OAuth2 |
-| Billing | stripe-python + Stripe Connect |
-| Blob storage | boto3 (S3) |
-| Vector store | Pinecone or Qdrant (configurable) |
-| Database | PostgreSQL + SQLAlchemy + Alembic |
-| Rate limiting | slowapi |
-| Testing | pytest + pytest-asyncio + httpx + moto (S3 mock) |
-| Deployment | Docker → fly.io / Railway / AWS ECS |
-
----
-
-## Development Rules
-
-1. **NEVER persist user data in plaintext.** The DB stores only auth, billing, storage metadata, and marketplace data. User context arrives in requests and is discarded. Cloud blobs are E2E encrypted client-side — backend only stores opaque bytes.
-2. **NEVER expose prompts.** System prompts are composed server-side from fragments. Responses are sanitized before sending. In plan mode, `prompt_template` fields are reference IDs only.
-3. **NEVER decrypt user blobs.** `app/storage/encryption.py` only verifies checksums. No decryption key ever reaches the backend.
-4. **Stateless request handling.** No server-side session state. All context comes from the client + JWT.
-5. **Type hints everywhere.** All functions have full type annotations.
-6. **Test every agent.** Each chat agent has unit tests with mocked LLM responses.
-7. **Structured logging.** JSON logs with request ID correlation.
-8. **Tier gates are enforced server-side.** Never trust client-reported tier. Always fetch from DB via `TierManager.get_tier(user_id)`.
-9. **One step at a time.** Implement one numbered step per session. When the step is fully done, mark all its checkboxes as `[x]` in this file and commit with message `step N complete: <outcome line>`.

README.md

@@ -83,7 +83,7 @@ Adiuva Cloud API is the FastAPI backend that powers the **Adiuva Electron deskto
 ## Key Features
 
 1. **LLM-powered orchestration** — GPT-4o-mini classifies user intent and routes to the appropriate domain agent.
-2. **4 specialized AI agents** — Tasks (8 tools), Projects (6 tools), Checkpoints (4 tools), Notes (5 tools), all powered by GPT-4o via LangChain.
+2. **4 specialized AI agents** — Tasks (8 tools), Projects (6 tools), Timelines (4 tools), Notes (5 tools), all powered by GPT-4o via LangChain.
 3. **Execution plans & playbooks** — Server-side prompt template registry; clients receive only opaque template IDs, never raw prompts.
 4. **E2E encrypted cloud storage** — The backend never decrypts user data; SHA-256 checksum verification uses constant-time comparison to prevent timing attacks.
 5. **Cloud vector store** — Pinecone or Qdrant with user-isolated namespaces and encrypted blob payloads.
@@ -449,7 +449,7 @@ The agent system uses a registry pattern with LangChain tool-calling agents powe
 |---|---|---|---|
 | **TaskAgent** | `task_agent` | 8 | Full task and comment CRUD. Status: `todo` / `in_progress` / `done`. Priority: `high` / `medium` / `low`. Tools: `list_tasks`, `create_task`, `update_task`, `delete_task`, `list_tasks_due_today`, `list_task_comments`, `add_task_comment`, `delete_task_comment` |
 | **ProjectAgent** | `project_agent` | 6 | Project lifecycle management. Status: `active` / `archived`. Prefers archiving over deletion. Tools: `list_projects`, `list_all_projects`, `get_project`, `create_project`, `update_project`, `delete_project` |
-| **CheckpointAgent** | `checkpoint_agent` | 4 | Project milestones. Requires `project_id` for creation. Supports AI-suggestion and approval workflows. Tools: `list_checkpoints`, `create_checkpoint`, `update_checkpoint`, `delete_checkpoint` |
+| **TimelineAgent** | `timeline_agent` | 4 | Project milestones. Requires `project_id` for creation. Supports AI-suggestion and approval workflows. Tools: `list_timelines`, `create_timeline`, `update_timeline`, `delete_timeline` |
 | **NoteAgent** | `note_agent` | 5 | Markdown note management. Optionally linked to projects. Tools: `list_notes`, `get_note`, `create_note`, `update_note`, `delete_note` |
 
 All agents use the model configured by `LLM_MODEL` (default: GPT-4o) with `temperature=0` via LiteLLM. Tools return JSON action descriptors that the Electron client interprets and applies locally.
@@ -504,7 +504,7 @@ Source: `app/core/orchestrator.py`, `app/core/execution_plan.py`
 
 ### Built-in Templates (6)
 
-`tpl_task_agent_default`, `tpl_checkpoint_agent_default`, `tpl_project_agent_default`, `tpl_note_agent_default`, `tpl_task_extract_from_project`, `tpl_note_weekly_summary`
+`tpl_task_agent_default`, `tpl_timeline_agent_default`, `tpl_project_agent_default`, `tpl_note_agent_default`, `tpl_task_extract_from_project`, `tpl_note_weekly_summary`
 
 ### Built-in Playbooks (2)
 
@@ -643,7 +643,7 @@ Source: `app/marketplace/`
   - Plugin ID must match `^[a-z0-9-]+$`
   - Permissions must be from the allowed set only
   - No binary blobs in the manifest
-- **Allowed permissions:** `read:tasks`, `write:tasks`, `read:projects`, `write:projects`, `read:notes`, `write:notes`, `read:checkpoints`, `write:checkpoints`, `read:calendar`, `write:calendar`
+- **Allowed permissions:** `read:tasks`, `write:tasks`, `read:projects`, `write:projects`, `read:notes`, `write:notes`, `read:timelines`, `write:timelines`, `read:calendar`, `write:calendar`
 - `get_pending(db)` — Lists plugins awaiting review.
 - `submit_review(db, plugin_id, reviewer_id, decision, notes)` — Records the review decision.
 
@@ -734,12 +734,12 @@ adiuva-api/
 │   ├── agents/                  # LLM-powered domain agents
 │   │   ├── task_agent.py        # Task & comment CRUD (8 tools)
 │   │   ├── project_agent.py     # Project lifecycle (6 tools)
-│   │   ├── checkpoint_agent.py  # Milestones (4 tools)
+│   │   ├── timeline_agent.py  # Milestones (4 tools)
 │   │   └── note_agent.py        # Markdown notes (5 tools)
 │   │
 │   ├── core/                    # Orchestration engine
 │   │   ├── agent_registry.py    # BaseAgent, ChatAgent, AgentRegistry
-│   │   ├── llm.py               # LiteLLM factory (get_llm, get_router_llm)
+│   │   ├── llm.py               # LiteLLM factory (get_llm)
 │   │   ├── orchestrator.py      # Intent classification & routing
 │   │   └── execution_plan.py    # Plan builder, templates, cache
 │   │

alembic/versions/001_initial_schema.py

@@ -21,18 +21,25 @@ depends_on: Union[str, Sequence[str], None] = None
 
 
 def upgrade() -> None:
-    # ── Enum types ────────────────────────────────────────────────────────
+    # ── Enum types — idempotent creation via exception handling ───────────
-    billing_tier = postgresql.ENUM(
+    op.execute("""
-        "free", "pro", "power", "team", name="billing_tier", create_type=False
+        DO $$ BEGIN
-    )
+            CREATE TYPE billing_tier AS ENUM ('free', 'pro', 'power', 'team');
-    plugin_status = postgresql.ENUM(
+        EXCEPTION WHEN duplicate_object THEN NULL;
-        "pending_review", "approved", "rejected", name="plugin_status", create_type=False
+        END $$;
-    )
+    """)
-    review_decision = postgresql.ENUM(
+    op.execute("""
-        "approved", "rejected", name="review_decision", create_type=False
+        DO $$ BEGIN
-    )
+            CREATE TYPE plugin_status AS ENUM ('pending_review', 'approved', 'rejected');
-    for enum in (billing_tier, plugin_status, review_decision):
+        EXCEPTION WHEN duplicate_object THEN NULL;
-        enum.create(op.get_bind(), checkfirst=True)
+        END $$;
+    """)
+    op.execute("""
+        DO $$ BEGIN
+            CREATE TYPE review_decision AS ENUM ('approved', 'rejected');
+        EXCEPTION WHEN duplicate_object THEN NULL;
+        END $$;
+    """)
 
     # ── users ─────────────────────────────────────────────────────────────
     op.create_table(
@@ -40,7 +47,7 @@ def upgrade() -> None:
         sa.Column("id", postgresql.UUID(as_uuid=False), nullable=False),
         sa.Column("email", sa.String(255), nullable=False),
         sa.Column("password_hash", sa.String(255), nullable=False),
-        sa.Column("tier", sa.Enum("free", "pro", "power", "team", name="billing_tier", create_type=False), nullable=False, server_default="free"),
+        sa.Column("tier", postgresql.ENUM("free", "pro", "power", "team", name="billing_tier", create_type=False), nullable=False, server_default="free"),
         sa.Column("stripe_customer_id", sa.String(255), nullable=True),
         sa.Column("created_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
         sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
@@ -70,7 +77,7 @@ def upgrade() -> None:
         sa.Column("id", postgresql.UUID(as_uuid=False), nullable=False),
         sa.Column("user_id", postgresql.UUID(as_uuid=False), nullable=False),
         sa.Column("stripe_subscription_id", sa.String(255), nullable=True),
-        sa.Column("tier", sa.Enum("free", "pro", "power", "team", name="billing_tier", create_type=False), nullable=False, server_default="free"),
+        sa.Column("tier", postgresql.ENUM("free", "pro", "power", "team", name="billing_tier", create_type=False), nullable=False, server_default="free"),
         sa.Column("status", sa.String(50), nullable=False, server_default="free"),
         sa.Column("current_period_end", sa.DateTime(timezone=True), nullable=True),
         sa.Column("created_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
@@ -125,7 +132,7 @@ def upgrade() -> None:
         sa.Column("category", sa.String(100), nullable=False, server_default=""),
         sa.Column("price_cents", sa.Integer, nullable=False, server_default="0"),
         sa.Column("permissions", sa.Text, nullable=False, server_default="[]"),
-        sa.Column("status", sa.Enum("pending_review", "approved", "rejected", name="plugin_status", create_type=False), nullable=False, server_default="pending_review"),
+        sa.Column("status", postgresql.ENUM("pending_review", "approved", "rejected", name="plugin_status", create_type=False), nullable=False, server_default="pending_review"),
         sa.Column("s3_package_key", sa.String(500), nullable=True),
         sa.Column("install_count", sa.Integer, nullable=False, server_default="0"),
         sa.Column("avg_rating", sa.Float, nullable=False, server_default="0.0"),
@@ -157,7 +164,7 @@ def upgrade() -> None:
         sa.Column("id", postgresql.UUID(as_uuid=False), nullable=False),
         sa.Column("plugin_id", sa.String(255), nullable=False),
         sa.Column("reviewer_id", postgresql.UUID(as_uuid=False), nullable=True),
-        sa.Column("decision", sa.Enum("approved", "rejected", name="review_decision", create_type=False), nullable=False),
+        sa.Column("decision", postgresql.ENUM("approved", "rejected", name="review_decision", create_type=False), nullable=False),
         sa.Column("notes", sa.Text, nullable=True),
         sa.Column("reviewed_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
         sa.Column("created_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),

alembic/versions/002_seed_plugins.py

@@ -37,12 +37,12 @@ _SEED_PLUGINS = [
     {
         "id": "plugin-slack-notify",
         "name": "Slack Notifier",
-        "description": "Post task and checkpoint updates to Slack channels.",
+        "description": "Post task and timeline updates to Slack channels.",
         "version": "1.2.0",
         "author_name": "Adiuva",
         "category": "communication",
         "price_cents": 499,
-        "permissions": json.dumps(["read:tasks", "read:checkpoints"]),
+        "permissions": json.dumps(["read:tasks", "read:timelines"]),
         "status": "approved",
         "s3_package_key": "plugins/plugin-slack-notify/1.2.0/package.zip",
         "install_count": 0,

alembic/versions/003_agent_tables.py

@@ -0,0 +1,127 @@
+"""Add agent config and run log tables: local_agent_configs, cloud_agent_configs, agent_run_logs.
+
+Revision ID: 003
+Revises: 002
+Create Date: 2026-03-05
+"""
+
+from __future__ import annotations
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+from sqlalchemy.dialects import postgresql
+
+revision: str = "003"
+down_revision: Union[str, None] = "002"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    # ── Enum types — idempotent creation ──────────────────────────────────
+    op.execute("""
+        DO $$ BEGIN
+            CREATE TYPE agent_type AS ENUM ('local', 'cloud');
+        EXCEPTION WHEN duplicate_object THEN NULL;
+        END $$;
+    """)
+    op.execute("""
+        DO $$ BEGIN
+            CREATE TYPE agent_run_status AS ENUM ('running', 'success', 'error', 'partial');
+        EXCEPTION WHEN duplicate_object THEN NULL;
+        END $$;
+    """)
+    op.execute("""
+        DO $$ BEGIN
+            CREATE TYPE cloud_provider AS ENUM ('gmail', 'teams', 'outlook');
+        EXCEPTION WHEN duplicate_object THEN NULL;
+        END $$;
+    """)
+
+    # ── local_agent_configs ───────────────────────────────────────────────
+    op.create_table(
+        "local_agent_configs",
+        sa.Column("id", postgresql.UUID(as_uuid=False), nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=False), nullable=False),
+        sa.Column("device_id", sa.String(255), nullable=False),
+        sa.Column("name", sa.String(255), nullable=False),
+        sa.Column("directory_paths", sa.JSON, nullable=False, server_default="[]"),
+        sa.Column("data_types", sa.JSON, nullable=False, server_default="[]"),
+        sa.Column("prompt_template", sa.Text, nullable=False, server_default=""),
+        sa.Column("file_extensions", sa.JSON, nullable=False, server_default="[]"),
+        sa.Column("schedule_cron", sa.String(100), nullable=False, server_default="0 */6 * * *"),
+        sa.Column("enabled", sa.Boolean, nullable=False, server_default=sa.true()),
+        sa.Column("last_run_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.PrimaryKeyConstraint("id"),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+    )
+    op.create_index("ix_local_agent_configs_user_id", "local_agent_configs", ["user_id"])
+
+    # ── cloud_agent_configs ───────────────────────────────────────────────
+    op.create_table(
+        "cloud_agent_configs",
+        sa.Column("id", postgresql.UUID(as_uuid=False), nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=False), nullable=False),
+        sa.Column(
+            "provider",
+            postgresql.ENUM("gmail", "teams", "outlook", name="cloud_provider", create_type=False),
+            nullable=False,
+        ),
+        sa.Column("name", sa.String(255), nullable=False),
+        sa.Column("data_types", sa.JSON, nullable=False, server_default="[]"),
+        sa.Column("prompt_template", sa.Text, nullable=False, server_default=""),
+        sa.Column("oauth_token_encrypted", sa.Text, nullable=True),
+        sa.Column("filter_config", sa.JSON, nullable=True),
+        sa.Column("schedule_cron", sa.String(100), nullable=False, server_default="0 */6 * * *"),
+        sa.Column("enabled", sa.Boolean, nullable=False, server_default=sa.true()),
+        sa.Column("last_run_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.PrimaryKeyConstraint("id"),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+    )
+    op.create_index("ix_cloud_agent_configs_user_id", "cloud_agent_configs", ["user_id"])
+
+    # ── agent_run_logs ─────────────────────────────────────────────────────
+    op.create_table(
+        "agent_run_logs",
+        sa.Column("id", postgresql.UUID(as_uuid=False), nullable=False),
+        # Plain string — not a FK because it references either local_agent_configs or
+        # cloud_agent_configs depending on agent_type.
+        sa.Column("agent_id", sa.String(255), nullable=False),
+        sa.Column(
+            "agent_type",
+            postgresql.ENUM("local", "cloud", name="agent_type", create_type=False),
+            nullable=False,
+        ),
+        sa.Column("user_id", postgresql.UUID(as_uuid=False), nullable=False),
+        sa.Column(
+            "status",
+            postgresql.ENUM("running", "success", "error", "partial", name="agent_run_status", create_type=False),
+            nullable=False,
+            server_default="running",
+        ),
+        sa.Column("items_processed", sa.Integer, nullable=False, server_default="0"),
+        sa.Column("items_created", sa.Integer, nullable=False, server_default="0"),
+        sa.Column("errors", sa.JSON, nullable=True),
+        sa.Column("started_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.Column("completed_at", sa.DateTime(timezone=True), nullable=True),
+        sa.PrimaryKeyConstraint("id"),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+    )
+    op.create_index("ix_agent_run_logs_user_id", "agent_run_logs", ["user_id"])
+    op.create_index("ix_agent_run_logs_agent_id", "agent_run_logs", ["agent_id"])
+
+
+def downgrade() -> None:
+    op.drop_table("agent_run_logs")
+    op.drop_table("cloud_agent_configs")
+    op.drop_table("local_agent_configs")
+
+    op.execute("DROP TYPE IF EXISTS cloud_provider;")
+    op.execute("DROP TYPE IF EXISTS agent_run_status;")
+    op.execute("DROP TYPE IF EXISTS agent_type;")

alembic/versions/004_add_memory_tables.py

@@ -0,0 +1,144 @@
+"""Add memory tables and user encryption_key column.
+
+Memory tables:
+  memory_core        — per-user key/value preferences (encrypted)
+  memory_associative — semantic memory with pgvector embedding (encrypted)
+  memory_episodic    — session summaries (encrypted)
+  memory_proactive   — behavioral patterns (encrypted)
+
+Also adds encryption_key column to users table.
+
+Revision ID: 004
+Revises: 003
+Create Date: 2026-03-08
+"""
+
+from __future__ import annotations
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+from sqlalchemy.dialects import postgresql
+
+revision: str = "004"
+down_revision: Union[str, None] = "003"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    # ── Enable pgvector extension (idempotent) ────────────────────────────────
+    op.execute("CREATE EXTENSION IF NOT EXISTS vector;")
+
+    # ── Add encryption_key to users ───────────────────────────────────────────
+    op.add_column(
+        "users",
+        sa.Column("encryption_key", sa.String(64), nullable=True),
+    )
+
+    # ── memory_core ───────────────────────────────────────────────────────────
+    op.create_table(
+        "memory_core",
+        sa.Column("id", postgresql.UUID(as_uuid=False), primary_key=True),
+        sa.Column(
+            "user_id",
+            postgresql.UUID(as_uuid=False),
+            sa.ForeignKey("users.id", ondelete="CASCADE"),
+            nullable=False,
+        ),
+        sa.Column("key", sa.String(255), nullable=False),
+        sa.Column("value_encrypted", sa.Text, nullable=False),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.func.now(),
+        ),
+    )
+    op.create_index("ix_memory_core_user_id", "memory_core", ["user_id"])
+
+    # ── memory_associative ────────────────────────────────────────────────────
+    # The embedding column uses pgvector's vector(1536) type.
+    op.create_table(
+        "memory_associative",
+        sa.Column("id", postgresql.UUID(as_uuid=False), primary_key=True),
+        sa.Column(
+            "user_id",
+            postgresql.UUID(as_uuid=False),
+            sa.ForeignKey("users.id", ondelete="CASCADE"),
+            nullable=False,
+        ),
+        sa.Column("content_encrypted", sa.Text, nullable=False),
+        sa.Column("entity_type", sa.String(100), nullable=True),
+        sa.Column("entity_id", sa.String(255), nullable=True),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.func.now(),
+        ),
+    )
+    # Add the pgvector column separately (not supported by generic sa types)
+    op.execute(
+        "ALTER TABLE memory_associative ADD COLUMN embedding vector(1536);"
+    )
+    op.create_index("ix_memory_associative_user_id", "memory_associative", ["user_id"])
+    # IVFFlat index for approximate nearest-neighbour search
+    op.execute(
+        "CREATE INDEX ix_memory_associative_embedding "
+        "ON memory_associative USING ivfflat (embedding vector_cosine_ops) WITH (lists = 100);"
+    )
+
+    # ── memory_episodic ───────────────────────────────────────────────────────
+    op.create_table(
+        "memory_episodic",
+        sa.Column("id", postgresql.UUID(as_uuid=False), primary_key=True),
+        sa.Column(
+            "user_id",
+            postgresql.UUID(as_uuid=False),
+            sa.ForeignKey("users.id", ondelete="CASCADE"),
+            nullable=False,
+        ),
+        sa.Column("summary_encrypted", sa.Text, nullable=False),
+        sa.Column("session_id", sa.String(255), nullable=False),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.func.now(),
+        ),
+    )
+    op.create_index("ix_memory_episodic_user_id", "memory_episodic", ["user_id"])
+    op.create_index("ix_memory_episodic_session_id", "memory_episodic", ["session_id"])
+
+    # ── memory_proactive ──────────────────────────────────────────────────────
+    op.create_table(
+        "memory_proactive",
+        sa.Column("id", postgresql.UUID(as_uuid=False), primary_key=True),
+        sa.Column(
+            "user_id",
+            postgresql.UUID(as_uuid=False),
+            sa.ForeignKey("users.id", ondelete="CASCADE"),
+            nullable=False,
+        ),
+        sa.Column("pattern_encrypted", sa.Text, nullable=False),
+        sa.Column("confidence", sa.Float, nullable=False, server_default="0.5"),
+        sa.Column("source", sa.String(50), nullable=False, server_default="inferred"),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.func.now(),
+        ),
+    )
+    op.create_index("ix_memory_proactive_user_id", "memory_proactive", ["user_id"])
+
+
+def downgrade() -> None:
+    op.drop_table("memory_proactive")
+    op.drop_table("memory_episodic")
+    op.drop_index("ix_memory_associative_embedding", "memory_associative")
+    op.drop_table("memory_associative")
+    op.drop_table("memory_core")
+    op.drop_column("users", "encryption_key")

alembic/versions/818478c251dc_add_name_and_surname_to_users_table.py

@@ -0,0 +1,30 @@
+"""add name and surname to users table
+
+Revision ID: 818478c251dc
+Revises: 004
+Create Date: 2026-03-10 15:10:42.811947
+
+"""
+from __future__ import annotations
+
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision: str = '818478c251dc'
+down_revision: Union[str, None] = '004'
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    op.add_column('users', sa.Column('name', sa.String(length=100), nullable=True))
+    op.add_column('users', sa.Column('surname', sa.String(length=100), nullable=True))
+
+
+def downgrade() -> None:
+    op.drop_column('users', 'surname')
+    op.drop_column('users', 'name')

alembic/versions/9a1f2d0b6c7e_deprecate_backend_agent_config_tables.py

@@ -0,0 +1,92 @@
+"""Deprecate backend agent config tables.
+
+The Electron client is now the source of truth for agent configuration
+(directory, extract targets, batch interval, custom prompt). Backend keeps
+billing checks and trigger/run logs only.
+
+Revision ID: 9a1f2d0b6c7e
+Revises: 818478c251dc
+Create Date: 2026-03-16
+"""
+
+from __future__ import annotations
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+from sqlalchemy.dialects import postgresql
+
+revision: str = "9a1f2d0b6c7e"
+down_revision: Union[str, None] = "818478c251dc"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+    existing = set(inspector.get_table_names())
+
+    if "cloud_agent_configs" in existing:
+        op.drop_index("ix_cloud_agent_configs_user_id", table_name="cloud_agent_configs")
+        op.drop_table("cloud_agent_configs")
+
+    if "local_agent_configs" in existing:
+        op.drop_index("ix_local_agent_configs_user_id", table_name="local_agent_configs")
+        op.drop_table("local_agent_configs")
+
+
+def downgrade() -> None:
+    op.create_table(
+        "local_agent_configs",
+        sa.Column("id", postgresql.UUID(as_uuid=False), nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=False), nullable=False),
+        sa.Column("device_id", sa.String(255), nullable=False),
+        sa.Column("name", sa.String(255), nullable=False),
+        sa.Column("directory_paths", sa.JSON, nullable=False, server_default="[]"),
+        sa.Column("data_types", sa.JSON, nullable=False, server_default="[]"),
+        sa.Column("prompt_template", sa.Text, nullable=False, server_default=""),
+        sa.Column("file_extensions", sa.JSON, nullable=False, server_default="[]"),
+        sa.Column("schedule_cron", sa.String(100), nullable=False, server_default="0 */6 * * *"),
+        sa.Column("enabled", sa.Boolean, nullable=False, server_default=sa.true()),
+        sa.Column("last_run_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.PrimaryKeyConstraint("id"),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+    )
+    op.create_index("ix_local_agent_configs_user_id", "local_agent_configs", ["user_id"])
+
+    op.execute(
+        """
+        DO $$ BEGIN
+            CREATE TYPE cloud_provider AS ENUM ('gmail', 'teams', 'outlook');
+        EXCEPTION WHEN duplicate_object THEN NULL;
+        END $$;
+        """
+    )
+
+    op.create_table(
+        "cloud_agent_configs",
+        sa.Column("id", postgresql.UUID(as_uuid=False), nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=False), nullable=False),
+        sa.Column(
+            "provider",
+            postgresql.ENUM("gmail", "teams", "outlook", name="cloud_provider", create_type=False),
+            nullable=False,
+        ),
+        sa.Column("name", sa.String(255), nullable=False),
+        sa.Column("data_types", sa.JSON, nullable=False, server_default="[]"),
+        sa.Column("prompt_template", sa.Text, nullable=False, server_default=""),
+        sa.Column("oauth_token_encrypted", sa.Text, nullable=True),
+        sa.Column("filter_config", sa.JSON, nullable=True),
+        sa.Column("schedule_cron", sa.String(100), nullable=False, server_default="0 */6 * * *"),
+        sa.Column("enabled", sa.Boolean, nullable=False, server_default=sa.true()),
+        sa.Column("last_run_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.PrimaryKeyConstraint("id"),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+    )
+    op.create_index("ix_cloud_agent_configs_user_id", "cloud_agent_configs", ["user_id"])

app/agents/__init__.py

@@ -1,5 +1,5 @@
-"""Import all agent modules to trigger @registry.register decorators."""
+"""Expose tool modules used by deep orchestrator-worker graphs."""
 
-from app.agents import checkpoint_agent, note_agent, project_agent, task_agent
+from app.agents import filesystem_agent, timeline_agent, note_agent, project_agent, task_agent
 
-__all__ = ["checkpoint_agent", "note_agent", "project_agent", "task_agent"]
+__all__ = ["filesystem_agent", "timeline_agent", "note_agent", "project_agent", "task_agent"]

app/agents/checkpoint_agent.py

@@ -1,121 +0,0 @@
-"""Checkpoint agent — project milestone management (list, create, update, delete)."""
-
-from __future__ import annotations
-
-import json
-from typing import Any
-
-from langchain_core.messages import HumanMessage, SystemMessage
-from langchain_core.tools import tool
-
-from app.core.agent_registry import ChatAgent, registry
-from app.core.llm import get_llm
-
-_SYSTEM_PROMPT = (
-    "You are a project checkpoint assistant. Checkpoints are milestone dates that\n"
-    "track progress on a project — they are not calendar events.\n\n"
-    "Rules:\n"
-    "  - project_id is REQUIRED for every create; confirm with the user if unknown\n"
-    "  - date is a Unix timestamp in milliseconds; convert human-readable dates\n"
-    "  - is_ai_suggested: 1 when proactively proposing a checkpoint, 0 otherwise\n"
-    "  - is_approved: 0 until the user explicitly confirms; then 1\n"
-    "  - For update_checkpoint, use -1 for integer fields you do not want to change\n"
-    "  - Listing without a project_id returns all checkpoints across projects\n"
-    "  - Always echo the title and formatted date in your confirmation."
-)
-
-
-@tool
-async def list_checkpoints(project_id: str = "") -> str:
-    """List checkpoints. Provide project_id to scope to a specific project."""
-    return json.dumps({
-        "action": "list",
-        "table": "checkpoints",
-        "filters": {"projectId": project_id or None},
-    })
-
-
-@tool
-async def create_checkpoint(
-    project_id: str,
-    title: str,
-    date: int,
-    is_ai_suggested: int = 0,
-    is_approved: int = 0,
-) -> str:
-    """Create a project checkpoint (milestone).
-    project_id: REQUIRED UUID of the parent project
-    title: descriptive name for the milestone
-    date: Unix timestamp in milliseconds
-    is_ai_suggested: 1 if proactively suggested, 0 if user-requested
-    is_approved: 0 until the user confirms
-    """
-    return json.dumps({
-        "action": "create_record",
-        "table": "checkpoints",
-        "data": {
-            "projectId": project_id,
-            "title": title,
-            "date": date,
-            "isAiSuggested": is_ai_suggested,
-            "isApproved": is_approved,
-        },
-    })
-
-
-@tool
-async def update_checkpoint(
-    checkpoint_id: str,
-    title: str = "",
-    date: int = -1,
-    is_approved: int = -1,
-) -> str:
-    """Update a checkpoint. Only pass fields that should change.
-    checkpoint_id: UUID of the checkpoint (required)
-    date: -1 means unchanged; any other value sets the new date (ms timestamp)
-    is_approved: -1 means unchanged; 0 or 1 sets the approval state
-    """
-    updates: dict[str, Any] = {}
-    if title:
-        updates["title"] = title
-    if date != -1:
-        updates["date"] = date
-    if is_approved != -1:
-        updates["isApproved"] = is_approved
-    return json.dumps({
-        "action": "update_record",
-        "table": "checkpoints",
-        "data": {"id": checkpoint_id, "updates": updates},
-    })
-
-
-@tool
-async def delete_checkpoint(checkpoint_id: str) -> str:
-    """Delete a checkpoint permanently by its UUID."""
-    return json.dumps({
-        "action": "delete_record",
-        "table": "checkpoints",
-        "data": {"id": checkpoint_id},
-    })
-
-
-@registry.register
-class CheckpointAgent(ChatAgent):
-    def get_name(self) -> str:
-        return "checkpoint_agent"
-
-    def get_description(self) -> str:
-        return "Manages project checkpoints (milestones): list, create, update, delete"
-
-    def get_tools(self) -> list[Any]:
-        return [list_checkpoints, create_checkpoint, update_checkpoint, delete_checkpoint]
-
-    async def handle(self, query: str, context: dict[str, Any]) -> str:
-        llm = get_llm()
-        messages = [
-            SystemMessage(content=_SYSTEM_PROMPT),
-            HumanMessage(
-                content=f"User query: {query}\nContext: {json.dumps(context)[:1000]}"
-            ),
-        ]
-        return await self._tool_loop(llm, messages, self.get_tools())

app/agents/filesystem_agent.py

@@ -0,0 +1,85 @@
+"""Filesystem agent — tools for reading local directories and files on Electron.
+
+These tools delegate to the Electron client via ``execute_on_client()`` using
+the same WS tool-call round-trip pattern as CRUD tools.  The Electron app
+handles actual disk I/O and responds with ``tool_result`` frames.
+"""
+
+from __future__ import annotations
+
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.core.ws_context import execute_on_client
+
+
+@tool
+async def list_directory(path: str) -> str:
+    """List files and folders in a local directory on the user's device.
+
+    Returns a formatted listing of entries with name, type (file/directory),
+    and full path.
+    """
+    result = await execute_on_client(
+        action="list_directory",
+        data={"path": path},
+    )
+    entries: list[dict[str, Any]] = result.get("entries", [])
+    if not entries:
+        return f"Directory '{path}' is empty or does not exist."
+    lines: list[str] = []
+    for entry in entries:
+        entry_type = entry.get("type", "unknown")
+        entry_name = entry.get("name", "")
+        entry_path = entry.get("path", "")
+        lines.append(f"- [{entry_type}] {entry_name}  ({entry_path})")
+    return f"Directory listing for '{path}' ({len(entries)} entries):\n" + "\n".join(lines)
+
+
+@tool
+async def read_file_content(path: str) -> str:
+    """Read the text content of a local file on the user's device.
+
+    Returns the file content as a string.  Large files may be truncated
+    by the Electron client.
+    """
+    result = await execute_on_client(
+        action="read_file_content",
+        data={"path": path},
+    )
+    content: str = result.get("content", "")
+    if not content:
+        return f"File '{path}' is empty or could not be read."
+    return content
+
+
+@tool
+async def get_file_metadata(path: str) -> str:
+    """Get metadata for a local file: size, creation date, modification date, extension.
+
+    Returns a formatted summary of the file's metadata.
+    """
+    result = await execute_on_client(
+        action="get_file_metadata",
+        data={"path": path},
+    )
+    size = result.get("size", "unknown")
+    created = result.get("createdAt", "unknown")
+    modified = result.get("modifiedAt", "unknown")
+    extension = result.get("extension", "unknown")
+    name = result.get("name", path)
+    return (
+        f"File: {name}\n"
+        f"  Extension: {extension}\n"
+        f"  Size: {size} bytes\n"
+        f"  Created: {created}\n"
+        f"  Modified: {modified}"
+    )
+
+
+FILESYSTEM_TOOLS: list[Any] = [
+    list_directory,
+    read_file_content,
+    get_file_metadata,
+]

app/agents/note_agent.py

@@ -2,16 +2,23 @@
 
 from __future__ import annotations
 
-import json
+import re
 from typing import Any
 
-from langchain_core.messages import HumanMessage, SystemMessage
 from langchain_core.tools import tool
 
-from app.core.agent_registry import ChatAgent, registry
+from app.core.llm import embed
-from app.core.llm import get_llm
+from app.core.ws_context import execute_on_client
 
-_SYSTEM_PROMPT = (
+_UUID_RE = re.compile(
+    r"^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$"
+)
+
+
+def _is_uuid(value: str) -> bool:
+    return bool(_UUID_RE.match(value))
+
+NOTE_SYSTEM_PROMPT = (
     "You are a note-taking assistant. You help users create, retrieve, update,\n"
     "and delete Markdown notes in their workspace.\n\n"
     "Rules:\n"
@@ -21,6 +28,7 @@ _SYSTEM_PROMPT = (
     "    before appending or replacing sections\n"
     "  - list_notes without project_id returns all notes; scope with project_id\n"
     "    when the user is working within a specific project\n"
+    "  - project_id must be a UUID; if you only know a project name, do not pass it as project_id\n"
     "  - Do not fabricate note content — reflect what the user provides or what\n"
     "    is already in the note (retrieved via get_note)."
 )
@@ -29,21 +37,27 @@ _SYSTEM_PROMPT = (
 @tool
 async def list_notes(project_id: str = "") -> str:
     """List notes, optionally scoped to a project by project_id."""
-    return json.dumps({
+    normalized_project_id = project_id if (project_id and _is_uuid(project_id)) else ""
-        "action": "list",
+    result = await execute_on_client(
-        "table": "notes",
+        action="select",
-        "filters": {"projectId": project_id or None},
+        table="notes",
-    })
+        filters={"projectId": normalized_project_id or None},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No notes found."
+    lines = [f"- {r['title']} (id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} note(s):\n" + "\n".join(lines)
 
 
 @tool
 async def get_note(note_id: str) -> str:
     """Fetch a single note by its UUID to read its full Markdown content."""
-    return json.dumps({
+    result = await execute_on_client(action="get", table="notes", data={"id": note_id})
-        "action": "get",
+    row = result.get("row")
-        "table": "notes",
+    if not row:
-        "data": {"id": note_id},
+        return f"Note {note_id} not found."
-    })
+    return f"Note '{row['title']}' (id: {row['id']}):\n\n{row['content']}"
 
 
 @tool
@@ -57,15 +71,24 @@ async def create_note(
     content: Markdown body text (required)
     project_id: optional UUID linking this note to a project
     """
-    return json.dumps({
+    result = await execute_on_client(
-        "action": "create_record",
+        action="insert",
-        "table": "notes",
+        table="notes",
-        "data": {
+        data={
             "title": title,
             "content": content,
             "projectId": project_id or None,
         },
-    })
+    )
+    row = result["row"]
+    # Index the note content in the vector store.
+    vector = await embed(content)
+    await execute_on_client(
+        action="vector_upsert",
+        data={"id": row["id"], "projectId": row.get("projectId"), "content": content},
+        vector=vector,
+    )
+    return f"Note created: '{row['title']}' (id: {row['id']})."
 
 
 @tool
@@ -83,40 +106,34 @@ async def update_note(
         updates["title"] = title
     if content:
         updates["content"] = content
-    return json.dumps({
+    result = await execute_on_client(
-        "action": "update_record",
+        action="update",
-        "table": "notes",
+        table="notes",
-        "data": {"id": note_id, "updates": updates},
+        data={"id": note_id, "updates": updates},
-    })
+    )
+    row = result["row"]
+    # Re-index if content changed.
+    if content:
+        vector = await embed(content)
+        await execute_on_client(
+            action="vector_upsert",
+            data={"id": note_id, "projectId": row.get("projectId"), "content": content},
+            vector=vector,
+        )
+    return f"Note updated: '{row['title']}' (id: {row['id']})."
 
 
 @tool
 async def delete_note(note_id: str) -> str:
     """Delete a note permanently by its UUID."""
-    return json.dumps({
+    await execute_on_client(action="delete", table="notes", data={"id": note_id})
-        "action": "delete_record",
+    return f"Note {note_id} deleted."
-        "table": "notes",
-        "data": {"id": note_id},
-    })
 
 
-@registry.register
+NOTE_TOOLS: list[Any] = [
-class NoteAgent(ChatAgent):
+    list_notes,
-    def get_name(self) -> str:
+    get_note,
-        return "note_agent"
+    create_note,
-
+    update_note,
-    def get_description(self) -> str:
+    delete_note,
-        return "Manages notes: list, get, create, update, delete"
-
-    def get_tools(self) -> list[Any]:
-        return [list_notes, get_note, create_note, update_note, delete_note]
-
-    async def handle(self, query: str, context: dict[str, Any]) -> str:
-        llm = get_llm()
-        messages = [
-            SystemMessage(content=_SYSTEM_PROMPT),
-            HumanMessage(
-                content=f"User query: {query}\nContext: {json.dumps(context)[:1000]}"
-            ),
 ]
-        return await self._tool_loop(llm, messages, self.get_tools())

app/agents/project_agent.py

@@ -2,16 +2,13 @@
 
 from __future__ import annotations
 
-import json
 from typing import Any
 
-from langchain_core.messages import HumanMessage, SystemMessage
 from langchain_core.tools import tool
 
-from app.core.agent_registry import ChatAgent, registry
+from app.core.ws_context import execute_on_client
-from app.core.llm import get_llm
 
-_SYSTEM_PROMPT = (
+PROJECT_SYSTEM_PROMPT = (
     "You are a project management assistant. You help users create, find,\n"
     "update, and archive projects in their workspace.\n\n"
     "Rules:\n"
@@ -36,14 +33,19 @@ async def list_projects(
     """List projects, optionally filtered by client_id.
     include_archived: 1 to include archived projects, 0 for active only (default).
     """
-    return json.dumps({
+    result = await execute_on_client(
-        "action": "list",
+        action="select",
-        "table": "projects",
+        table="projects",
-        "filters": {
+        filters={
             "clientId": client_id or None,
             "includeArchived": bool(include_archived),
         },
-    })
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No projects found."
+    lines = [f"- {r['name']} (status: {r['status']}, id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} project(s):\n" + "\n".join(lines)
 
 
 @tool
@@ -51,20 +53,25 @@ async def list_all_projects() -> str:
     """List every project regardless of client or status.
     Use only when the user wants a complete cross-client overview.
     """
-    return json.dumps({
+    result = await execute_on_client(action="select", table="projects")
-        "action": "list_all",
+    rows = result.get("rows", [])
-        "table": "projects",
+    if not rows:
-    })
+        return "No projects found."
+    lines = [f"- {r['name']} (status: {r['status']}, id: {r['id']})" for r in rows]
+    return f"All projects ({len(rows)}):\n" + "\n".join(lines)
 
 
 @tool
 async def get_project(project_id: str) -> str:
     """Fetch a single project by its UUID."""
-    return json.dumps({
+    result = await execute_on_client(action="get", table="projects", data={"id": project_id})
-        "action": "get",
+    row = result.get("row")
-        "table": "projects",
+    if not row:
-        "data": {"id": project_id},
+        return f"Project {project_id} not found."
-    })
+    return (
+        f"Project: '{row['name']}' (id: {row['id']}, status: {row['status']}, "
+        f"clientId: {row.get('clientId', 'none')})"
+    )
 
 
 @tool
@@ -76,14 +83,13 @@ async def create_project(
     name: human-readable project name (required)
     client_id: optional UUID of the owning client
     """
-    return json.dumps({
+    result = await execute_on_client(
-        "action": "create_record",
+        action="insert",
-        "table": "projects",
+        table="projects",
-        "data": {
+        data={"name": name, "clientId": client_id or None},
-            "name": name,
+    )
-            "clientId": client_id or None,
+    row = result["row"]
-        },
+    return f"Project created: '{row['name']}' (id: {row['id']})"
-    })
 
 
 @tool
@@ -108,11 +114,13 @@ async def update_project(
         updates["status"] = status
     if ai_summary:
         updates["aiSummary"] = ai_summary
-    return json.dumps({
+    result = await execute_on_client(
-        "action": "update_record",
+        action="update",
-        "table": "projects",
+        table="projects",
-        "data": {"id": project_id, "updates": updates},
+        data={"id": project_id, "updates": updates},
-    })
+    )
+    row = result["row"]
+    return f"Project updated: '{row['name']}' (id: {row['id']}, status: {row['status']})"
 
 
 @tool
@@ -121,23 +129,11 @@ async def delete_project(project_id: str) -> str:
     IMPORTANT: prefer update_project(status='archived') unless the user
     has explicitly confirmed they want permanent deletion.
     """
-    return json.dumps({
+    await execute_on_client(action="delete", table="projects", data={"id": project_id})
-        "action": "delete_record",
+    return f"Project {project_id} permanently deleted."
-        "table": "projects",
-        "data": {"id": project_id},
-    })
 
 
-@registry.register
+PROJECT_TOOLS: list[Any] = [
-class ProjectAgent(ChatAgent):
-    def get_name(self) -> str:
-        return "project_agent"
-
-    def get_description(self) -> str:
-        return "Manages projects: list, get, create, update, archive, delete"
-
-    def get_tools(self) -> list[Any]:
-        return [
     list_projects,
     list_all_projects,
     get_project,
@@ -145,13 +141,3 @@ class ProjectAgent(ChatAgent):
     update_project,
     delete_project,
 ]
-
-    async def handle(self, query: str, context: dict[str, Any]) -> str:
-        llm = get_llm()
-        messages = [
-            SystemMessage(content=_SYSTEM_PROMPT),
-            HumanMessage(
-                content=f"User query: {query}\nContext: {json.dumps(context)[:1000]}"
-            ),
-        ]
-        return await self._tool_loop(llm, messages, self.get_tools())

app/agents/task_agent.py

@@ -2,16 +2,23 @@
 
 from __future__ import annotations
 
-import json
+from datetime import datetime, timezone
+import re
 from typing import Any
 
-from langchain_core.messages import HumanMessage, SystemMessage
 from langchain_core.tools import tool
 
-from app.core.agent_registry import ChatAgent, registry
+from app.core.ws_context import execute_on_client
-from app.core.llm import get_llm
 
-_SYSTEM_PROMPT = (
+_UUID_RE = re.compile(
+    r"^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$"
+)
+
+
+def _is_uuid(value: str) -> bool:
+    return bool(_UUID_RE.match(value))
+
+TASK_SYSTEM_PROMPT = (
     "You are a task management assistant for a project workspace.\n"
     "You create, update, list, and track tasks and their comments.\n\n"
     "Rules:\n"
@@ -22,7 +29,7 @@ _SYSTEM_PROMPT = (
     "  - project_id is optional; link to a project when the user mentions one\n"
     "  - is_ai_suggested: 1 only when proactively proposing a task the user\n"
     "    did not explicitly request; 0 otherwise\n"
-    "  - is_approved defaults to 0; set to 1 only when the user confirms\n"
+    "  - is_ai_suggested: 1 only when proactively proposing a task the user did not explicitly request; 0 otherwise\n"
     "  - Use list_tasks_due_today for 'what's due today' queries\n"
     "  - For update_task, use -1 for integer fields you do not want to change\n"
     "  - Always confirm the action in plain, user-friendly language."
@@ -41,16 +48,25 @@ async def list_tasks(
 ) -> str:
     """List tasks, optionally filtered by project_id, status (todo|in_progress|done),
     a search string, or an order_by field name (dueDate|priority|createdAt)."""
-    return json.dumps({
+    normalized_project_id = project_id if (project_id and _is_uuid(project_id)) else ""
-        "action": "list",
+    result = await execute_on_client(
-        "table": "tasks",
+        action="select",
-        "filters": {
+        table="tasks",
-            "projectId": project_id or None,
+        filters={
+            "projectId": normalized_project_id or None,
             "status": status or None,
             "search": search or None,
             "orderBy": order_by or None,
         },
-    })
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No tasks found matching the given filters."
+    lines = [
+        f"- {r['title']} (status: {r['status']}, priority: {r['priority']}, id: {r['id']})"
+        for r in rows
+    ]
+    return f"Found {len(rows)} task(s):\n" + "\n".join(lines)
 
 
 @tool
@@ -63,7 +79,6 @@ async def create_task(
     due_date: int = 0,
     project_id: str = "",
     is_ai_suggested: int = 0,
-    is_approved: int = 0,
 ) -> str:
     """Create a new task.
     title: task title (required)
@@ -74,12 +89,11 @@ async def create_task(
     due_date: Unix timestamp in milliseconds; 0 means no due date
     project_id: optional UUID of the parent project
     is_ai_suggested: 1 if proactively suggested, 0 if user-requested
-    is_approved: 0 until the user confirms; 1 when confirmed
     """
-    return json.dumps({
+    result = await execute_on_client(
-        "action": "create_record",
+        action="insert",
-        "table": "tasks",
+        table="tasks",
-        "data": {
+        data={
             "title": title,
             "description": description or None,
             "status": status,
@@ -88,9 +102,13 @@ async def create_task(
             "dueDate": due_date or None,
             "projectId": project_id or None,
             "isAiSuggested": is_ai_suggested,
-            "isApproved": is_approved,
         },
-    })
+    )
+    row = result["row"]
+    return (
+        f"Task created: '{row['title']}' "
+        f"(id: {row['id']}, status: {row['status']}, priority: {row['priority']})"
+    )
 
 
 @tool
@@ -103,12 +121,10 @@ async def update_task(
     assignees: str = "",
     due_date: int = -1,
     project_id: str = "",
-    is_approved: int = -1,
 ) -> str:
     """Update fields on an existing task. Only pass fields you want to change.
     task_id: the task's UUID (required)
     due_date: -1 means unchanged; 0 clears the due date; any positive value sets it
-    is_approved: -1 means unchanged; 0 or 1 sets the value
     """
     updates: dict[str, Any] = {}
     if title:
@@ -125,32 +141,41 @@ async def update_task(
         updates["dueDate"] = due_date or None
     if project_id:
         updates["projectId"] = project_id
-    if is_approved != -1:
+    result = await execute_on_client(
-        updates["isApproved"] = is_approved
+        action="update",
-    return json.dumps({
+        table="tasks",
-        "action": "update_record",
+        data={"id": task_id, "updates": updates},
-        "table": "tasks",
+    )
-        "data": {"id": task_id, "updates": updates},
+    row = result["row"]
-    })
+    return f"Task updated: '{row['title']}' (id: {row['id']}, status: {row['status']})"
 
 
 @tool
 async def delete_task(task_id: str) -> str:
     """Delete a task permanently by its UUID."""
-    return json.dumps({
+    await execute_on_client(action="delete", table="tasks", data={"id": task_id})
-        "action": "delete_record",
+    return f"Task {task_id} deleted."
-        "table": "tasks",
-        "data": {"id": task_id},
-    })
 
 
 @tool
 async def list_tasks_due_today() -> str:
     """List all tasks whose due date falls on today's date."""
-    return json.dumps({
+    now = datetime.now(tz=timezone.utc)
-        "action": "list_due_today",
+    start_ms = int(datetime(now.year, now.month, now.day, tzinfo=timezone.utc).timestamp() * 1000)
-        "table": "tasks",
+    end_ms = start_ms + 86_400_000 - 1  # last ms of today
-    })
+    result = await execute_on_client(
+        action="select",
+        table="tasks",
+        filters={"dueDateFrom": start_ms, "dueDateTo": end_ms},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No tasks are due today."
+    lines = [
+        f"- {r['title']} (priority: {r['priority']}, status: {r['status']}, id: {r['id']})"
+        for r in rows
+    ]
+    return f"Tasks due today ({len(rows)}):\n" + "\n".join(lines)
 
 
 # ── Task comment tools ────────────────────────────────────────────────
@@ -159,11 +184,16 @@ async def list_tasks_due_today() -> str:
 @tool
 async def list_task_comments(task_id: str) -> str:
     """List all comments on a task by its UUID."""
-    return json.dumps({
+    result = await execute_on_client(
-        "action": "list",
+        action="select",
-        "table": "taskComments",
+        table="taskComments",
-        "filters": {"taskId": task_id},
+        filters={"taskId": task_id},
-    })
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return f"No comments found for task {task_id}."
+    lines = [f"- [{r['author']}]: {r['content']} (id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} comment(s):\n" + "\n".join(lines)
 
 
 @tool
@@ -173,40 +203,30 @@ async def add_task_comment(task_id: str, author: str, content: str) -> str:
     author: name or ID of the comment author
     content: comment text
     """
-    return json.dumps({
+    result = await execute_on_client(
-        "action": "create_record",
+        action="insert",
-        "table": "taskComments",
+        table="taskComments",
-        "data": {
+        data={"taskId": task_id, "author": author, "content": content},
-            "taskId": task_id,
+    )
-            "author": author,
+    row = result.get("row", {})
-            "content": content,
+    row_author = row.get("author", author)
-        },
+    # Electron payloads can vary (taskId vs task_id). Fall back to input task_id.
-    })
+    row_task_id = row.get("taskId") or row.get("task_id") or task_id
+    row_comment_id = row.get("id", "unknown")
+    return f"Comment added by {row_author} on task {row_task_id} (comment id: {row_comment_id})."
 
 
 @tool
 async def delete_task_comment(comment_id: str) -> str:
     """Delete a task comment by its UUID."""
-    return json.dumps({
+    await execute_on_client(action="delete", table="taskComments", data={"id": comment_id})
-        "action": "delete_record",
+    return f"Comment {comment_id} deleted."
-        "table": "taskComments",
-        "data": {"id": comment_id},
-    })
 
 
 # ── Agent ─────────────────────────────────────────────────────────────
 
 
-@registry.register
+TASK_TOOLS: list[Any] = [
-class TaskAgent(ChatAgent):
-    def get_name(self) -> str:
-        return "task_agent"
-
-    def get_description(self) -> str:
-        return "Manages tasks and comments: list, create, update, delete, due-today, comments"
-
-    def get_tools(self) -> list[Any]:
-        return [
     list_tasks,
     create_task,
     update_task,
@@ -216,13 +236,3 @@ class TaskAgent(ChatAgent):
     add_task_comment,
     delete_task_comment,
 ]
-
-    async def handle(self, query: str, context: dict[str, Any]) -> str:
-        llm = get_llm()
-        messages = [
-            SystemMessage(content=_SYSTEM_PROMPT),
-            HumanMessage(
-                content=f"User query: {query}\nContext: {json.dumps(context)[:1000]}"
-            ),
-        ]
-        return await self._tool_loop(llm, messages, self.get_tools())

app/agents/timeline_agent.py

@@ -0,0 +1,114 @@
+"""Timeline agent — project milestone management (list, create, update, delete)."""
+
+from __future__ import annotations
+
+import re
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.core.ws_context import execute_on_client
+
+_UUID_RE = re.compile(
+    r"^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$"
+)
+
+
+def _is_uuid(value: str) -> bool:
+    return bool(_UUID_RE.match(value))
+
+TIMELINE_SYSTEM_PROMPT = (
+    "You are a project timeline assistant. Timelines are milestone dates that\n"
+    "track progress on a project — they are not calendar events.\n\n"
+    "Rules:\n"
+    "  - project_id is REQUIRED for every create; confirm with the user if unknown\n"
+    "  - For listing, project_id must be a UUID; never pass plain names as project_id\n"
+    "  - date is a Unix timestamp in milliseconds; convert human-readable dates\n"
+    "  - is_ai_suggested: 1 when proactively proposing a timeline, 0 otherwise\n"
+    "  - is_ai_suggested: 1 when proactively proposing a timeline, 0 otherwise\n"
+    "  - For update_timeline, use -1 for integer fields you do not want to change\n"
+    "  - Listing without a project_id returns all timelines across projects\n"
+    "  - Always echo the title and formatted date in your confirmation."
+)
+
+
+@tool
+async def list_timelines(project_id: str = "") -> str:
+    """List timelines. Provide project_id to scope to a specific project."""
+    normalized_project_id = project_id if (project_id and _is_uuid(project_id)) else ""
+    result = await execute_on_client(
+        action="select",
+        table="timelines",
+        filters={"projectId": normalized_project_id or None},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No timelines found."
+    lines = [f"- {r['title']} (date: {r['date']}, id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} timeline(s):\n" + "\n".join(lines)
+
+
+@tool
+async def create_timeline(
+    project_id: str,
+    title: str,
+    date: int,
+    is_ai_suggested: int = 0,
+) -> str:
+    """Create a project timeline (milestone).
+    project_id: REQUIRED UUID of the parent project
+    title: descriptive name for the milestone
+    date: Unix timestamp in milliseconds
+    is_ai_suggested: 1 if proactively suggested, 0 if user-requested
+    """
+    result = await execute_on_client(
+        action="insert",
+        table="timelines",
+        data={
+            "projectId": project_id,
+            "title": title,
+            "date": date,
+            "isAiSuggested": is_ai_suggested,
+        },
+    )
+    row = result["row"]
+    return f"Timeline created: '{row['title']}' (id: {row['id']}, date: {row['date']})"
+
+
+@tool
+async def update_timeline(
+    timeline_id: str,
+    title: str = "",
+    date: int = -1,
+) -> str:
+    """Update a timeline. Only pass fields that should change.
+    timeline_id: UUID of the timeline (required)
+    date: -1 means unchanged; any other value sets the new date (ms timestamp)
+    """
+    updates: dict[str, Any] = {}
+    if title:
+        updates["title"] = title
+    if date != -1:
+        updates["date"] = date
+    result = await execute_on_client(
+        action="update",
+        table="timelines",
+        data={"id": timeline_id, "updates": updates},
+    )
+    row = result["row"]
+    return f"Timeline updated: '{row['title']}' (id: {row['id']})"
+
+
+@tool
+async def delete_timeline(timeline_id: str) -> str:
+    """Delete a timeline permanently by its UUID."""
+    await execute_on_client(action="delete", table="timelines", data={"id": timeline_id})
+    return f"Timeline {timeline_id} deleted."
+
+
+TIMELINE_TOOLS: list[Any] = [
+    list_timelines,
+    create_timeline,
+    update_timeline,
+    delete_timeline,
+]

app/api/middleware/auth.py

@@ -55,11 +55,26 @@ async def get_current_user(
         raise credentials_exc
 
     # Live tier lookup — subscription row is the authoritative source.
-    from app.models import Subscription  # noqa: PLC0415
+    # In dev, fall back to 'power' (unlimited) so quota limits don't
+    # block local development when no Stripe subscription exists.
+    from app.models import Subscription, User  # noqa: PLC0415
 
     result = await db.execute(
         select(Subscription.tier).where(Subscription.user_id == user_id)
     )
-    tier: str = result.scalar_one_or_none() or "free"
+    default_tier = "power" if settings.ENV == "dev" else "free"
+    tier: str = result.scalar_one_or_none() or default_tier
 
-    return UserProfile(id=user_id, email=email, tier=tier)  # type: ignore[arg-type]
+    # Fetch name/surname from user row.
+    user_result = await db.execute(
+        select(User.name, User.surname).where(User.id == user_id)
+    )
+    user_row = user_result.one_or_none()
+
+    return UserProfile(
+        id=user_id,
+        email=email,
+        name=user_row.name if user_row else None,
+        surname=user_row.surname if user_row else None,
+        tier=tier,
+    )  # type: ignore[arg-type]

app/api/routes/agent_setup.py

@@ -0,0 +1,406 @@
+"""Chatbot Journey — WS-based guided conversation to build an agent prompt_template.
+
+The journey is driven entirely through WebSocket frames (no REST endpoints).
+The device WS handler dispatches ``journey_start`` and ``journey_message``
+frames to the functions exported here.
+
+Journey flow:
+  1. FE sends ``journey_start`` frame with basic agent config (directory,
+     data_types, schedule).
+  2. Server creates an in-memory session, sets up a WS executor so the
+     setup LLM can use file-system tools, does a first directory scrape,
+     and sends back a ``journey_reply`` with the first question.
+  3. FE sends ``journey_message`` frames for each user reply.
+  4. Server appends the user message, calls the LLM (which may read files
+     via tools), and sends back a ``journey_reply``.
+  5. After 3-5 turns the LLM wraps up by emitting a ``prompt_template``
+     block delimited by ``PROMPT_TEMPLATE_START`` / ``PROMPT_TEMPLATE_END``.
+  6. Server parses the block, sends ``journey_reply`` with ``done=True``
+     and the template.  FE stores it locally.
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+import time
+import uuid
+from dataclasses import dataclass, field
+from typing import Any
+
+from langchain_core.messages import AIMessage, HumanMessage, SystemMessage, ToolMessage
+
+from app.agents.filesystem_agent import FILESYSTEM_TOOLS
+from app.core.llm import get_llm
+
+logger = logging.getLogger(__name__)
+
+# ── Session TTL ───────────────────────────────────────────────────────────
+
+_SESSION_TTL_SECONDS: int = 1800  # 30 minutes
+
+# Sentinel strings used to delimit the LLM-produced prompt_template.
+_TEMPLATE_START = "PROMPT_TEMPLATE_START"
+_TEMPLATE_END = "PROMPT_TEMPLATE_END"
+
+# Minimum turns before we consider nudging the LLM to wrap up.
+_MIN_TURNS_BEFORE_NUDGE: int = 3
+# Hard cap to avoid infinite loops (safety net, not the primary stopping criterion).
+_MAX_TURNS: int = 15
+# Max tool-calling steps per LLM invocation.
+_MAX_TOOL_STEPS: int = 6
+
+# ── In-memory session store ───────────────────────────────────────────────
+
+
+@dataclass
+class JourneySession:
+    session_id: str
+    user_id: str
+    agent_type: str  # "local" | "cloud"
+    directory: str
+    data_types: list[str]
+    history: list[dict[str, Any]] = field(default_factory=list)
+    system_prompt: str = ""
+    created_at: float = field(default_factory=time.monotonic)
+
+    def is_expired(self) -> bool:
+        return (time.monotonic() - self.created_at) > _SESSION_TTL_SECONDS
+
+
+# session_id → session
+_sessions: dict[str, JourneySession] = {}
+
+
+def get_journey_session(session_id: str, user_id: str) -> JourneySession | None:
+    """Retrieve session; return None on missing, expired, or wrong owner."""
+    s = _sessions.get(session_id)
+    if s is None or s.is_expired():
+        _sessions.pop(session_id, None)
+        return None
+    if s.user_id != user_id:
+        return None
+    return s
+
+
+# ── System prompt builder ─────────────────────────────────────────────────
+
+_SYSTEM_PROMPT_TEMPLATE = """\
+You are a friendly assistant helping a freelancer configure a data-extraction agent.
+Your job is to understand exactly what data the user wants to extract from their
+local directory and produce a detailed prompt_template that a separate AI will use
+as its instruction set.
+
+The extraction agent already has this base behaviour built in:
+  - Reads each file using file-system tools.
+  - Creates records (tasks, notes, timelines, projects) via CRUD tools.
+  - Sets isAiSuggested=1 on every new record.
+  - Only extracts data explicitly present in the files — it never invents information.
+The user's custom prompt is appended AFTER this base behaviour, so focus on
+what to look for and how to map it — not on the general extraction mechanics.
+
+You have access to file-system tools to explore the user's directory:
+- list_directory: to see folder structure
+- read_file_content: to peek at file contents
+- get_file_metadata: to check file info
+
+The user's configured directory is: {directory}
+Target data types: {data_types}
+
+IMPORTANT — project assignment is handled automatically by the main agent runner
+before the custom prompt is ever used.  You MUST NOT ask the user about projects,
+projectId, or how to link records to projects.  Never include projectId logic or
+project creation instructions in the generated prompt_template.
+
+Start by exploring the directory to understand its structure.  Then ask concise,
+focused questions one at a time.  Cover these topics (not necessarily in this order):
+  1. The type and format of the source content (confirmed by your exploration).
+  2. How fields should be mapped (e.g. filename → task title).
+  3. Priority or status rules (e.g. "urgent" keyword → high priority).
+  4. Any special handling, date extraction, or exclusions.
+
+Once you reach 90% confidence, output the final prompt_template between these exact
+markers on their own lines:
+
+{template_start}
+<the complete extraction prompt here>
+{template_end}
+
+The prompt_template must be a self-contained instruction for an AI that reads files
+and must perform CRUD operations using tools to create records.  It should specify:
+  - What entity types to create (tasks, notes, timelines) — never projects.
+  - How to map file content to record fields (camelCase: title, status, priority,
+    dueDate, content, etc.) — never include projectId.
+  - That isAiSuggested must be set to 1 on every new record.
+  - Concrete examples of mappings based on what you discovered in the directory.
+
+{existing_section}\
+Keep asking clarifying questions until you are at least 90% confident you have
+enough information to generate an accurate prompt_template.  Once you reach that
+confidence level, stop asking and produce the final template immediately.
+Begin by exploring the directory, then ask your first question.\
+"""
+
+
+def _build_system_prompt(
+    directory: str,
+    data_types: list[str],
+    existing_template: str | None = None,
+) -> str:
+    existing_section = (
+        f"\nThe user already has the following prompt_template — refine it based on their answers:\n"
+        f"---\n{existing_template}\n---\n"
+        if existing_template
+        else ""
+    )
+    return _SYSTEM_PROMPT_TEMPLATE.format(
+        directory=directory,
+        data_types=", ".join(data_types),
+        template_start=_TEMPLATE_START,
+        template_end=_TEMPLATE_END,
+        existing_section=existing_section,
+    )
+
+
+# ── Template extraction ───────────────────────────────────────────────────
+
+
+def _extract_template(text: str) -> str | None:
+    """Return the text between PROMPT_TEMPLATE_START and PROMPT_TEMPLATE_END, or None."""
+    if _TEMPLATE_START not in text or _TEMPLATE_END not in text:
+        return None
+    start_idx = text.index(_TEMPLATE_START) + len(_TEMPLATE_START)
+    end_idx = text.index(_TEMPLATE_END)
+    return text[start_idx:end_idx].strip() or None
+
+
+# ── LLM call with tool support ───────────────────────────────────────────
+
+
+def _as_text(content: Any) -> str:
+    if content is None:
+        return ""
+    if isinstance(content, str):
+        return content
+    if isinstance(content, list):
+        parts: list[str] = []
+        for item in content:
+            if isinstance(item, str):
+                parts.append(item)
+            elif isinstance(item, dict):
+                text = item.get("text")
+                if isinstance(text, str):
+                    parts.append(text)
+        return "".join(parts)
+    return str(content)
+
+
+async def _call_llm_with_tools(
+    system_prompt: str,
+    history: list[dict[str, Any]],
+    tools: list[Any],
+) -> str:
+    """Build LangChain messages from history and invoke the LLM with tools.
+
+    Handles tool-calling loops: if the LLM calls tools, execute them and
+    continue until a final text response is produced.
+    """
+    messages: list[Any] = [SystemMessage(content=system_prompt)]
+    for turn in history:
+        if turn["role"] == "user":
+            messages.append(HumanMessage(content=turn["content"]))
+        else:
+            messages.append(AIMessage(content=turn["content"]))
+
+    llm = get_llm(model=None, temperature=0.4)
+    llm_with_tools = llm.bind_tools(tools)
+    tool_map = {tool_def.name: tool_def for tool_def in tools}
+
+    for _ in range(_MAX_TOOL_STEPS):
+        response: AIMessage = await llm_with_tools.ainvoke(messages)
+        messages.append(response)
+
+        if not response.tool_calls:
+            return _as_text(response.content)
+
+        for call in response.tool_calls:
+            call_name = str(call.get("name", ""))
+            call_args = call.get("args", {})
+            logger.info(
+                "agent_setup: journey tool_call name=%s args=%s",
+                call_name,
+                json.dumps(call_args, ensure_ascii=True)[:500],
+            )
+
+            tool_fn = tool_map.get(call_name)
+            if tool_fn is None:
+                tool_output = f"Unknown tool: {call_name}"
+            else:
+                tool_output = await tool_fn.ainvoke(call_args)
+
+            logger.info(
+                "agent_setup: journey tool_result name=%s output=%s",
+                call_name,
+                str(tool_output)[:800],
+            )
+            messages.append(ToolMessage(content=str(tool_output), tool_call_id=call["id"]))
+
+    # Fallback: exceeded max steps.
+    final = await llm.ainvoke(messages)
+    return _as_text(final.content)
+
+
+# ── Journey handlers (called from device_ws.py) ──────────────────────────
+
+
+async def handle_journey_start(
+    user_id: str,
+    frame: dict[str, Any],
+) -> dict[str, Any]:
+    """Handle a ``journey_start`` WS frame.
+
+    Creates a session, runs the setup LLM with directory exploration,
+    and returns the ``journey_reply`` payload.
+    """
+    agent_type = frame.get("agent_type", "local")
+    directory = frame.get("directory", "")
+    data_types = frame.get("data_types", [])
+    existing_template = frame.get("existing_template")
+
+    # Use the session_id provided by the FE so the reply matches the
+    # listener key; fall back to a generated one if absent.
+    session_id = frame.get("session_id") or str(uuid.uuid4())
+    system_prompt = _build_system_prompt(directory, data_types, existing_template)
+
+    session = JourneySession(
+        session_id=session_id,
+        user_id=user_id,
+        agent_type=agent_type,
+        directory=directory,
+        data_types=data_types,
+        system_prompt=system_prompt,
+    )
+
+    # The LLM will explore the directory using FILESYSTEM_TOOLS via the
+    # ws_context executor (already set by the WS handler before calling us).
+    # Seed with an initial user message — some providers (e.g. GitHub Copilot)
+    # require at least one user/input message to be present.
+    seed_history: list[dict[str, Any]] = [
+        {"role": "user", "content": "Hi, I'm ready to set up my agent. Please explore my directory and ask me your first question."},
+    ]
+    ai_reply = await _call_llm_with_tools(
+        system_prompt=system_prompt,
+        history=seed_history,
+        tools=list(FILESYSTEM_TOOLS),
+    )
+
+    session.history.extend(seed_history)
+    session.history.append({"role": "assistant", "content": ai_reply})
+    _sessions[session_id] = session
+
+    logger.info(
+        "agent_setup: journey session %s started for user %s (directory=%s)",
+        session_id,
+        user_id,
+        directory,
+    )
+
+    # Check if the LLM produced the template on the first turn (unlikely but possible).
+    prompt_template = _extract_template(ai_reply)
+    done = prompt_template is not None
+
+    display_message = ai_reply
+    if done:
+        display_message = (
+            ai_reply[: ai_reply.index(_TEMPLATE_START)].strip()
+            or "Here is your agent configuration. You can save it or continue refining."
+        )
+        _sessions.pop(session_id, None)
+
+    return {
+        "type": "journey_reply",
+        "session_id": session_id,
+        "message": display_message,
+        "done": done,
+        "prompt_template": prompt_template,
+    }
+
+
+async def handle_journey_message(
+    user_id: str,
+    frame: dict[str, Any],
+) -> dict[str, Any]:
+    """Handle a ``journey_message`` WS frame.
+
+    Appends the user message, calls the LLM, and returns the
+    ``journey_reply`` payload.
+    """
+    session_id = frame.get("session_id", "")
+    message = frame.get("message", "")
+
+    session = get_journey_session(session_id, user_id)
+    if session is None:
+        return {
+            "type": "journey_reply",
+            "session_id": session_id,
+            "message": "Journey session not found or expired. Please start a new setup.",
+            "done": True,
+            "prompt_template": None,
+        }
+
+    # Append user turn.
+    session.history.append({"role": "user", "content": message})
+
+    # Call the LLM with tools.
+    ai_reply = await _call_llm_with_tools(
+        system_prompt=session.system_prompt,
+        history=session.history,
+        tools=list(FILESYSTEM_TOOLS),
+    )
+
+    session.history.append({"role": "assistant", "content": ai_reply})
+
+    # Check if the LLM produced the final template.
+    prompt_template = _extract_template(ai_reply)
+    done = prompt_template is not None
+
+    # If the LLM didn't produce a template, nudge it once it has asked enough
+    # questions (>= _MIN_TURNS_BEFORE_NUDGE) or hits the hard safety cap.
+    if not done:
+        turns = sum(1 for t in session.history if t["role"] == "user")
+        if turns >= _MAX_TURNS:
+            nudge_content = (
+                "[System: You have enough information. Please generate the final "
+                f"prompt_template now, wrapped in {_TEMPLATE_START} / {_TEMPLATE_END} markers.]"
+            )
+            session.history.append({"role": "user", "content": nudge_content})
+
+            nudge_reply = await _call_llm_with_tools(
+                system_prompt=session.system_prompt,
+                history=session.history,
+                tools=list(FILESYSTEM_TOOLS),
+            )
+            session.history.append({"role": "assistant", "content": nudge_reply})
+
+            prompt_template = _extract_template(nudge_reply)
+            if prompt_template is not None:
+                done = True
+                ai_reply = nudge_reply
+
+    display_message = ai_reply
+    if done:
+        display_message = (
+            ai_reply[: ai_reply.index(_TEMPLATE_START)].strip()
+            if _TEMPLATE_START in ai_reply
+            else "Here is your agent configuration. You can save it or continue refining."
+        )
+        _sessions.pop(session_id, None)
+        logger.info("agent_setup: journey session %s completed for user %s", session_id, user_id)
+
+    return {
+        "type": "journey_reply",
+        "session_id": session_id,
+        "message": display_message,
+        "done": done,
+        "prompt_template": prompt_template,
+    }

app/api/routes/agents.py

@@ -0,0 +1,222 @@
+"""Agent routes.
+
+Backend responsibilities are intentionally minimal:
+    GET  /agents/catalog         — static catalog for UI display
+    POST /agents/can-create      — billing eligibility check
+    POST /agents/trigger         — trigger a local agent run
+
+Agent configuration is owned by the Electron app and is not persisted
+in backend agent-config tables.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import uuid
+from datetime import datetime, timedelta, timezone
+
+from fastapi import APIRouter, Depends, HTTPException, status
+from sqlalchemy import func, select
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from app.api.deps import get_current_user
+from app.billing.tier_manager import FEATURES
+from app.core.agent_runner import is_agent_running, run_local_agent
+from app.core.device_manager import device_manager
+from app.db import get_session
+from app.models import AgentRunLog, LocalAgentConfig
+from app.schemas import (
+    AgentCatalogItem,
+    AgentCreationCheckRequest,
+    AgentCreationCheckResponse,
+    AgentRunLogResponse,
+    AgentTriggerRequest,
+    UserProfile,
+)
+
+router = APIRouter(prefix="/agents", tags=["agents"])
+
+
+# ── Datetime helpers ──────────────────────────────────────────────────
+
+def _dt_ms(dt: datetime) -> int:
+    return int(dt.timestamp() * 1000)
+
+
+def _dt_ms_opt(dt: datetime | None) -> int | None:
+    return int(dt.timestamp() * 1000) if dt else None
+
+
+def _to_data_types(values: list[str]) -> list[str]:
+    normalize = {
+        "task": "tasks",           "tasks": "tasks",
+        "note": "notes",           "notes": "notes",
+        "timeline": "timelines",   "timelines": "timelines",   "timelineEvents": "timelines",
+        "project": "projects",     "projects": "projects",
+    }
+    seen: set[str] = set()
+    result: list[str] = []
+    for v in values:
+        mapped = normalize.get(v)
+        if mapped and mapped not in seen:
+            seen.add(mapped)
+            result.append(mapped)
+    return result
+
+
+def _to_run_log_response(log: AgentRunLog) -> AgentRunLogResponse:
+    return AgentRunLogResponse(
+        id=log.id,
+        agent_id=log.agent_id,
+        agent_type=log.agent_type,  # type: ignore[arg-type]
+        status=log.status,  # type: ignore[arg-type]
+        items_processed=log.items_processed,
+        items_created=log.items_created,
+        errors=log.errors or [],
+        started_at=_dt_ms(log.started_at),
+        completed_at=_dt_ms_opt(log.completed_at),
+    )
+
+
+def _enforce_agent_limit(tier: str, current_count: int) -> int:
+    limit: int = FEATURES.get(tier, FEATURES["free"])["batch_active"]
+    if limit != -1 and current_count >= limit:
+        raise HTTPException(
+            status_code=status.HTTP_403_FORBIDDEN,
+            detail=f"Agent limit ({limit}) reached for your tier. Upgrade to create more.",
+        )
+    return limit
+
+
+async def _enforce_run_frequency(
+    tier: str,
+    user_id: str,
+    db: AsyncSession,
+) -> None:
+    """Raise HTTP 402 if the user has exceeded their daily batch run limit."""
+    limit: int = FEATURES.get(tier, FEATURES["free"])["batch_runs_per_day"]
+    if limit == -1:
+        return  # unlimited
+
+    today_start = datetime.now(timezone.utc).replace(
+        hour=0, minute=0, second=0, microsecond=0
+    )
+    result = await db.execute(
+        select(func.count(AgentRunLog.id)).where(
+            AgentRunLog.user_id == user_id,
+            AgentRunLog.started_at >= today_start,
+        )
+    )
+    runs_today: int = result.scalar_one()
+
+    if runs_today >= limit:
+        raise HTTPException(
+            status_code=status.HTTP_402_PAYMENT_REQUIRED,
+            detail=f"Daily batch run limit ({limit}) reached for your tier. Upgrade for more runs.",
+        )
+
+
+# ── Catalog ───────────────────────────────────────────────────────────
+
+@router.get("/catalog", response_model=list[AgentCatalogItem])
+async def get_agent_catalog(
+    current_user: UserProfile = Depends(get_current_user),
+) -> list[AgentCatalogItem]:
+    """Return the static list of available agent types and their descriptions."""
+    return [
+        AgentCatalogItem(
+            type="local_directory",
+            name="Local Directory Monitor",
+            description="Watches local directories, extracts data from files using AI",
+        ),
+        AgentCatalogItem(
+            type="gmail",
+            name="Gmail Connector",
+            description="Scans Gmail inbox, extracts tasks/notes from emails",
+        ),
+        AgentCatalogItem(
+            type="teams",
+            name="Microsoft Teams Connector",
+            description="Monitors Teams messages, extracts action items",
+        ),
+        AgentCatalogItem(
+            type="outlook",
+            name="Outlook Connector",
+            description="Scans Outlook inbox, extracts tasks/notes",
+        ),
+    ]
+
+
+@router.post("/can-create", response_model=AgentCreationCheckResponse)
+async def can_create_agent(
+    body: AgentCreationCheckRequest,
+    current_user: UserProfile = Depends(get_current_user),
+) -> AgentCreationCheckResponse:
+    """Check if the user can create one more agent based on billing tier.
+
+    Since configuration is client-owned, the Electron app sends its current
+    active agent count and the backend applies tier limits.
+    """
+    limit: int = FEATURES.get(current_user.tier, FEATURES["free"])["batch_active"]
+    allowed = limit == -1 or body.active_agents < limit
+    return AgentCreationCheckResponse(
+        allowed=allowed,
+        tier=current_user.tier,
+        active_agents=body.active_agents,
+        limit=limit,
+    )
+
+
+@router.post("/trigger", response_model=AgentRunLogResponse, status_code=status.HTTP_202_ACCEPTED)
+async def trigger_agent_run(
+    body: AgentTriggerRequest,
+    current_user: UserProfile = Depends(get_current_user),
+    db: AsyncSession = Depends(get_session),
+) -> AgentRunLogResponse:
+    """Trigger a local agent run using client-provided configuration."""
+    _enforce_agent_limit(current_user.tier, body.active_agents)
+    await _enforce_run_frequency(current_user.tier, current_user.id, db)
+
+    config = LocalAgentConfig(
+        id=str(uuid.uuid4()),
+        user_id=current_user.id,
+        device_id=body.device_id,
+        name="Local Directory Monitor",
+        directory_paths=[body.directory],
+        data_types=_to_data_types(body.what_to_extract),
+        prompt_template=body.custom_agent_prompt,
+        file_extensions=[],
+        schedule_cron=body.batch_interval,
+        enabled=True,
+    )
+
+    # Use the FE's stable agent_id if provided, fall back to the ephemeral config id.
+    stable_agent_id = body.agent_id or config.id
+
+    if is_agent_running(stable_agent_id):
+        raise HTTPException(
+            status_code=status.HTTP_409_CONFLICT,
+            detail="Agent is already running. Only one run per agent is allowed at a time.",
+        )
+
+    run_log = AgentRunLog(
+        agent_id=stable_agent_id,
+        agent_type="local",
+        user_id=current_user.id,
+        status="running",
+    )
+    db.add(run_log)
+    await db.commit()
+    await db.refresh(run_log)
+
+    run_context = {
+        "type": "agent_batch",
+        "run_id": run_log.id,
+        "agent_id": stable_agent_id,
+    }
+
+    asyncio.create_task(
+        run_local_agent(current_user.id, config, run_log, device_manager, run_context)
+    )
+
+    return _to_run_log_response(run_log)

app/api/routes/auth.py

@@ -13,6 +13,7 @@ import uuid
 from datetime import datetime, timedelta, timezone
 
 import bcrypt
+from cryptography.fernet import Fernet
 from fastapi import APIRouter, Depends, HTTPException, status
 from jose import jwt
 from pydantic import BaseModel
@@ -65,6 +66,8 @@ def _make_access_token(user_id: str, email: str, tier: str) -> tuple[str, int]:
 class _RegisterRequest(BaseModel):
     email: str
     password: str
+    name: str | None = None
+    surname: str | None = None
 
 
 class _LoginRequest(BaseModel):
@@ -92,8 +95,11 @@ async def register(
     user = User(
         id=str(uuid.uuid4()),
         email=body.email,
+        name=body.name,
+        surname=body.surname,
         password_hash=_hash_password(body.password),
         tier="free",
+        encryption_key=Fernet.generate_key().decode(),
     )
     db.add(user)
     await db.flush()  # get user.id without committing
@@ -191,7 +197,39 @@ async def refresh(
     )
 
 
+class _UpdateProfileRequest(BaseModel):
+    name: str | None = None
+    surname: str | None = None
+
+
 @router.get("/me", response_model=UserProfile)
 async def me(current_user: UserProfile = Depends(get_current_user)) -> UserProfile:
     """Return the profile for the authenticated user."""
     return current_user
+
+
+@router.put("/me", response_model=UserProfile)
+async def update_profile(
+    body: _UpdateProfileRequest,
+    current_user: UserProfile = Depends(get_current_user),
+    db: AsyncSession = Depends(get_session),
+) -> UserProfile:
+    """Update the authenticated user's name and surname."""
+    result = await db.execute(select(User).where(User.id == current_user.id))
+    user = result.scalar_one()
+
+    if body.name is not None:
+        user.name = body.name
+    if body.surname is not None:
+        user.surname = body.surname
+
+    await db.commit()
+    await db.refresh(user)
+
+    return UserProfile(
+        id=user.id,
+        email=user.email,
+        name=user.name,
+        surname=user.surname,
+        tier=current_user.tier,
+    )

app/api/routes/chat.py

@@ -1,78 +1,29 @@
-"""Chat routes: POST /chat and WebSocket /chat/stream."""
+"""Chat routes: POST /chat (REST fallback).
+
+WebSocket chat is handled by the unified device WS endpoint (/api/v1/ws/device).
+"""
 
 from __future__ import annotations
 
-import asyncio
+from fastapi import APIRouter, Depends
-import json
-
-from fastapi import APIRouter, Depends, WebSocket, WebSocketDisconnect
 from fastapi.responses import JSONResponse
-from jose import JWTError, jwt
 
 from app.api.deps import get_current_user
-from app.config.settings import settings
+from app.core.deep_agent import run_home
-from app.core.orchestrator import orchestrate, orchestrate_stream
 from app.schemas import ChatRequest, UserProfile
 
 router = APIRouter(prefix="/chat", tags=["chat"])
 
-_HEARTBEAT_INTERVAL = 30  # seconds
-
 
 @router.post("")
 async def chat(
     body: ChatRequest,
     current_user: UserProfile = Depends(get_current_user),
 ) -> JSONResponse:
-    """Route a chat message through the orchestrator.
+    """REST fallback for home chat when websocket streaming is unavailable."""
-
+    response = await run_home(
-    Returns ``ChatResponse`` for ``execution_mode='direct'``,
+        user_id=current_user.id,
-    or ``ExecutionPlan`` for ``execution_mode='plan'``.
+        message=body.message,
-    """
+        context=body.context.model_dump(),
-    result = await orchestrate(body)
+    )
-    return JSONResponse(content=result.model_dump())
+    return JSONResponse(content={"response": response})
-
-
-@router.websocket("/stream")
-async def chat_stream(websocket: WebSocket) -> None:
-    """Streaming chat via WebSocket.
-
-    Auth: ``?token=<jwt>`` query param (Bearer not possible during WS handshake).
-
-    Protocol:
-      1. Client sends ``ChatRequest`` as the first JSON text frame.
-      2. Server streams response text chunks.
-      3. Final frame: JSON ``{"done": true, "response": "...", "actions": [...]}``.
-      4. Server pings every 30 s to keep the connection alive.
-    """
-    # Authenticate before accepting the connection
-    token = websocket.query_params.get("token", "")
-    try:
-        payload = jwt.decode(token, settings.JWT_SECRET, algorithms=[settings.JWT_ALGORITHM])
-        user_id: str | None = payload.get("sub")
-        if not user_id:
-            raise JWTError("missing sub")
-    except JWTError:
-        await websocket.close(code=1008)  # 1008 = Policy Violation
-        return
-
-    await websocket.accept()
-
-    try:
-        raw = await websocket.receive_text()
-        body = ChatRequest.model_validate_json(raw)
-
-        async def _heartbeat() -> None:
-            while True:
-                await asyncio.sleep(_HEARTBEAT_INTERVAL)
-                await websocket.send_text(json.dumps({"ping": True}))
-
-        heartbeat_task = asyncio.create_task(_heartbeat())
-        try:
-            async for chunk in orchestrate_stream(body):
-                await websocket.send_text(chunk)
-        finally:
-            heartbeat_task.cancel()
-
-    except WebSocketDisconnect:
-        pass

app/api/routes/device_ws.py

@@ -0,0 +1,417 @@
+"""Device WebSocket endpoint.
+
+Persistent connection from Electron devices to the backend.
+
+  WS  /api/v1/ws/device?token=<jwt>
+
+Auth: JWT passed as ``?token=`` query parameter (Bearer header is not
+available during the WebSocket handshake).
+
+Protocol:
+  1. Client connects → JWT validated → connection accepted.
+  2. Client sends ``device_hello`` frame: ``{ type, device_id, agent_ids }``.
+  3. Backend registers the connection in ``DeviceConnectionManager``.
+  4. Session enters message dispatch loop + heartbeat.
+
+Incoming frame dispatch:
+  - ``tool_result``      → resolves a pending tool-call Future.
+  - ``journey_start``    → starts a guided setup journey session.
+  - ``journey_message``  → continues a journey conversation.
+  - ``pong``             → heartbeat acknowledgement (updates last-seen).
+  - unknown types        → logged, ignored.
+
+Outgoing heartbeat: ``{ "type": "ping" }`` every 30 s.
+
+On disconnect:
+  - Unregisters from DeviceConnectionManager.
+  - Marks all in-progress AgentRunLog rows for this user as ``error``
+    with message "device disconnected".
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import logging
+from uuid import uuid4
+
+from fastapi import APIRouter, WebSocket, WebSocketDisconnect
+from jose import JWTError, jwt
+from sqlalchemy import update
+
+from app.api.routes.agent_setup import handle_journey_message, handle_journey_start
+from app.config.settings import settings
+from app.core.agent_runner import trigger_pending_runs
+from app.core.deep_agent import run_floating_stream, run_home_stream
+from app.core.device_manager import device_manager
+from app.core.memory_middleware import MemoryMiddleware
+from app.core.output_formatter import StreamFormatter
+from app.core.ws_context import clear_client_executor, set_client_executor
+from app.db import async_session
+from app.models import AgentRunLog
+from app.schemas import WsFrameType
+
+logger = logging.getLogger(__name__)
+
+router = APIRouter(prefix="/ws", tags=["device-ws"])
+
+_HEARTBEAT_INTERVAL = 30  # seconds
+_PONG_TIMEOUT = 10  # seconds — grace window after a ping
+
+
+@router.websocket("/device")
+async def device_ws(websocket: WebSocket) -> None:
+    """Persistent WebSocket endpoint for Electron device connections.
+
+    Authentication is via ``?token=<jwt>`` query parameter.
+    """
+    # ── 1. Authenticate before accepting ─────────────────────────────
+    token = websocket.query_params.get("token", "")
+    try:
+        payload = jwt.decode(
+            token, settings.JWT_SECRET, algorithms=[settings.JWT_ALGORITHM]
+        )
+        user_id: str | None = payload.get("sub")
+        if not user_id:
+            raise JWTError("missing sub")
+    except JWTError:
+        await websocket.close(code=1008)  # Policy Violation
+        return
+
+    await websocket.accept()
+
+    # ── 2. Await device_hello frame ───────────────────────────────────
+    try:
+        raw = await asyncio.wait_for(websocket.receive_text(), timeout=15.0)
+    except (asyncio.TimeoutError, WebSocketDisconnect):
+        await websocket.close(code=1008)
+        return
+
+    try:
+        hello = json.loads(raw)
+        if hello.get("type") != WsFrameType.device_hello:
+            raise ValueError("expected device_hello as first frame")
+        device_id: str = hello["device_id"]
+        agent_ids: list[str] = hello.get("agent_ids", [])
+    except (KeyError, ValueError, json.JSONDecodeError) as exc:
+        logger.warning("device_ws: invalid device_hello from user=%s: %s", user_id, exc)
+        await websocket.close(code=1008)
+        return
+
+    # ── 3. Register connection ────────────────────────────────────────
+    device_manager.register(user_id, device_id, websocket)
+    logger.info(
+        "device_ws: connected user=%s device=%s agents=%s",
+        user_id,
+        device_id,
+        agent_ids,
+    )
+
+    # Trigger any overdue agent runs now that the device is connected.
+    asyncio.create_task(trigger_pending_runs(user_id, device_id, device_manager))
+
+    # ── 4. Concurrent message loop + heartbeat ────────────────────────
+    try:
+        await asyncio.gather(
+            _message_loop(websocket, user_id),
+            _heartbeat_loop(websocket),
+        )
+    except WebSocketDisconnect:
+        pass
+    except Exception as exc:
+        logger.warning("device_ws: unhandled exception user=%s: %s", user_id, exc)
+    finally:
+        device_manager.unregister(user_id)
+        logger.info("device_ws: disconnected user=%s device=%s", user_id, device_id)
+        await _mark_runs_disconnected(user_id)
+
+
+# ── Message dispatch loop ─────────────────────────────────────────────
+
+async def _message_loop(websocket: WebSocket, user_id: str) -> None:
+    """Receive frames from Electron and dispatch to the appropriate handler."""
+    async for raw in websocket.iter_text():
+        try:
+            frame: dict = json.loads(raw)
+        except json.JSONDecodeError:
+            logger.warning("device_ws: invalid JSON from user=%s", user_id)
+            continue
+
+        frame_type = frame.get("type")
+
+        if frame_type == WsFrameType.tool_result:
+            call_id = frame.get("id")
+            if call_id:
+                device_manager.resolve_pending_call(user_id, call_id, frame)
+            else:
+                logger.warning(
+                    "device_ws: tool_result missing id from user=%s", user_id
+                )
+
+        elif frame_type == WsFrameType.home_request:
+            asyncio.create_task(
+                _handle_home_request(websocket, user_id, frame)
+            )
+
+        elif frame_type == WsFrameType.floating_request:
+            asyncio.create_task(
+                _handle_floating_request(websocket, user_id, frame)
+            )
+
+        elif frame_type == WsFrameType.journey_start:
+            asyncio.create_task(
+                _handle_journey_start(websocket, user_id, frame)
+            )
+
+        elif frame_type == WsFrameType.journey_message:
+            asyncio.create_task(
+                _handle_journey_message(websocket, user_id, frame)
+            )
+
+        elif frame_type == "pong":
+            # Heartbeat ack — nothing to do, connection is alive.
+            pass
+
+        else:
+            logger.debug(
+                "device_ws: unknown frame type %r from user=%s", frame_type, user_id
+            )
+
+
+# ── v3 Chat Handlers ──────────────────────────────────────────────────
+
+async def _make_ws_executor(websocket: WebSocket, user_id: str):
+    """Return a callback that sends tool_call frames and awaits tool_result."""
+    async def _executor(payload: dict) -> dict:
+        payload["type"] = WsFrameType.tool_call
+        await websocket.send_text(json.dumps(payload))
+        future = device_manager.create_pending_call(user_id, payload["id"])
+        return await future
+    return _executor
+
+
+async def _handle_home_request(
+    websocket: WebSocket,
+    user_id: str,
+    frame: dict,
+) -> None:
+    """Handle a home_request frame — streams HomeFormatter output back on the socket."""
+    request_id = frame.get("request_id") or str(uuid4())
+    message: str = frame.get("message", "")
+    session_id: str = frame.get("session_id") or str(uuid4())
+    logger.info(
+        "device_ws: home_request_start user=%s req=%s session=%s msg=%s",
+        user_id,
+        request_id,
+        session_id,
+        message[:200],
+    )
+
+    # ── Memory: enrich context before LLM call ────────────────────────
+    async with async_session() as db:
+        memory = MemoryMiddleware(db)
+        memory_context = await memory.enrich_context(
+            user_id,
+            message,
+            trace_id=request_id,
+            session_id=session_id,
+        )
+
+    context: dict = {
+        "conversation_history": frame.get("conversation_history", []),
+        "_debug": {"request_id": request_id, "session_id": session_id, "user_id": user_id},
+        **memory_context,
+    }
+
+    executor = await _make_ws_executor(websocket, user_id)
+    set_client_executor(executor)
+    response_chunks: list[str] = []
+    try:
+        event_stream = run_home_stream(user_id, message, context)
+        formatter = StreamFormatter(request_id=request_id)
+        async for ws_frame in formatter.format(event_stream):
+            await websocket.send_text(ws_frame.model_dump_json())
+            # Collect text chunks to build the full response for episode storage
+            if ws_frame.type == "stream_text":  # type: ignore[union-attr]
+                response_chunks.append(ws_frame.chunk)  # type: ignore[union-attr]
+    except Exception as exc:
+        logger.error(
+            "device_ws: home_request failed user=%s req=%s: %s",
+            user_id, request_id, exc,
+        )
+    finally:
+        clear_client_executor()
+
+    # ── Memory: store episode after response ──────────────────────────
+    async with async_session() as db:
+        memory = MemoryMiddleware(db)
+        await memory.store_episode(
+            user_id, session_id, message, "".join(response_chunks), trace_id=request_id
+        )
+    logger.info(
+        "device_ws: home_request_end user=%s req=%s session=%s response_chars=%d",
+        user_id,
+        request_id,
+        session_id,
+        len("".join(response_chunks)),
+    )
+
+
+async def _handle_floating_request(
+    websocket: WebSocket,
+    user_id: str,
+    frame: dict,
+) -> None:
+    """Handle a floating_request frame — streams FloatingFormatter output back on the socket."""
+    request_id = frame.get("request_id") or str(uuid4())
+    message: str = frame.get("message", "")
+    session_id: str = frame.get("session_id") or str(uuid4())
+    scope: dict = frame.get("scope", {})
+    logger.info(
+        "device_ws: floating_request_start user=%s req=%s session=%s scope=%s msg=%s",
+        user_id,
+        request_id,
+        session_id,
+        json.dumps(scope, ensure_ascii=True)[:200],
+        message[:200],
+    )
+
+    # ── Memory: enrich context before LLM call ────────────────────────
+    async with async_session() as db:
+        memory = MemoryMiddleware(db)
+        memory_context = await memory.enrich_context(
+            user_id,
+            message,
+            trace_id=request_id,
+            session_id=session_id,
+        )
+
+    context: dict = {
+        "scope": scope,
+        "_debug": {"request_id": request_id, "session_id": session_id, "user_id": user_id},
+        **memory_context,
+    }
+
+    executor = await _make_ws_executor(websocket, user_id)
+    set_client_executor(executor)
+    response_chunks: list[str] = []
+    try:
+        event_stream = run_floating_stream(user_id, message, context)
+        formatter = StreamFormatter(request_id=request_id)
+        async for ws_frame in formatter.format(event_stream):
+            await websocket.send_text(ws_frame.model_dump_json())
+            if ws_frame.type == "stream_text":  # type: ignore[union-attr]
+                response_chunks.append(ws_frame.chunk)  # type: ignore[union-attr]
+    except Exception as exc:
+        logger.error(
+            "device_ws: floating_request failed user=%s req=%s: %s",
+            user_id, request_id, exc,
+        )
+    finally:
+        clear_client_executor()
+
+    # ── Memory: store episode after response ──────────────────────────
+    async with async_session() as db:
+        memory = MemoryMiddleware(db)
+        await memory.store_episode(
+            user_id, session_id, message, "".join(response_chunks), trace_id=request_id
+        )
+    logger.info(
+        "device_ws: floating_request_end user=%s req=%s session=%s response_chars=%d",
+        user_id,
+        request_id,
+        session_id,
+        len("".join(response_chunks)),
+    )
+
+
+# ── v4 Journey Handlers ─────────────────────────────────────────────
+
+
+async def _handle_journey_start(
+    websocket: WebSocket,
+    user_id: str,
+    frame: dict,
+) -> None:
+    """Handle a journey_start frame — explores directory and sends first question."""
+    executor = await _make_ws_executor(websocket, user_id)
+    set_client_executor(executor)
+    try:
+        reply = await handle_journey_start(user_id, frame)
+        await websocket.send_text(json.dumps(reply))
+    except Exception as exc:
+        logger.error(
+            "device_ws: journey_start failed user=%s: %s", user_id, exc
+        )
+        await websocket.send_text(json.dumps({
+            "type": "journey_reply",
+            "session_id": frame.get("session_id", ""),
+            "message": f"Failed to start journey: {exc}",
+            "done": True,
+            "prompt_template": None,
+        }))
+    finally:
+        clear_client_executor()
+
+
+async def _handle_journey_message(
+    websocket: WebSocket,
+    user_id: str,
+    frame: dict,
+) -> None:
+    """Handle a journey_message frame — continues the journey conversation."""
+    executor = await _make_ws_executor(websocket, user_id)
+    set_client_executor(executor)
+    try:
+        reply = await handle_journey_message(user_id, frame)
+        await websocket.send_text(json.dumps(reply))
+    except Exception as exc:
+        session_id = frame.get("session_id", "")
+        logger.error(
+            "device_ws: journey_message failed user=%s session=%s: %s",
+            user_id, session_id, exc,
+        )
+        await websocket.send_text(json.dumps({
+            "type": "journey_reply",
+            "session_id": session_id,
+            "message": f"Journey error: {exc}",
+            "done": True,
+            "prompt_template": None,
+        }))
+    finally:
+        clear_client_executor()
+
+
+# ── Heartbeat ─────────────────────────────────────────────────────────
+
+async def _heartbeat_loop(websocket: WebSocket) -> None:
+    """Send a ping frame every 30 s to keep the connection alive."""
+    while True:
+        await asyncio.sleep(_HEARTBEAT_INTERVAL)
+        await websocket.send_text(json.dumps({"type": "ping"}))
+
+
+# ── Disconnect cleanup ────────────────────────────────────────────────
+
+async def _mark_runs_disconnected(user_id: str) -> None:
+    """Mark all in-progress AgentRunLog rows as 'error' for this user."""
+    try:
+        async with async_session() as db:
+            await db.execute(
+                update(AgentRunLog)
+                .where(
+                    AgentRunLog.user_id == user_id,
+                    AgentRunLog.status == "running",
+                )
+                .values(
+                    status="error",
+                    errors=["device disconnected"],
+                )
+            )
+            await db.commit()
+    except Exception as exc:
+        logger.error(
+            "device_ws: failed to mark runs as disconnected for user=%s: %s",
+            user_id,
+            exc,
+        )

app/api/routes/plans.py

@@ -1,37 +0,0 @@
-"""Plans routes: GET /plans/playbook and GET /plans/playbook/{plan_id}."""
-
-from __future__ import annotations
-
-from fastapi import APIRouter, Depends, HTTPException, status
-
-from app.api.deps import get_current_user
-from app.core.execution_plan import plan_cache
-from app.schemas import ExecutionPlan, UserProfile
-
-router = APIRouter(prefix="/plans", tags=["plans"])
-
-
-@router.get("/playbook", response_model=list[ExecutionPlan])
-async def list_playbooks(
-    current_user: UserProfile = Depends(get_current_user),
-) -> list[ExecutionPlan]:
-    """Return all cached execution plan playbooks for the authenticated user.
-
-    TODO(Step11): filter by tier — power+ plans gated behind batch_builder feature.
-    """
-    return plan_cache.get_all_playbooks()
-
-
-@router.get("/playbook/{plan_id}", response_model=ExecutionPlan)
-async def get_playbook(
-    plan_id: str,
-    current_user: UserProfile = Depends(get_current_user),
-) -> ExecutionPlan:
-    """Return a specific execution plan playbook by ID."""
-    plan = plan_cache.get_plan(plan_id)
-    if plan is None:
-        raise HTTPException(
-            status_code=status.HTTP_404_NOT_FOUND,
-            detail=f"Plan not found: {plan_id}",
-        )
-    return plan

app/api/routes/vectors.py

@@ -1,4 +1,4 @@
-"""Vectors routes: upsert, search, and delete cloud vector store entries."""
+"""Vectors routes: upsert, search, delete cloud vector store entries, and embed text."""
 
 from __future__ import annotations
 
@@ -6,6 +6,7 @@ from fastapi import APIRouter, Depends
 from pydantic import BaseModel
 
 from app.api.deps import get_current_user
+from app.core.llm import embed
 from app.schemas import (
     UserProfile,
     VectorSearchRequest,
@@ -24,6 +25,14 @@ class _VectorDeleteRequest(BaseModel):
     ids: list[str]
 
 
+class _EmbedRequest(BaseModel):
+    text: str
+
+
+class _EmbedResponse(BaseModel):
+    vector: list[float]
+
+
 @router.post("/vectors/upsert", response_model=dict)
 async def upsert_vectors(
     body: VectorUpsertRequest,
@@ -54,3 +63,17 @@ async def delete_vectors(
     """Delete vectors by ID, scoped to the authenticated user."""
     await _vector_store.delete(current_user.id, body.ids)
     return {"ok": True}
+
+
+@router.post("/vectors/embed", response_model=_EmbedResponse)
+async def embed_text(
+    body: _EmbedRequest,
+    current_user: UserProfile = Depends(get_current_user),
+) -> _EmbedResponse:
+    """Generate a 1536-dim embedding vector for the given text.
+
+    Uses ``text-embedding-3-small`` via OpenAI.  Auth required (JWT).
+    Used by backend tools (note_agent) and Electron (vectordb.ts) alike.
+    """
+    vector = await embed(body.text)
+    return _EmbedResponse(vector=vector)

app/billing/tier_manager.py

@@ -21,6 +21,7 @@ FEATURES: dict[str, dict[str, Any]] = {
     "free": {
         "agents": 3,
         "batch_active": 2,
+        "batch_runs_per_day": 5,
         "cloud_storage_gb": 0,
         "backup_gb": 0,
         "providers": 1,
@@ -31,6 +32,7 @@ FEATURES: dict[str, dict[str, Any]] = {
     "pro": {
         "agents": -1,           # unlimited
         "batch_active": 10,
+        "batch_runs_per_day": 50,
         "cloud_storage_gb": 5,
         "backup_gb": 5,
         "providers": -1,
@@ -41,6 +43,7 @@ FEATURES: dict[str, dict[str, Any]] = {
     "power": {
         "agents": -1,
         "batch_active": -1,     # unlimited
+        "batch_runs_per_day": -1,  # unlimited
         "cloud_storage_gb": 25,
         "backup_gb": 25,
         "providers": -1,
@@ -51,6 +54,7 @@ FEATURES: dict[str, dict[str, Any]] = {
     "team": {
         "agents": -1,
         "batch_active": -1,
+        "batch_runs_per_day": -1,  # unlimited
         "cloud_storage_gb": -1,  # unlimited
         "backup_gb": -1,         # unlimited
         "providers": -1,
@@ -77,16 +81,18 @@ class TierManager:
     async def get_tier(self, user_id: str, db: AsyncSession) -> BillingTier:
         """Return the current billing tier for ``user_id`` from the DB.
 
-        Falls back to ``'free'`` when no subscription row exists.
+        Falls back to ``'power'`` in dev (unlimited) or ``'free'`` in prod
+        when no subscription row exists.
         """
         from app.models import Subscription  # noqa: PLC0415
+        from app.config.settings import settings  # noqa: PLC0415
 
         result = await db.execute(
             select(Subscription.tier).where(Subscription.user_id == user_id)
         )
         tier: str | None = result.scalar_one_or_none()
         if tier is None or tier not in FEATURES:
-            return "free"
+            return "power" if settings.ENV == "dev" else "free"
         return tier  # type: ignore[return-value]
 
     # ── Feature access ───────────────────────────────────────────────────

app/config/settings.py

@@ -1,5 +1,5 @@
 from typing import Literal
-from pydantic_settings import BaseSettings
+from pydantic_settings import BaseSettings, SettingsConfigDict
 
 
 class Settings(BaseSettings):
@@ -26,17 +26,36 @@ class Settings(BaseSettings):
     OPENAI_API_KEY: str = ""
     ANTHROPIC_API_KEY: str = ""
     GOOGLE_API_KEY: str = ""
+    CEREBRAS_API_KEY: str = ""
 
     LLM_MODEL: str = "gpt-4o"
-    LLM_ROUTER_MODEL: str = "gpt-4o-mini"
+    LLM_EMBED_MODEL: str = "text-embedding-3-small"
+
+    # GitHub Copilot OAuth token storage directory.
+    # Leave empty to use the LiteLLM default (~/.config/litellm/github_copilot).
+    # In Docker, set this to a path backed by a named volume so tokens survive restarts.
+    GITHUB_COPILOT_TOKEN_DIR: str = ""
+
+    # OAuth client credentials — used for Gmail and Microsoft (Outlook/Teams) flows.
+    GMAIL_CLIENT_ID: str = ""
+    GMAIL_CLIENT_SECRET: str = ""
+    MS_CLIENT_ID: str = ""
+    MS_CLIENT_SECRET: str = ""
+    # MS_TENANT_ID: set to 'common' to allow multi-tenant (personal + work accounts).
+    MS_TENANT_ID: str = "common"
+
+    # Fernet key (URL-safe base64, 32-byte key) for at-rest encryption of OAuth
+    # tokens stored in cloud_agent_configs.oauth_token_encrypted.
+    # Generate with: from cryptography.fernet import Fernet; Fernet.generate_key()
+    OAUTH_ENCRYPTION_KEY: str = ""
 
     CORS_ORIGINS: list[str] = ["app://.", "http://localhost:3000", "http://localhost:5173"]
 
     ENV: Literal["dev", "prod"] = "dev"
 
-    class Config:
+    model_config = SettingsConfigDict(
-        env_file = ".env"
+        env_file=".env", env_file_encoding="utf-8", extra="ignore"
-        env_file_encoding = "utf-8"
+    )
 
 
 settings = Settings()

app/core/agent_registry.py

@@ -1,4 +1,4 @@
-"""Agent Registry — base classes and singleton registry for chat agents."""
+"""Minimal agent base types retained for compatibility with batch runners."""
 
 from __future__ import annotations
 
@@ -7,7 +7,7 @@ from typing import Any
 
 
 class BaseAgent(ABC):
-    """Common base for all agents."""
+    """Common base for non-chat agents still using the old base contract."""
 
     def __init__(
         self,
@@ -27,111 +27,4 @@ class BaseAgent(ABC):
 
     @property
     def skills(self) -> list[str]:
-        """Override in subclasses to advertise capabilities."""
         return []
-
-
-class ChatAgent(BaseAgent):
-    """Base class for LLM-powered chat agents."""
-
-    @abstractmethod
-    async def handle(self, query: str, context: dict[str, Any]) -> str:
-        """Process a user query and return a text response."""
-        ...
-
-    @abstractmethod
-    def get_tools(self) -> list[Any]:
-        """Return LangChain tool definitions available to this agent."""
-        ...
-
-    async def _tool_loop(
-        self,
-        llm: Any,
-        messages: list[Any],
-        tools: list[Any],
-        max_iter: int = 5,
-    ) -> str:
-        """Shared tool-calling loop.
-
-        Binds *tools* to *llm*, invokes iteratively until the model stops
-        requesting tool calls or *max_iter* is reached, and returns the
-        final text response.
-        """
-        from langchain_core.messages import AIMessage, ToolMessage
-
-        llm_with_tools = llm.bind_tools(tools) if tools else llm
-
-        for _ in range(max_iter):
-            response: AIMessage = await llm_with_tools.ainvoke(messages)
-            messages.append(response)
-
-            if not response.tool_calls:
-                return str(response.content)
-
-            # Execute each requested tool call
-            tool_map = {t.name: t for t in tools}
-            for call in response.tool_calls:
-                tool_fn = tool_map.get(call["name"])
-                if tool_fn is None:
-                    result = f"Unknown tool: {call['name']}"
-                else:
-                    result = await tool_fn.ainvoke(call["args"])
-                messages.append(
-                    ToolMessage(content=str(result), tool_call_id=call["id"])
-                )
-
-        # Exhausted iterations — ask model for a final answer without tools
-        response = await llm.ainvoke(messages)
-        return str(response.content)
-
-
-class AgentRegistry:
-    """Singleton registry for ChatAgent subclasses."""
-
-    _instance: AgentRegistry | None = None
-
-    def __init__(self) -> None:
-        self._agents: dict[str, type[ChatAgent]] = {}
-
-    def __new__(cls) -> AgentRegistry:
-        if cls._instance is None:
-            cls._instance = super().__new__(cls)
-            cls._instance._agents = {}
-        return cls._instance
-
-    # ── public API ───────────────────────────────────────────────────
-
-    def register(self, agent_class: type[ChatAgent]) -> type[ChatAgent]:
-        """Class decorator — registers an agent by its name."""
-        instance = agent_class()
-        name = instance.get_name()
-        self._agents[name] = agent_class
-        return agent_class
-
-    def get(self, name: str) -> ChatAgent:
-        """Return a fresh instance of the named agent."""
-        cls = self._agents.get(name)
-        if cls is None:
-            raise KeyError(f"Agent not found: {name}")
-        return cls()
-
-    def list_agents(self) -> list[dict[str, str]]:
-        """Return ``[{name, description}]`` for the orchestrator prompt."""
-        result: list[dict[str, str]] = []
-        for cls in self._agents.values():
-            inst = cls()
-            result.append(
-                {"name": inst.get_name(), "description": inst.get_description()}
-            )
-        return result
-
-    async def call_agent(
-        self, name: str, query: str, context: dict[str, Any]
-    ) -> str:
-        """Instantiate the named agent and call its ``handle`` method."""
-        agent = self.get(name)
-        return await agent.handle(query, context)
-
-
-# Module-level singleton
-registry = AgentRegistry()

app/core/deep_agent.py

@@ -0,0 +1,846 @@
+"""Single-agent runners for home and floating chat contexts."""
+
+from __future__ import annotations
+
+import json
+import logging
+import re
+from datetime import date
+from collections.abc import AsyncGenerator
+from typing import Any, Literal
+
+from langchain_core.messages import AIMessage, HumanMessage, SystemMessage, ToolMessage
+from langchain_core.tools import tool
+
+from app.agents.note_agent import NOTE_TOOLS
+from app.agents.project_agent import PROJECT_TOOLS
+from app.agents.task_agent import TASK_TOOLS
+from app.agents.timeline_agent import TIMELINE_TOOLS
+from app.core.llm import get_llm
+from app.core.memory_middleware import MemoryMiddleware
+from app.core.ws_context import clear_tool_result_collector, execute_on_client, set_tool_result_collector
+from app.db import async_session
+
+logger = logging.getLogger(__name__)
+
+FloatingDomainType = Literal["task", "timeline", "project", "node"]
+FloatingDomainSection = Literal["task", "timeline", "note"]
+
+_HOME_SINGLE_AGENT_SYSTEM = (
+    "You are the home assistant with direct access to all tools: tasks, projects, notes, timelines, and memory tools. "
+    "Always use tools for factual data retrieval before answering. "
+    "When the user asks to remember, forget, or update what you know about them, use memory tools. "
+    "If context.context.resolved_project_id exists, use it as project_id for scoped list calls. "
+    "Return markdown and use tags when relevant: <project>[ids]</project>, <task>[ids]</task>, "
+    "<note>[ids]</note>, <timeline>[ids]</timeline>, <chart>{json}</chart>. "
+    "When listing tasks or timelines, each id tag must be on its own line with no prefix/suffix text. "
+    "Never put titles, priorities, or dates on the same line as <task> or <timeline> tags. "
+    "For questions about upcoming timelines (e.g. 'prossimi eventi'), include only future items in the current month unless the user asks a different range. "
+    "For upcoming tasks, after tag lines add a short recommendation based on due date and priority."
+)
+
+_FLOATING_SINGLE_AGENT_SYSTEM = (
+    "You are the floating assistant with direct access to all tools: tasks, projects, notes, timelines, and memory tools. "
+    "Stay focused on the floating scope in context.scope and answer concisely. "
+    "Return plain text only. Do not output XML/HTML-like tags such as <task>, <project>, <note>, <timeline>, or any bracketed id tag wrappers. "
+    "Always use tools for factual data retrieval before answering. "
+    "When the user asks to remember, forget, or update what you know about them, use memory tools. "
+    "If context.context.resolved_project_id exists, use it as project_id for scoped list calls. "
+)
+
+_FLOATING_DOMAIN_CLASSIFIER_SYSTEM = (
+    "You are a strict domain classifier for websocket floating requests. "
+    "Return ONLY a JSON object with keys: type, id, section. "
+    "Allowed type values: task, timeline, project, node. "
+    "Allowed section values: task, timeline, note, or null. "
+    "Rules: infer from user message intent first; do not blindly trust scope.type. "
+    "If user asks tasks/timeline/notes for a project, set type=project and section accordingly. "
+    "If project id is unknown but context.resolved_project_id exists, use it as id. "
+    "If id is unknown, use null. "
+    "No markdown, no prose, JSON only."
+)
+
+
+def _as_text(content: Any) -> str:
+    if content is None:
+        return ""
+    if isinstance(content, str):
+        return content
+    if isinstance(content, list):
+        parts: list[str] = []
+        for item in content:
+            if isinstance(item, str):
+                parts.append(item)
+            elif isinstance(item, dict):
+                text = item.get("text")
+                if isinstance(text, str):
+                    parts.append(text)
+        return "".join(parts)
+    return str(content)
+
+
+def _candidate_tokens(message: str) -> list[str]:
+    tokens = re.findall(r"[a-zA-Z0-9_-]+", message.lower())
+    return [token for token in tokens if len(token) >= 3]
+
+
+async def _resolve_project_id_from_message(message: str) -> str | None:
+    """Resolve likely project UUID from user message using client project list."""
+    try:
+        result = await execute_on_client(action="select", table="projects")
+    except Exception as exc:
+        logger.warning("deep_agent: project resolve select failed: %s", exc)
+        return None
+
+    rows = result.get("rows", [])
+    if not isinstance(rows, list) or not rows:
+        return None
+
+    tokens = _candidate_tokens(message)
+    scored: list[tuple[int, dict[str, Any]]] = []
+    for row in rows:
+        if not isinstance(row, dict):
+            continue
+        name = str(row.get("name", "")).lower()
+        score = sum(1 for token in tokens if token in name)
+        if score > 0:
+            scored.append((score, row))
+
+    if not scored:
+        return None
+
+    scored.sort(key=lambda item: item[0], reverse=True)
+    top_score = scored[0][0]
+    top_rows = [row for score, row in scored if score == top_score]
+    if len(top_rows) != 1:
+        return None
+
+    project_id = top_rows[0].get("id")
+    return project_id if isinstance(project_id, str) else None
+
+
+def _needs_project_resolution(message: str) -> bool:
+    lowered = message.lower()
+    return any(keyword in lowered for keyword in ["project", "progetto", "progetti", "whitelist"])
+
+
+async def _prepare_context(message: str, context: dict[str, Any]) -> dict[str, Any]:
+    prepared = dict(context)
+    if _needs_project_resolution(message):
+        resolved_project_id = await _resolve_project_id_from_message(message)
+        if resolved_project_id:
+            prepared["resolved_project_id"] = resolved_project_id
+            logger.info("deep_agent: resolved_project_id=%s", resolved_project_id)
+    return prepared
+
+
+def _all_tools() -> list[Any]:
+    return [*TASK_TOOLS, *PROJECT_TOOLS, *NOTE_TOOLS, *TIMELINE_TOOLS]
+
+
+def _trace_id_from_context(context: dict[str, Any]) -> str | None:
+    debug = context.get("_debug")
+    if isinstance(debug, dict):
+        request_id = debug.get("request_id")
+        if isinstance(request_id, str) and request_id:
+            return request_id
+    return None
+
+
+def _context_for_model(context: dict[str, Any]) -> dict[str, Any]:
+    sanitized = dict(context)
+    sanitized.pop("_debug", None)
+    return sanitized
+
+
+_TAG_LINE_RE = re.compile(r"<(task|timeline)>\[[^\]]+\]</\1>")
+_TIMELINE_DMY_RE = re.compile(r"(?P<d>\d{2})/(?P<m>\d{2})/(?P<y>\d{4})")
+
+
+def _is_upcoming_timeline_query(message: str) -> bool:
+    lowered = message.lower()
+    has_upcoming = "prossim" in lowered or "upcoming" in lowered or "next" in lowered
+    has_timeline_topic = any(
+        token in lowered
+        for token in ("event", "evento", "eventi", "timeline", "milestone", "scaden")
+    )
+    return has_upcoming and has_timeline_topic
+
+
+def _timeline_date_in_current_month_or_future(dmy: str) -> bool:
+    match = _TIMELINE_DMY_RE.search(dmy)
+    if not match:
+        return True
+    try:
+        parsed = date(
+            int(match.group("y")),
+            int(match.group("m")),
+            int(match.group("d")),
+        )
+    except ValueError:
+        return True
+
+    today = date.today()
+    return parsed >= today and parsed.year == today.year and parsed.month == today.month
+
+
+def _normalize_tagged_list_lines(text: str, message: str) -> str:
+    if not text:
+        return text
+
+    upcoming_timeline_only = _is_upcoming_timeline_query(message)
+    output_lines: list[str] = []
+
+    for line in text.splitlines():
+        matches = list(_TAG_LINE_RE.finditer(line))
+        if not matches:
+            output_lines.append(line)
+            continue
+
+        had_non_tag_text = _TAG_LINE_RE.sub("", line).strip(" -\t0123456789.*:)")
+        if not had_non_tag_text and len(matches) == 1:
+            tag_text = matches[0].group(0)
+            if (
+                upcoming_timeline_only
+                and "<timeline>" in tag_text
+                and not _timeline_date_in_current_month_or_future(line)
+            ):
+                continue
+            output_lines.append(tag_text)
+            continue
+
+        for match in matches:
+            tag_text = match.group(0)
+            if (
+                upcoming_timeline_only
+                and "<timeline>" in tag_text
+                and not _timeline_date_in_current_month_or_future(line)
+            ):
+                continue
+            output_lines.append(tag_text)
+
+    return "\n".join(output_lines)
+
+
+_GENERIC_TAG_RE = re.compile(r"</?(task|project|note|timeline|chart)>", re.IGNORECASE)
+_BRACKETED_ID_RE = re.compile(r"\[(?:[0-9a-fA-F-]{8,}|[A-Za-z0-9_-]{8,})\]")
+_FLOATING_EMPTY_FALLBACK = "No results found."
+
+
+def _strip_floating_markup_fragment(text: str) -> str:
+    if not text:
+        return text
+    cleaned = _GENERIC_TAG_RE.sub("", text)
+    return _BRACKETED_ID_RE.sub("", cleaned)
+
+
+def _strip_floating_markup(text: str) -> str:
+    """Ensure floating responses stay plain text with no XML-like tag wrappers."""
+    if not text:
+        return text
+
+    cleaned = _strip_floating_markup_fragment(text)
+    # Collapse excessive spaces introduced by tag/id removal while preserving lines.
+    lines = [re.sub(r"[ \t]{2,}", " ", line).strip() for line in cleaned.splitlines()]
+    return "\n".join(line for line in lines if line)
+
+
+def _fallback_from_raw_floating_text(raw_text: str) -> str:
+    fallback = _strip_floating_markup_fragment(raw_text or "")
+    fallback = re.sub(r"[ \t]{2,}", " ", fallback).strip()
+    return fallback or _FLOATING_EMPTY_FALLBACK
+
+
+class _FloatingStreamSanitizer:
+    """Streaming sanitizer that removes floating markup without buffering the full answer."""
+
+    def __init__(self) -> None:
+        self._pending = ""
+
+    @staticmethod
+    def _split_safe_boundary(text: str) -> tuple[str, str]:
+        boundary = len(text)
+
+        last_lt = text.rfind("<")
+        if last_lt != -1 and ">" not in text[last_lt:]:
+            boundary = min(boundary, last_lt)
+
+        last_lb = text.rfind("[")
+        if last_lb != -1 and "]" not in text[last_lb:]:
+            boundary = min(boundary, last_lb)
+
+        if boundary == len(text):
+            return text, ""
+        return text[:boundary], text[boundary:]
+
+    def feed(self, chunk: str) -> str:
+        combined = f"{self._pending}{chunk}"
+        safe_text, self._pending = self._split_safe_boundary(combined)
+        return _strip_floating_markup_fragment(safe_text)
+
+    def finalize(self) -> str:
+        # Drop dangling unfinished wrappers at the very end.
+        tail = re.sub(r"<[^>\n]*$", "", self._pending)
+        tail = re.sub(r"\[[^\]\n]*$", "", tail)
+        self._pending = ""
+        return _strip_floating_markup_fragment(tail)
+
+
+def _normalize_memory_label(path_or_label: str) -> str:
+    value = path_or_label.strip()
+    if value.startswith("/memories/"):
+        value = value[len("/memories/"):]
+    value = value.strip("/")
+    return value
+
+
+def _memory_tools(user_id: str, trace_id: str | None) -> list[Any]:
+    @tool
+    async def memory_list_blocks() -> str:
+        """List all core memory blocks currently stored for the user."""
+        logger.info("deep_agent: memory_list_blocks trace=%s user=%s", trace_id or "-", user_id)
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            blocks = await memory.list_core_blocks(user_id)
+        if not blocks:
+            return "No memory blocks found."
+        lines = [f"- {b['label']}: {b['value']}" for b in blocks]
+        return "Memory blocks:\n" + "\n".join(lines)
+
+    @tool
+    async def memory_get(path_or_label: str) -> str:
+        """Get one memory block by label or /memories/<label> path."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_get trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            value = await memory.get_core_block(user_id, label)
+        if value is None:
+            return f"Memory block '{label}' not found."
+        return f"Memory block '{label}':\n{value}"
+
+    @tool
+    async def memory_create(path_or_label: str, value: str) -> str:
+        """Create or overwrite a memory block value by label or /memories/<label> path."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_create trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            await memory.update_core(user_id, label, value, trace_id=trace_id)
+        return f"Memory block '{label}' saved."
+
+    @tool
+    async def memory_append(path_or_label: str, content: str) -> str:
+        """Append content to a memory block, creating it if missing."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_append trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            await memory.append_core(user_id, label, content)
+        return f"Memory block '{label}' appended."
+
+    @tool
+    async def memory_replace(path_or_label: str, old_string: str, new_string: str) -> str:
+        """Replace one exact string in a memory block."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_replace trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            changed = await memory.replace_core(user_id, label, old_string, new_string)
+        if not changed:
+            return f"No replacement made in '{label}' (old string not found)."
+        return f"Memory block '{label}' updated."
+
+    @tool
+    async def memory_delete(path_or_label: str) -> str:
+        """Delete a memory block by label or /memories/<label> path."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_delete trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            deleted = await memory.delete_core(user_id, label)
+        if not deleted:
+            return f"Memory block '{label}' not found."
+        return f"Memory block '{label}' deleted."
+
+    @tool
+    async def archival_memory_insert(content: str) -> str:
+        """Insert a long-term archival memory entry."""
+        logger.info("deep_agent: archival_memory_insert trace=%s user=%s", trace_id or "-", user_id)
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            await memory.insert_archival(user_id, content, source="assistant")
+        return "Archival memory saved."
+
+    @tool
+    async def archival_memory_search(query: str, top_k: int = 5) -> str:
+        """Search long-term archival memory by semantic fallback (keyword currently)."""
+        logger.info("deep_agent: archival_memory_search trace=%s user=%s query=%s", trace_id or "-", user_id, query[:80])
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            results = await memory.search_archival(user_id, query, top_k=top_k)
+        if not results:
+            return "No archival memory results found."
+        lines = [f"- {item}" for item in results]
+        return "Archival memory results:\n" + "\n".join(lines)
+
+    @tool
+    async def conversation_search(query: str, top_k: int = 5) -> str:
+        """Search recall memory from prior episodic conversation summaries."""
+        logger.info("deep_agent: conversation_search trace=%s user=%s query=%s", trace_id or "-", user_id, query[:80])
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            results = await memory.search_recall(user_id, query, top_k=top_k)
+        if not results:
+            return "No recall memory results found."
+        lines = [f"- {item}" for item in results]
+        return "Recall memory results:\n" + "\n".join(lines)
+
+    return [
+        memory_list_blocks,
+        memory_get,
+        memory_create,
+        memory_append,
+        memory_replace,
+        memory_delete,
+        archival_memory_insert,
+        archival_memory_search,
+        conversation_search,
+    ]
+
+
+def _all_tools_for_user(user_id: str, trace_id: str | None) -> list[Any]:
+    return [*_all_tools(), *_memory_tools(user_id, trace_id)]
+
+
+def _detect_domain_section(message: str) -> FloatingDomainSection | None:
+    lowered = message.lower()
+    if any(keyword in lowered for keyword in ["timeline", "milestone", "release", "schedule"]):
+        return "timeline"
+    if any(keyword in lowered for keyword in ["task", "tasks", "todo", "attivit", "azione"]):
+        return "task"
+    if any(keyword in lowered for keyword in ["note", "notes", "memo", "document"]):
+        return "note"
+    return None
+
+
+def _normalize_domain_payload(payload: dict[str, Any], fallback_id: str | None) -> dict[str, str | None]:
+    type_raw = str(payload.get("type") or "").strip().lower()
+    domain_type: FloatingDomainType = "task"
+    if type_raw in {"task", "timeline", "project", "node"}:
+        domain_type = type_raw
+
+    id_value = payload.get("id")
+    domain_id = id_value if isinstance(id_value, str) and id_value.strip() else None
+    if domain_type == "project" and not domain_id:
+        domain_id = fallback_id
+
+    section_raw = payload.get("section")
+    section: FloatingDomainSection | None = None
+    if isinstance(section_raw, str):
+        section_candidate = section_raw.strip().lower()
+        if section_candidate in {"task", "timeline", "note"}:
+            section = section_candidate
+
+    if domain_type != "project":
+        section = None
+
+    return {
+        "type": domain_type,
+        "id": domain_id,
+        "section": section,
+    }
+
+
+def _parse_json_object(text: str) -> dict[str, Any] | None:
+    raw = text.strip()
+    if not raw:
+        return None
+    try:
+        parsed = json.loads(raw)
+        return parsed if isinstance(parsed, dict) else None
+    except json.JSONDecodeError:
+        pass
+
+    match = re.search(r"\{.*\}", raw, re.DOTALL)
+    if not match:
+        return None
+    try:
+        parsed = json.loads(match.group(0))
+    except json.JSONDecodeError:
+        return None
+    return parsed if isinstance(parsed, dict) else None
+
+
+def _infer_floating_domain_rule_based(message: str, context: dict[str, Any]) -> dict[str, str | None]:
+    section = _detect_domain_section(message)
+    scope = context.get("scope") if isinstance(context, dict) else None
+    resolved_project_id = context.get("resolved_project_id") if isinstance(context, dict) else None
+    project_id = resolved_project_id if isinstance(resolved_project_id, str) and resolved_project_id else None
+
+    if isinstance(scope, dict):
+        scope_type = str(scope.get("type") or "").strip().lower()
+        scope_id = scope.get("id")
+        scope_id_value = scope_id if isinstance(scope_id, str) and scope_id else None
+
+        if scope_type in {"task", "tasks"}:
+            return {"type": "task", "id": scope_id_value, "section": None}
+        if scope_type in {"project", "projects"}:
+            project_scope_id = scope_id_value or project_id
+            return {
+                "type": "project",
+                "id": project_scope_id,
+                "section": section,
+            }
+        if scope_type in {"note", "notes"}:
+            return {
+                "type": "node",
+                "id": scope_id_value,
+                "section": None,
+            }
+        if scope_type in {"timeline", "timelines"}:
+            return {"type": "timeline", "id": scope_id_value, "section": None}
+
+    lowered = message.lower()
+    if any(keyword in lowered for keyword in ["project", "progetto", "client"]) or project_id:
+        return {
+            "type": "project",
+            "id": project_id,
+            "section": section,
+        }
+    if section == "timeline":
+        return {"type": "timeline", "id": None, "section": None}
+    if section == "note":
+        return {"type": "node", "id": None, "section": None}
+    return {"type": "task", "id": None, "section": None}
+
+
+async def _infer_floating_domain(message: str, context: dict[str, Any]) -> dict[str, str | None]:
+    resolved_project_id = context.get("resolved_project_id") if isinstance(context, dict) else None
+    project_id = resolved_project_id if isinstance(resolved_project_id, str) and resolved_project_id else None
+
+    classifier_context = {
+        "scope": context.get("scope") if isinstance(context.get("scope"), dict) else None,
+        "resolved_project_id": project_id,
+    }
+
+    try:
+        llm = get_llm()
+        response = await llm.ainvoke(
+            [
+                SystemMessage(content=_FLOATING_DOMAIN_CLASSIFIER_SYSTEM),
+                HumanMessage(
+                    content=(
+                        f"Message:\n{message}\n\n"
+                        f"Context:\n{json.dumps(classifier_context, ensure_ascii=True)}"
+                    )
+                ),
+            ]
+        )
+        parsed = _parse_json_object(_as_text(response.content))
+        if parsed is not None:
+            domain = _normalize_domain_payload(parsed, project_id)
+            logger.info(
+                "deep_agent: floating_domain_classified type=%s id=%s section=%s",
+                domain.get("type"),
+                domain.get("id"),
+                domain.get("section"),
+            )
+            return domain
+        logger.warning("deep_agent: floating_domain classifier returned non-json output")
+    except Exception as exc:
+        logger.warning("deep_agent: floating_domain classifier failed: %s", exc)
+
+    return _infer_floating_domain_rule_based(message, context)
+
+
+async def _run_single_agent(
+    *,
+    user_id: str,
+    system_prompt: str,
+    message: str,
+    context: dict[str, Any],
+    max_steps: int = 6,
+) -> str:
+    trace_id = _trace_id_from_context(context)
+    llm = get_llm()
+    tools = _all_tools_for_user(user_id, trace_id)
+    model_context = _context_for_model(context)
+    logger.info("deep_agent: run_single_agent_start trace=%s user=%s", trace_id or "-", user_id)
+    llm_with_tools = llm.bind_tools(tools)
+    messages: list[Any] = [
+        SystemMessage(content=system_prompt),
+        HumanMessage(
+            content=(
+                f"User message:\n{message}\n\n"
+                f"Context:\n{json.dumps({'context': model_context}, ensure_ascii=True)[:3500]}"
+            )
+        ),
+    ]
+
+    tool_calls_count = 0
+    collected: list[dict[str, Any]] = []
+    set_tool_result_collector(collected)
+    try:
+        for _ in range(max_steps):
+            response: AIMessage = await llm_with_tools.ainvoke(messages)
+            messages.append(response)
+
+            if not response.tool_calls:
+                final_text = _as_text(response.content)
+                logger.info(
+                    "deep_agent: run_single_agent_end trace=%s user=%s tool_calls=%d response_chars=%d",
+                    trace_id or "-",
+                    user_id,
+                    tool_calls_count,
+                    len(final_text),
+                )
+                return final_text
+
+            tool_map = {tool_def.name: tool_def for tool_def in tools}
+            for call in response.tool_calls:
+                tool_calls_count += 1
+                call_id = str(call.get("id", ""))
+                call_name = str(call.get("name", ""))
+                call_args = call.get("args", {})
+                logger.info(
+                    "deep_agent: AI->Tool tool_call_id=%s tool=%s args=%s",
+                    call_id,
+                    call_name,
+                    json.dumps(call_args, ensure_ascii=True)[:800],
+                )
+
+                tool_fn = tool_map.get(call_name)
+                if tool_fn is None:
+                    tool_output = f"Unknown tool: {call_name}"
+                else:
+                    tool_output = await tool_fn.ainvoke(call_args)
+
+                logger.info(
+                    "deep_agent: Tool->AI tool_call_id=%s tool=%s output=%s",
+                    call_id,
+                    call_name,
+                    str(tool_output)[:1200],
+                )
+
+                messages.append(ToolMessage(content=str(tool_output), tool_call_id=call["id"]))
+
+        final = await llm.ainvoke(messages)
+        final_text = _as_text(final.content)
+        logger.info(
+            "deep_agent: run_single_agent_end trace=%s user=%s tool_calls=%d response_chars=%d fallback=1",
+            trace_id or "-",
+            user_id,
+            tool_calls_count,
+            len(final_text),
+        )
+        return final_text
+    finally:
+        clear_tool_result_collector()
+
+
+async def _run_single_agent_stream(
+    *,
+    user_id: str,
+    system_prompt: str,
+    message: str,
+    context: dict[str, Any],
+    max_steps: int = 6,
+) -> AsyncGenerator[tuple[str, Any], None]:
+    trace_id = _trace_id_from_context(context)
+    llm = get_llm()
+    tools = _all_tools_for_user(user_id, trace_id)
+    model_context = _context_for_model(context)
+    logger.info("deep_agent: run_single_agent_stream_start trace=%s user=%s", trace_id or "-", user_id)
+    llm_with_tools = llm.bind_tools(tools)
+    messages: list[Any] = [
+        SystemMessage(content=system_prompt),
+        HumanMessage(
+            content=(
+                f"User message:\n{message}\n\n"
+                f"Context:\n{json.dumps({'context': model_context}, ensure_ascii=True)[:3500]}"
+            )
+        ),
+    ]
+
+    tool_calls_count = 0
+    streamed_chars = 0
+    collected: list[dict[str, Any]] = []
+    set_tool_result_collector(collected)
+    try:
+        for _ in range(max_steps):
+            response: AIMessage = await llm_with_tools.ainvoke(messages)
+            messages.append(response)
+
+            if not response.tool_calls:
+                emitted_any = False
+                async for chunk in llm.astream(messages):
+                    token = _as_text(getattr(chunk, "content", ""))
+                    if token:
+                        streamed_chars += len(token)
+                        emitted_any = True
+                        yield "token", token
+
+                # Some providers return final text in `response.content` but stream no chunks.
+                if not emitted_any:
+                    fallback_text = _as_text(response.content)
+                    if fallback_text:
+                        streamed_chars += len(fallback_text)
+                        yield "token", fallback_text
+                logger.info(
+                    "deep_agent: run_single_agent_stream_end trace=%s user=%s tool_calls=%d response_chars=%d",
+                    trace_id or "-",
+                    user_id,
+                    tool_calls_count,
+                    streamed_chars,
+                )
+                return
+
+            tool_map = {tool_def.name: tool_def for tool_def in tools}
+            for call in response.tool_calls:
+                tool_calls_count += 1
+                call_id = str(call.get("id", ""))
+                call_name = str(call.get("name", ""))
+                call_args = call.get("args", {})
+                logger.info(
+                    "deep_agent: AI->Tool tool_call_id=%s tool=%s args=%s",
+                    call_id,
+                    call_name,
+                    json.dumps(call_args, ensure_ascii=True)[:800],
+                )
+
+                tool_fn = tool_map.get(call_name)
+                if tool_fn is None:
+                    tool_output = f"Unknown tool: {call_name}"
+                else:
+                    tool_output = await tool_fn.ainvoke(call_args)
+
+                logger.info(
+                    "deep_agent: Tool->AI tool_call_id=%s tool=%s output=%s",
+                    call_id,
+                    call_name,
+                    str(tool_output)[:1200],
+                )
+
+                messages.append(ToolMessage(content=str(tool_output), tool_call_id=call["id"]))
+
+        async for chunk in llm.astream(messages):
+            token = _as_text(getattr(chunk, "content", ""))
+            if token:
+                streamed_chars += len(token)
+                yield "token", token
+        logger.info(
+            "deep_agent: run_single_agent_stream_end trace=%s user=%s tool_calls=%d response_chars=%d fallback=1",
+            trace_id or "-",
+            user_id,
+            tool_calls_count,
+            streamed_chars,
+        )
+    finally:
+        clear_tool_result_collector()
+
+
+async def run_home(user_id: str, message: str, context: dict[str, Any]) -> str:
+    prepared_context = await _prepare_context(message, context)
+    response = await _run_single_agent(
+        user_id=user_id,
+        system_prompt=_HOME_SINGLE_AGENT_SYSTEM,
+        message=message,
+        context=prepared_context,
+    )
+    return _normalize_tagged_list_lines(response, message)
+
+
+async def run_floating(user_id: str, message: str, context: dict[str, Any]) -> tuple[str, dict[str, str | None]]:
+    prepared_context = await _prepare_context(message, context)
+    domain = await _infer_floating_domain(message, prepared_context)
+    response = await _run_single_agent(
+        user_id=user_id,
+        system_prompt=_FLOATING_SINGLE_AGENT_SYSTEM,
+        message=message,
+        context=prepared_context,
+    )
+    sanitized = _strip_floating_markup(response)
+    if not sanitized and response:
+        sanitized = _fallback_from_raw_floating_text(response)
+    return sanitized, domain
+
+
+async def run_home_stream(
+    user_id: str,
+    message: str,
+    context: dict[str, Any],
+) -> AsyncGenerator[tuple[str, Any], None]:
+    prepared_context = await _prepare_context(message, context)
+    text_chunks: list[str] = []
+    async for event in _run_single_agent_stream(
+        user_id=user_id,
+        system_prompt=_HOME_SINGLE_AGENT_SYSTEM,
+        message=message,
+        context=prepared_context,
+    ):
+        event_type, data = event
+        if event_type != "token":
+            yield event
+            continue
+        text_chunks.append(str(data or ""))
+
+    normalized = _normalize_tagged_list_lines("".join(text_chunks), message)
+    if normalized:
+        yield "token", normalized
+
+
+async def run_floating_stream(
+    user_id: str,
+    message: str,
+    context: dict[str, Any],
+) -> AsyncGenerator[tuple[str, Any], None]:
+    prepared_context = await _prepare_context(message, context)
+    domain = await _infer_floating_domain(message, prepared_context)
+    yield "floating_domain", domain
+
+    sanitizer = _FloatingStreamSanitizer()
+    emitted_sanitized = False
+    raw_chunks: list[str] = []
+    async for event in _run_single_agent_stream(
+        user_id=user_id,
+        system_prompt=_FLOATING_SINGLE_AGENT_SYSTEM,
+        message=message,
+        context=prepared_context,
+    ):
+        event_type, data = event
+        if event_type != "token":
+            yield event
+            continue
+
+        raw_chunk = str(data or "")
+        raw_chunks.append(raw_chunk)
+        sanitized_chunk = sanitizer.feed(raw_chunk)
+        if sanitized_chunk:
+            emitted_sanitized = True
+            yield "token", sanitized_chunk
+
+    tail = sanitizer.finalize()
+    if tail:
+        emitted_sanitized = True
+        yield "token", tail
+
+    if not emitted_sanitized and raw_chunks:
+        yield "token", _fallback_from_raw_floating_text("".join(raw_chunks))
+
+
+async def update_core_memory(user_id: str, key: str, value: str) -> None:
+    """Compatibility helper kept for callers that expect explicit memory update API."""
+    async with async_session() as db:
+        memory = MemoryMiddleware(db)
+        await memory.update_core(user_id, key, value)

app/core/device_manager.py

@@ -0,0 +1,151 @@
+"""Device connection manager.
+
+Maintains in-memory state for all active Electron → backend WebSocket
+connections.  One connection per user (latest replaces previous).
+
+The manager handles the **tool-call round-trip** pattern:
+  - Backend sends ``tool_call`` frame → Electron executes the action →
+    returns ``tool_result`` frame.
+  - ``create_pending_call`` registers a Future keyed by ``call_id``.
+  - ``resolve_pending_call`` fulfils the Future; callers awaiting it
+    receive the result dict from Electron.
+
+This pattern is used by all tools (CRUD, file-system, etc.) via
+``execute_on_client()`` in ``ws_context.py``.
+
+The ``device_manager`` module-level singleton is imported by both the
+device WS route and the agent runner.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import logging
+from dataclasses import dataclass, field
+
+from fastapi import WebSocket
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class DeviceConnection:
+    """State for a single connected Electron device."""
+
+    ws: WebSocket
+    device_id: str
+    # Futures indexed by tool_call id — resolved when tool_result arrives.
+    pending_calls: dict[str, asyncio.Future[dict]] = field(default_factory=dict)
+
+
+class DeviceConnectionManager:
+    """Singleton registry of active Electron WebSocket connections.
+
+    Thread/task safety note: asyncio is single-threaded by design.  All
+    mutations happen inside await-points on the main event loop, so no
+    locking is required for the in-memory dicts.
+    """
+
+    def __init__(self) -> None:
+        self._connections: dict[str, DeviceConnection] = {}
+
+    # ── Registration ──────────────────────────────────────────────────
+
+    def register(self, user_id: str, device_id: str, ws: WebSocket) -> None:
+        """Store the active connection for *user_id*, replacing any previous one."""
+        if user_id in self._connections:
+            old = self._connections[user_id]
+            logger.info(
+                "device_manager: replacing existing connection for user=%s device=%s",
+                user_id,
+                old.device_id,
+            )
+            # Cancel any futures that were waiting on the old connection.
+            for fut in old.pending_calls.values():
+                if not fut.done():
+                    fut.cancel()
+        self._connections[user_id] = DeviceConnection(ws=ws, device_id=device_id)
+        logger.info(
+            "device_manager: registered user=%s device=%s", user_id, device_id
+        )
+
+    def unregister(self, user_id: str) -> None:
+        """Remove the connection for *user_id* and cancel any pending futures."""
+        conn = self._connections.pop(user_id, None)
+        if conn is None:
+            return
+        for fut in conn.pending_calls.values():
+            if not fut.done():
+                fut.cancel()
+        logger.info("device_manager: unregistered user=%s", user_id)
+
+    # ── Presence queries ──────────────────────────────────────────────
+
+    def get_ws(self, user_id: str) -> WebSocket | None:
+        """Return the active WebSocket for *user_id*, or ``None`` if offline."""
+        conn = self._connections.get(user_id)
+        return conn.ws if conn else None
+
+    def is_online(self, user_id: str, device_id: str | None = None) -> bool:
+        """Return ``True`` if the user has an active connection.
+
+        If *device_id* is provided also checks that it matches the connected device.
+        """
+        conn = self._connections.get(user_id)
+        if conn is None:
+            return False
+        if device_id is not None:
+            return conn.device_id == device_id
+        return True
+
+    # ── Frame sending ─────────────────────────────────────────────────
+
+    async def send_frame(self, user_id: str, frame: dict) -> None:
+        """Send *frame* as a JSON text message to the device.
+
+        Raises ``RuntimeError`` if the user is not connected.
+        """
+        conn = self._connections.get(user_id)
+        if conn is None:
+            raise RuntimeError(
+                f"send_frame: user {user_id!r} is not connected"
+            )
+        await conn.ws.send_text(json.dumps(frame))
+
+    # ── Tool-call round-trip ──────────────────────────────────────────
+
+    def create_pending_call(
+        self, user_id: str, call_id: str
+    ) -> asyncio.Future[dict]:
+        """Register a Future that will be resolved when the tool_result arrives.
+
+        Raises ``RuntimeError`` if the user is not connected.
+        """
+        conn = self._connections.get(user_id)
+        if conn is None:
+            raise RuntimeError(
+                f"create_pending_call: user {user_id!r} is not connected"
+            )
+        loop = asyncio.get_event_loop()
+        fut: asyncio.Future[dict] = loop.create_future()
+        conn.pending_calls[call_id] = fut
+        return fut
+
+    def resolve_pending_call(
+        self, user_id: str, call_id: str, result: dict
+    ) -> None:
+        """Fulfil the Future registered under *call_id* with the Electron result.
+
+        No-ops if the call_id is unknown (already timed out or cancelled).
+        """
+        conn = self._connections.get(user_id)
+        if conn is None:
+            return
+        fut = conn.pending_calls.pop(call_id, None)
+        if fut is not None and not fut.done():
+            fut.set_result(result)
+
+
+# Module-level singleton — import this everywhere.
+device_manager = DeviceConnectionManager()

app/core/execution_plan.py

@@ -1,222 +0,0 @@
-"""Execution Plan generator — builder, template registry, and LRU plan cache."""
-
-from __future__ import annotations
-
-from collections import OrderedDict
-from typing import Any
-
-from app.schemas import ExecutionPlan, PlanStep
-
-
-# ── Prompt Template Registry ──────────────────────────────────────────
-
-
-class PromptTemplateRegistry:
-    """Server-side store mapping template IDs to prompt text.
-
-    Clients only ever receive template IDs (e.g. ``"tpl_task_agent_default"``).
-    The actual prompt text is resolved here on the server, keeping prompt IP
-    out of API responses.
-    """
-
-    def __init__(self) -> None:
-        self._templates: dict[str, str] = {}
-
-    def register(self, template_id: str, prompt_text: str) -> None:
-        self._templates[template_id] = prompt_text
-
-    def get(self, template_id: str) -> str:
-        """Resolve a template ID to its prompt text.
-
-        Raises ``KeyError`` if the template is not registered.
-        """
-        text = self._templates.get(template_id)
-        if text is None:
-            raise KeyError(f"Template not found: {template_id!r}")
-        return text
-
-    def has(self, template_id: str) -> bool:
-        return template_id in self._templates
-
-    def list_ids(self) -> list[str]:
-        """Return all registered template IDs (never the text)."""
-        return list(self._templates.keys())
-
-
-# ── Execution Plan Builder ────────────────────────────────────────────
-
-
-class ExecutionPlanBuilder:
-    """Fluent builder for ``ExecutionPlan`` objects.
-
-    Example::
-
-        plan = (
-            ExecutionPlanBuilder("task_agent")
-            .add_llm_step("tpl_task_agent_default", {"message": user_msg})
-            .add_data_step("create_record", data_from_step=0)
-            .build()
-        )
-    """
-
-    def __init__(self, agent: str) -> None:
-        self._agent = agent
-        self._steps: list[PlanStep] = []
-
-    # ── step adders ──────────────────────────────────────────────────
-
-    def add_step(
-        self, action: str, params: dict[str, Any] | None = None
-    ) -> ExecutionPlanBuilder:
-        """Append a generic action step with optional parameters."""
-        self._steps.append(PlanStep(action=action, variables=params))
-        return self
-
-    def add_llm_step(
-        self, template_id: str, variables: dict[str, Any] | None = None
-    ) -> ExecutionPlanBuilder:
-        """Append an LLM step referencing a server-side template by ID."""
-        self._steps.append(
-            PlanStep(action="llm", prompt_template=template_id, variables=variables)
-        )
-        return self
-
-    def add_data_step(self, action: str, data_from_step: int) -> ExecutionPlanBuilder:
-        """Append a step whose input comes from the output of an earlier step."""
-        self._steps.append(PlanStep(action=action, data_from_step=data_from_step))
-        return self
-
-    # ── build ────────────────────────────────────────────────────────
-
-    def build(self) -> ExecutionPlan:
-        """Validate step references and return the ``ExecutionPlan``.
-
-        Raises ``ValueError`` if any ``data_from_step`` references a
-        non-existent or future step index.
-        """
-        for i, step in enumerate(self._steps):
-            if step.data_from_step is not None:
-                if not (0 <= step.data_from_step < i):
-                    raise ValueError(
-                        f"Step {i}: data_from_step={step.data_from_step} must "
-                        f"reference a preceding step index in range 0..{i - 1}"
-                    )
-        return ExecutionPlan(agent=self._agent, steps=list(self._steps))
-
-
-# ── Plan Cache (LRU) ──────────────────────────────────────────────────
-
-
-class PlanCache:
-    """In-memory LRU cache for ``ExecutionPlan`` objects.
-
-    Plans stored here are accessible as playbooks via ``get_all_playbooks()``.
-    The cache also serves as a runtime memoisation layer so that repeated
-    identical intent classifications can skip re-building the plan.
-    """
-
-    def __init__(self, maxsize: int = 1000) -> None:
-        self._maxsize = maxsize
-        self._cache: OrderedDict[str, ExecutionPlan] = OrderedDict()
-
-    def cache_plan(self, key: str, plan: ExecutionPlan) -> None:
-        """Store *plan* under *key*, evicting the LRU entry if at capacity."""
-        if key in self._cache:
-            del self._cache[key]  # remove so re-insertion places it at the end
-        elif len(self._cache) >= self._maxsize:
-            self._cache.popitem(last=False)  # evict least-recently-used
-        self._cache[key] = plan
-
-    def get_plan(self, key: str) -> ExecutionPlan | None:
-        """Return the cached plan for *key*, or ``None`` if not present.
-
-        Accessing a plan marks it as most-recently used.
-        """
-        if key not in self._cache:
-            return None
-        self._cache.move_to_end(key)
-        return self._cache[key]
-
-    def get_all_playbooks(self) -> list[ExecutionPlan]:
-        """Return all cached plans (most-recently used last)."""
-        return list(self._cache.values())
-
-
-# ── Module-level singletons ───────────────────────────────────────────
-
-template_registry = PromptTemplateRegistry()
-plan_cache = PlanCache()
-
-
-def _register_builtin_templates() -> None:
-    """Register the built-in server-side prompt templates.
-
-    These strings never leave the server.  Clients only receive the IDs.
-    """
-    _tpls: dict[str, str] = {
-        "tpl_task_agent_default": (
-            "You are a task management assistant. Help the user create, update, "
-            "list, and track tasks. Use correct status values (todo, in_progress, "
-            "done) and priority values (high, medium, low) from the workspace model."
-        ),
-        "tpl_checkpoint_agent_default": (
-            "You are a project checkpoint assistant. Help the user create and manage "
-            "milestone checkpoints on their projects. Every checkpoint requires a "
-            "project_id and a date expressed as a Unix timestamp in milliseconds."
-        ),
-        "tpl_project_agent_default": (
-            "You are a project management assistant. Help the user create, find, "
-            "update, and archive projects. Projects have a name, an optional client, "
-            "and a status of either active or archived."
-        ),
-        "tpl_note_agent_default": (
-            "You are a note-taking assistant. Help the user create, retrieve, update, "
-            "and delete Markdown notes. Notes can optionally be linked to a project."
-        ),
-        "tpl_task_extract_from_project": (
-            "Extract all actionable tasks from the provided project context. "
-            "Return a structured list of tasks, each with a title, inferred priority "
-            "(high, medium, or low), suggested status (todo), and a due_date in "
-            "milliseconds where a deadline can be inferred."
-        ),
-        "tpl_note_weekly_summary": (
-            "Generate a weekly project summary note from the provided workspace data. "
-            "Include: tasks completed this week, tasks due soon, active projects, "
-            "and upcoming checkpoints. Format the output as clean Markdown."
-        ),
-    }
-    for tid, text in _tpls.items():
-        template_registry.register(tid, text)
-
-
-def _load_playbooks() -> None:
-    """Pre-build and cache the built-in playbooks."""
-    playbooks: list[tuple[str, ExecutionPlan]] = [
-        (
-            "create_tasks_from_project",
-            ExecutionPlanBuilder("project_agent")
-            .add_llm_step(
-                "tpl_task_extract_from_project",
-                {"source": "project_context"},
-            )
-            .add_data_step("create_record", data_from_step=0)
-            .build(),
-        ),
-        (
-            "generate_weekly_note",
-            ExecutionPlanBuilder("note_agent")
-            .add_llm_step(
-                "tpl_note_weekly_summary",
-                {"period": "last_7_days"},
-            )
-            .add_data_step("create_record", data_from_step=0)
-            .build(),
-        ),
-    ]
-    for key, plan in playbooks:
-        plan_cache.cache_plan(key, plan)
-
-
-# Initialise on module load
-_register_builtin_templates()
-_load_playbooks()

app/core/llm.py

@@ -1,6 +1,6 @@
 """LLM factory — centralised model instantiation via LiteLLM.
 
-Every agent and the orchestrator call ``get_llm()`` or ``get_router_llm()``
+Every agent and the orchestrator call ``get_llm()``
 instead of directly constructing a provider-specific class.  The model string
 follows the `LiteLLM model naming convention
 <https://docs.litellm.ai/docs/providers>`_:
@@ -11,17 +11,36 @@ follows the `LiteLLM model naming convention
 * Ollama:     ``ollama/llama3``
 * Bedrock:    ``bedrock/anthropic.claude-v2``
 
-Switch providers by changing **LLM_MODEL** / **LLM_ROUTER_MODEL** in ``.env``
+Switch providers by changing **LLM_MODEL** in ``.env``
 — no code changes required.
 """
 
 from __future__ import annotations
 
+import os
+import warnings
+
+from openai import AsyncOpenAI
+import litellm
+
 from langchain_openai import ChatOpenAI
+from langchain_litellm import ChatLiteLLM
 from litellm import get_supported_openai_params  # noqa: F401 – validates install
 
 from app.config.settings import settings
 
+# Some models (e.g. gpt-5, o-series) reject unsupported params like temperature.
+# Drop them silently instead of raising UnsupportedParamsError.
+litellm.drop_params = True
+
+# Some provider responses include a plain dict in the `usage` field where a
+# richer Pydantic model is expected. This warning is noisy but non-fatal.
+warnings.filterwarnings(
+    "ignore",
+    message=r"PydanticSerializationUnexpectedValue\(Expected `ResponseAPIUsage`",
+    category=UserWarning,
+)
+
 
 def _api_key_for_model(model: str) -> str | None:
     """Return the most appropriate API key for the given LiteLLM model string."""
@@ -29,6 +48,12 @@ def _api_key_for_model(model: str) -> str | None:
         return settings.ANTHROPIC_API_KEY or None
     if model.startswith("gemini/") or model.startswith("google/"):
         return settings.GOOGLE_API_KEY or None
+    if model.startswith("cerebras/"):
+        return settings.CEREBRAS_API_KEY or None
+    if model.startswith("github_copilot/"):
+        # GitHub Copilot uses OAuth device-flow tokens managed by LiteLLM.
+        # No API key is required; returning None lets LiteLLM handle auth.
+        return None
     # Default: OpenAI-compatible (covers plain model names like "gpt-4o")
     return settings.OPENAI_API_KEY or None
 
@@ -37,7 +62,7 @@ def get_llm(
     *,
     model: str | None = None,
     temperature: float = 0,
-) -> ChatOpenAI:
+) -> ChatOpenAI | ChatLiteLLM:
     """Return a LangChain chat model backed by LiteLLM.
 
     LiteLLM exposes an OpenAI-compatible API, so we use ``ChatOpenAI`` pointed
@@ -53,6 +78,16 @@ def get_llm(
         Sampling temperature.  ``0`` = deterministic.
     """
     model = model or settings.LLM_MODEL
+
+    # Point LiteLLM to the custom token directory when configured.
+    if settings.GITHUB_COPILOT_TOKEN_DIR:
+        os.environ.setdefault("GITHUB_COPILOT_TOKEN_DIR", settings.GITHUB_COPILOT_TOKEN_DIR)
+
+    # Use ChatLiteLLM for provider-prefixed models (github_copilot/, anthropic/, etc.)
+    # so LiteLLM handles routing and auth. ChatOpenAI for plain OpenAI model names.
+    if "/" in model:
+        return ChatLiteLLM(model=model, temperature=temperature)
+
     return ChatOpenAI(
         model=model,
         temperature=temperature,
@@ -60,9 +95,23 @@ def get_llm(
     )
 
 
-def get_router_llm(
+async def embed(text: str) -> list[float]:
-    *,
+    """Return an embedding vector for *text*.
-    temperature: float = 0,
+
-) -> ChatOpenAI:
+    Uses ``settings.LLM_EMBED_MODEL`` so the same provider switch in ``.env``
-    """Return the lighter model used for intent classification / routing."""
+    (e.g. ``github_copilot/text-embedding-3-small``) applies here without any
-    return get_llm(model=settings.LLM_ROUTER_MODEL, temperature=temperature)
+    code changes.  Falls back to the raw AsyncOpenAI client for plain OpenAI
+    model names to preserve existing behaviour.
+    """
+    model = settings.LLM_EMBED_MODEL
+
+    if model.startswith("github_copilot/") or "/" in model:
+        # Use LiteLLM for all provider-prefixed models (Copilot, Bedrock, etc.)
+        # so the provider's auth mechanism is applied correctly.
+        response = await litellm.aembedding(model=model, input=[text])
+        return response.data[0]["embedding"]
+
+    # Plain OpenAI model name — use the raw AsyncOpenAI client (existing path).
+    client = AsyncOpenAI(api_key=settings.OPENAI_API_KEY)
+    response = await client.embeddings.create(model=model, input=text)
+    return response.data[0].embedding

app/core/memory_middleware.py

@@ -0,0 +1,441 @@
+"""Memory Middleware — enrich requests with memory context and store interactions.
+
+Four-tier memory model (MemGPT-style):
+  core         — persistent key/value user preferences, always injected
+  associative  — semantic similarity search via pgvector (top-k)
+  episodic     — recent session summaries (last N)
+  proactive    — behavioral patterns above confidence threshold
+
+All memory content is encrypted at rest using the per-user Fernet key
+stored in User.encryption_key. Decryption happens in-memory only.
+
+Usage:
+    memory = MemoryMiddleware(db_session)
+    context = await memory.enrich_context(user_id, message)
+    # ... run agent ...
+    await memory.store_episode(user_id, session_id, message, response)
+"""
+
+from __future__ import annotations
+
+import logging
+import uuid
+from typing import Any
+
+from cryptography.fernet import Fernet, InvalidToken
+from sqlalchemy import select
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from app.models import (
+    MemoryAssociative,
+    MemoryCore,
+    MemoryEpisodic,
+    MemoryProactive,
+    User,
+)
+
+logger = logging.getLogger(__name__)
+
+# Tuning constants
+_ASSOCIATIVE_TOP_K = 5
+_EPISODIC_RECENT_N = 10
+_PROACTIVE_CONFIDENCE_THRESHOLD = 0.6
+
+
+class MemoryMiddleware:
+    """Enrich orchestrator context with memory and persist interactions after."""
+
+    def __init__(self, db: AsyncSession) -> None:
+        self._db = db
+
+    # ── Public API ────────────────────────────────────────────────────────────
+
+    async def enrich_context(
+        self,
+        user_id: str,
+        message: str,
+        trace_id: str | None = None,
+        session_id: str | None = None,
+    ) -> dict[str, Any]:
+        """Build memory context dict to inject into the orchestrator before LLM call.
+
+        Returns a dict with keys:
+          core_memory        — {key: plaintext_value, ...}
+          associative_memory — [plaintext_content, ...]  (top-k by keyword match)
+          episodic_memory    — [plaintext_summary, ...]  (most recent N)
+          proactive_hints    — [plaintext_pattern, ...]  (above threshold)
+        """
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return {}
+
+        core = await self._load_core(user_id, fernet)
+        associative = await self._load_associative(user_id, message, fernet)
+        episodic = await self._load_episodic(user_id, fernet, session_id=session_id)
+        proactive = await self._load_proactive(user_id, fernet)
+
+        user_dbg = await self._get_user_debug(user_id)
+        logger.info(
+            "memory: enrich_context trace=%s user=%s tier=%s core=%d associative=%d episodic=%d proactive=%d",
+            trace_id or "-",
+            user_id,
+            user_dbg.get("tier") or "-",
+            len(core),
+            len(associative),
+            len(episodic),
+            len(proactive),
+        )
+
+        return {
+            "core_memory": core,
+            "associative_memory": associative,
+            "episodic_memory": episodic,
+            "proactive_hints": proactive,
+        }
+
+    async def store_episode(
+        self,
+        user_id: str,
+        session_id: str,
+        message: str,
+        response: str,
+        trace_id: str | None = None,
+    ) -> None:
+        """Summarise and store a completed interaction in episodic memory.
+
+        The summary is a simple heuristic concatenation (no LLM call) to keep
+        latency low. Full LLM summarisation can be added in a later step.
+        """
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return
+
+        summary = f"User: {message[:200]}\nAssistant: {response[:200]}"
+        encrypted = _encrypt(fernet, summary)
+
+        row = MemoryEpisodic(
+            id=str(uuid.uuid4()),
+            user_id=user_id,
+            summary_encrypted=encrypted,
+            session_id=session_id,
+        )
+        self._db.add(row)
+        try:
+            await self._db.commit()
+            user_dbg = await self._get_user_debug(user_id)
+            logger.info(
+                "memory: store_episode trace=%s user=%s tier=%s session=%s",
+                trace_id or "-",
+                user_id,
+                user_dbg.get("tier") or "-",
+                session_id,
+            )
+        except Exception as exc:
+            logger.error("memory: store_episode failed user=%s: %s", user_id, exc)
+            await self._db.rollback()
+
+    async def update_core(self, user_id: str, key: str, value: str, trace_id: str | None = None) -> None:
+        """Upsert a core memory key/value for a user."""
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return
+
+        encrypted = _encrypt(fernet, value)
+
+        result = await self._db.execute(
+            select(MemoryCore).where(
+                MemoryCore.user_id == user_id,
+                MemoryCore.key == key,
+            )
+        )
+        existing = result.scalar_one_or_none()
+        if existing is not None:
+            existing.value_encrypted = encrypted
+        else:
+            self._db.add(MemoryCore(
+                id=str(uuid.uuid4()),
+                user_id=user_id,
+                key=key,
+                value_encrypted=encrypted,
+            ))
+        try:
+            await self._db.commit()
+            user_dbg = await self._get_user_debug(user_id)
+            logger.info(
+                "memory: update_core trace=%s user=%s tier=%s key=%s",
+                trace_id or "-",
+                user_id,
+                user_dbg.get("tier") or "-",
+                key,
+            )
+        except Exception as exc:
+            logger.error("memory: update_core failed user=%s key=%s: %s", user_id, key, exc)
+            await self._db.rollback()
+
+    async def list_core_blocks(self, user_id: str) -> list[dict[str, str]]:
+        """Return core memory as editable blocks (label/value)."""
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return []
+
+        result = await self._db.execute(
+            select(MemoryCore)
+            .where(MemoryCore.user_id == user_id)
+            .order_by(MemoryCore.key.asc())
+        )
+        rows = result.scalars().all()
+        out: list[dict[str, str]] = []
+        for row in rows:
+            plaintext = _safe_decrypt(fernet, row.value_encrypted)
+            if plaintext is not None:
+                out.append({"label": row.key, "value": plaintext})
+        logger.debug("memory: list_core_blocks user=%s count=%d", user_id, len(out))
+        return out
+
+    async def get_core_block(self, user_id: str, label: str) -> str | None:
+        """Return a single core memory block value by label."""
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return None
+
+        result = await self._db.execute(
+            select(MemoryCore).where(
+                MemoryCore.user_id == user_id,
+                MemoryCore.key == label,
+            )
+        )
+        row = result.scalar_one_or_none()
+        if row is None:
+            logger.debug("memory: get_core_block user=%s label=%s found=0", user_id, label)
+            return None
+        value = _safe_decrypt(fernet, row.value_encrypted)
+        logger.debug("memory: get_core_block user=%s label=%s found=%d", user_id, label, 1 if value is not None else 0)
+        return value
+
+    async def delete_core(self, user_id: str, label: str) -> bool:
+        """Delete a core memory block by label. Returns True if deleted."""
+        result = await self._db.execute(
+            select(MemoryCore).where(
+                MemoryCore.user_id == user_id,
+                MemoryCore.key == label,
+            )
+        )
+        row = result.scalar_one_or_none()
+        if row is None:
+            logger.debug("memory: delete_core user=%s label=%s found=0", user_id, label)
+            return False
+
+        await self._db.delete(row)
+        try:
+            await self._db.commit()
+            logger.info("memory: delete_core user=%s label=%s", user_id, label)
+            return True
+        except Exception as exc:
+            logger.error("memory: delete_core failed user=%s label=%s: %s", user_id, label, exc)
+            await self._db.rollback()
+            return False
+
+    async def append_core(self, user_id: str, label: str, content: str) -> None:
+        """Append content to a core block, creating it if missing."""
+        current = await self.get_core_block(user_id, label)
+        if current is None:
+            await self.update_core(user_id, label, content)
+            logger.info("memory: append_core user=%s label=%s created=1", user_id, label)
+            return
+        await self.update_core(user_id, label, f"{current}\n{content}")
+        logger.info("memory: append_core user=%s label=%s created=0", user_id, label)
+
+    async def replace_core(self, user_id: str, label: str, old: str, new: str) -> bool:
+        """Replace one exact string inside a core block. Returns False if not found."""
+        current = await self.get_core_block(user_id, label)
+        if current is None or old not in current:
+            logger.debug("memory: replace_core user=%s label=%s changed=0", user_id, label)
+            return False
+        await self.update_core(user_id, label, current.replace(old, new, 1))
+        logger.info("memory: replace_core user=%s label=%s changed=1", user_id, label)
+        return True
+
+    async def insert_archival(self, user_id: str, content: str, source: str = "manual") -> None:
+        """Insert a long-term archival memory entry."""
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return
+
+        encrypted = _encrypt(fernet, content)
+        row = MemoryAssociative(
+            id=str(uuid.uuid4()),
+            user_id=user_id,
+            content_encrypted=encrypted,
+            embedding=None,
+            entity_type=source,
+            entity_id=None,
+        )
+        self._db.add(row)
+        try:
+            await self._db.commit()
+            logger.info("memory: insert_archival user=%s source=%s", user_id, source)
+        except Exception as exc:
+            logger.error("memory: insert_archival failed user=%s: %s", user_id, exc)
+            await self._db.rollback()
+
+    async def search_archival(self, user_id: str, query: str, top_k: int = 5) -> list[str]:
+        """Search archival memory (keyword fallback; semantic ranking can replace this)."""
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return []
+
+        result = await self._db.execute(
+            select(MemoryAssociative)
+            .where(MemoryAssociative.user_id == user_id)
+            .order_by(MemoryAssociative.updated_at.desc())
+            .limit(100)
+        )
+        rows = result.scalars().all()
+        needle = query.strip().lower()
+        out: list[str] = []
+        for row in rows:
+            plaintext = _safe_decrypt(fernet, row.content_encrypted)
+            if plaintext is None:
+                continue
+            if not needle or needle in plaintext.lower():
+                out.append(plaintext)
+            if len(out) >= max(top_k, 1):
+                break
+        logger.info("memory: search_archival user=%s query=%s hits=%d", user_id, query[:80], len(out))
+        return out
+
+    async def search_recall(self, user_id: str, query: str, top_k: int = 5) -> list[str]:
+        """Search recall memory (episodic summaries) by keyword."""
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return []
+
+        result = await self._db.execute(
+            select(MemoryEpisodic)
+            .where(MemoryEpisodic.user_id == user_id)
+            .order_by(MemoryEpisodic.created_at.desc())
+            .limit(100)
+        )
+        rows = result.scalars().all()
+        needle = query.strip().lower()
+        out: list[str] = []
+        for row in rows:
+            plaintext = _safe_decrypt(fernet, row.summary_encrypted)
+            if plaintext is None:
+                continue
+            if not needle or needle in plaintext.lower():
+                out.append(plaintext)
+            if len(out) >= max(top_k, 1):
+                break
+        logger.info("memory: search_recall user=%s query=%s hits=%d", user_id, query[:80], len(out))
+        return out
+
+    # ── Private helpers ───────────────────────────────────────────────────────
+
+    async def _get_fernet(self, user_id: str) -> Fernet | None:
+        """Load the user's Fernet key from DB. Returns None if missing."""
+        result = await self._db.execute(select(User).where(User.id == user_id))
+        user = result.scalar_one_or_none()
+        if user is None or not user.encryption_key:
+            logger.warning("memory: no encryption_key for user=%s", user_id)
+            return None
+        return Fernet(user.encryption_key.encode())
+
+    async def _get_user_debug(self, user_id: str) -> dict[str, str | None]:
+        """Load lightweight user debug fields for trace logs."""
+        result = await self._db.execute(select(User).where(User.id == user_id))
+        user = result.scalar_one_or_none()
+        if user is None:
+            return {"tier": None}
+        return {
+            "tier": user.tier,
+        }
+
+    async def _load_core(self, user_id: str, fernet: Fernet) -> dict[str, str]:
+        result = await self._db.execute(
+            select(MemoryCore).where(MemoryCore.user_id == user_id)
+        )
+        rows = result.scalars().all()
+        out: dict[str, str] = {}
+        for row in rows:
+            plaintext = _safe_decrypt(fernet, row.value_encrypted)
+            if plaintext is not None:
+                out[row.key] = plaintext
+        return out
+
+    async def _load_associative(
+        self, user_id: str, message: str, fernet: Fernet
+    ) -> list[str]:
+        """Load top-k associative memories.
+
+        Production: uses pgvector cosine similarity on the message embedding.
+        Current implementation: keyword-based fallback (no external embedding call)
+        so tests pass without a live OpenAI key.
+        """
+        result = await self._db.execute(
+            select(MemoryAssociative)
+            .where(MemoryAssociative.user_id == user_id)
+            .order_by(MemoryAssociative.updated_at.desc())
+            .limit(_ASSOCIATIVE_TOP_K)
+        )
+        rows = result.scalars().all()
+        out: list[str] = []
+        for row in rows:
+            plaintext = _safe_decrypt(fernet, row.content_encrypted)
+            if plaintext is not None:
+                out.append(plaintext)
+        return out
+
+    async def _load_episodic(
+        self,
+        user_id: str,
+        fernet: Fernet,
+        session_id: str | None = None,
+    ) -> list[str]:
+        query = select(MemoryEpisodic).where(MemoryEpisodic.user_id == user_id)
+        if session_id:
+            query = query.where(MemoryEpisodic.session_id == session_id)
+        result = await self._db.execute(
+            query
+            .order_by(MemoryEpisodic.created_at.desc())
+            .limit(_EPISODIC_RECENT_N)
+        )
+        rows = result.scalars().all()
+        out: list[str] = []
+        for row in rows:
+            plaintext = _safe_decrypt(fernet, row.summary_encrypted)
+            if plaintext is not None:
+                out.append(plaintext)
+        return out
+
+    async def _load_proactive(self, user_id: str, fernet: Fernet) -> list[str]:
+        result = await self._db.execute(
+            select(MemoryProactive)
+            .where(
+                MemoryProactive.user_id == user_id,
+                MemoryProactive.confidence >= _PROACTIVE_CONFIDENCE_THRESHOLD,
+            )
+            .order_by(MemoryProactive.confidence.desc())
+        )
+        rows = result.scalars().all()
+        out: list[str] = []
+        for row in rows:
+            plaintext = _safe_decrypt(fernet, row.pattern_encrypted)
+            if plaintext is not None:
+                out.append(plaintext)
+        return out
+
+
+# ── Encryption helpers ────────────────────────────────────────────────────────
+
+def _encrypt(fernet: Fernet, plaintext: str) -> str:
+    return fernet.encrypt(plaintext.encode()).decode()
+
+
+def _safe_decrypt(fernet: Fernet, ciphertext: str) -> str | None:
+    """Decrypt and return plaintext, or None on error (corrupted/wrong key)."""
+    try:
+        return fernet.decrypt(ciphertext.encode()).decode()
+    except (InvalidToken, Exception) as exc:
+        logger.warning("memory: decrypt failed: %s", exc)
+        return None

app/core/orchestrator.py

@@ -1,168 +0,0 @@
-"""Orchestrator — LLM-based intent router and agent pipeline."""
-
-from __future__ import annotations
-
-import json
-from typing import Any, AsyncGenerator
-
-from langchain_core.messages import HumanMessage, SystemMessage
-
-from app.core.agent_registry import AgentRegistry
-from app.core.llm import get_router_llm
-from app.core.agent_registry import registry as _default_registry
-from app.schemas import ChatRequest, ChatResponse, ExecutionPlan
-
-_FALLBACK_AGENT = "task_agent"
-
-_CLASSIFY_SYSTEM = (
-    "You are an intent classifier. Given the user message and context, decide "
-    "which agent to route to.\n"
-    "Available agents: {agents}\n"
-    "Respond with just the agent name, nothing else."
-)
-
-_SYNTHESIZE_HUMAN = (
-    "Combine the following agent results into one coherent response.\n\n"
-    "Agent results:\n{results}\n\n"
-    "Original message: {message}"
-)
-
-
-def _make_llm():
-    return get_router_llm()
-
-
-async def classify_intent(
-    message: str,
-    context: dict[str, Any],
-    reg: AgentRegistry,
-) -> str:
-    """Use gpt-4o-mini to classify intent and return the matching agent name.
-
-    Falls back to ``task_agent`` when the registry is empty or the model
-    returns a name that is not registered.
-    """
-    agents = reg.list_agents()
-    if not agents:
-        return _FALLBACK_AGENT
-
-    system = _CLASSIFY_SYSTEM.format(agents=json.dumps(agents))
-    # Truncate context to keep the classification prompt short
-    human = f"Message: {message}\nContext summary: {json.dumps(context)[:500]}"
-
-    llm = _make_llm()
-    response = await llm.ainvoke(
-        [SystemMessage(content=system), HumanMessage(content=human)]
-    )
-
-    agent_name = str(response.content).strip().lower()
-    known = {a["name"] for a in agents}
-    return agent_name if agent_name in known else _FALLBACK_AGENT
-
-
-async def route_single(
-    agent_name: str,
-    message: str,
-    context: dict[str, Any],
-    reg: AgentRegistry,
-) -> ChatResponse:
-    """Route to a single agent and wrap the result in a ``ChatResponse``."""
-    response_text = await reg.call_agent(agent_name, message, context)
-    return ChatResponse(response=response_text)
-
-
-async def route_pipeline(
-    agent_names: list[str],
-    message: str,
-    context: dict[str, Any],
-    reg: AgentRegistry,
-) -> ChatResponse:
-    """Execute agents sequentially; each agent receives previous results in context.
-
-    A final LLM synthesis call merges all results into one coherent response.
-    """
-    previous_results: list[str] = []
-
-    for agent_name in agent_names:
-        ctx = {**context, "previous_results": list(previous_results)}
-        result = await reg.call_agent(agent_name, message, ctx)
-        previous_results.append(result)
-
-    results_str = "\n\n".join(
-        f"[{name}]: {res}" for name, res in zip(agent_names, previous_results)
-    )
-    human = _SYNTHESIZE_HUMAN.format(results=results_str, message=message)
-    llm = _make_llm()
-    synthesis = await llm.ainvoke([HumanMessage(content=human)])
-    return ChatResponse(response=str(synthesis.content))
-
-
-def _build_plan(agent_name: str, message: str) -> ExecutionPlan:
-    """Build an ``ExecutionPlan`` for the resolved agent.
-
-    Uses ``ExecutionPlanBuilder`` with the server-side template registry.
-    If a default template exists for the agent, an LLM step is emitted;
-    otherwise a plain ``handle`` action step is used.
-    """
-    from app.core.execution_plan import ExecutionPlanBuilder, template_registry
-
-    template_id = f"tpl_{agent_name}_default"
-    builder = ExecutionPlanBuilder(agent_name)
-    if template_registry.has(template_id):
-        builder.add_llm_step(template_id, {"message": message})
-    else:
-        builder.add_step("handle", {"message": message})
-    return builder.build()
-
-
-async def orchestrate(
-    request: ChatRequest,
-    reg: AgentRegistry | None = None,
-) -> ChatResponse | ExecutionPlan:
-    """Main orchestration entry point.
-
-    * Classifies the user's intent to select an agent.
-    * ``execution_mode == 'direct'``: routes to the agent and returns a
-      ``ChatResponse``.
-    * ``execution_mode == 'plan'``: returns an ``ExecutionPlan`` with the
-      resolved agent and a template-ID-only step (prompt IP stays server-side).
-    """
-    if reg is None:
-        reg = _default_registry
-
-    context = request.context.model_dump()
-    agent_name = await classify_intent(request.message, context, reg)
-
-    if request.execution_mode == "direct":
-        return await route_single(agent_name, request.message, context, reg)
-
-    # plan mode — return plan, do not execute
-    return _build_plan(agent_name, request.message)
-
-
-async def orchestrate_stream(
-    request: ChatRequest,
-    reg: AgentRegistry | None = None,
-) -> AsyncGenerator[str, None]:
-    """Streaming orchestration — yields text chunks then a final JSON frame.
-
-    The final frame is a JSON object:
-    ``{"done": true, "response": "...", "actions": []}``.
-
-    Agents do not yet support token-level streaming; the full response is
-    fetched first, then emitted in fixed-size chunks.  Token-level streaming
-    will be wired in Step 6 when agents expose ``astream()``.
-    """
-    if reg is None:
-        reg = _default_registry
-
-    context = request.context.model_dump()
-    agent_name = await classify_intent(request.message, context, reg)
-    response_text = await reg.call_agent(agent_name, request.message, context)
-
-    chunk_size = 50
-    for i in range(0, len(response_text), chunk_size):
-        yield response_text[i : i + chunk_size]
-
-    final = ChatResponse(response=response_text)
-    yield json.dumps({"done": True, **final.model_dump()})

app/core/output_formatter.py

@@ -0,0 +1,47 @@
+"""Output formatter for deep-agent stream events."""
+
+from __future__ import annotations
+
+from collections.abc import AsyncGenerator
+from typing import Any
+
+from app.schemas import WsFloatingDomain, WsStreamEnd, WsStreamStart, WsStreamText
+
+WsFrame = WsStreamStart | WsStreamText | WsStreamEnd | WsFloatingDomain
+
+
+class StreamFormatter:
+    """Convert `(event_type, data)` stream events into websocket frame models."""
+
+    def __init__(self, request_id: str) -> None:
+        self.request_id = request_id
+
+    async def format(
+        self,
+        event_stream: AsyncGenerator[tuple[str, Any], None],
+    ) -> AsyncGenerator[WsFrame, None]:
+        started = False
+
+        async for event_type, data in event_stream:
+            if event_type == "floating_domain":
+                if isinstance(data, dict):
+                    yield WsFloatingDomain(
+                        request_id=self.request_id,
+                        domain=data,
+                    )
+                continue
+
+            if event_type != "token":
+                continue
+
+            if not started:
+                yield WsStreamStart(request_id=self.request_id)
+                started = True
+
+            text = str(data or "")
+            if text:
+                yield WsStreamText(request_id=self.request_id, chunk=text)
+
+        if not started:
+            yield WsStreamStart(request_id=self.request_id)
+        yield WsStreamEnd(request_id=self.request_id)

app/core/ws_context.py

@@ -0,0 +1,92 @@
+"""WebSocket client executor context.
+
+Holds a per-request async callback that tools call to execute CRUD
+operations on the Electron client's local SQLite / LanceDB databases.
+The callback sends a `tool_call` WS frame and awaits the `tool_result`.
+"""
+
+from __future__ import annotations
+
+from contextvars import ContextVar
+from typing import Any, Callable, Coroutine
+from uuid import uuid4
+
+# Holds the execute callback for the current WS session.
+# Set by the chat WS handler before the orchestrator runs; cleared after.
+_client_executor: ContextVar[Callable[[dict], Coroutine[Any, Any, dict]]] = ContextVar(
+    "_client_executor"
+)
+
+# Optional collector that captures raw execute_on_client results.
+# Set by _tool_loop / _tool_loop_stream to populate ChatAgent.tool_results.
+_tool_result_collector: ContextVar[list[dict] | None] = ContextVar(
+    "_tool_result_collector", default=None
+)
+
+
+def set_tool_result_collector(lst: list[dict]) -> None:
+    """Register *lst* as the collector for this async context."""
+    _tool_result_collector.set(lst)
+
+
+def clear_tool_result_collector() -> None:
+    """Clear the collector (best-effort)."""
+    _tool_result_collector.set(None)
+
+
+def set_client_executor(fn: Callable[[dict], Coroutine[Any, Any, dict]]) -> None:
+    """Bind *fn* as the executor for the current async context (task/coroutine)."""
+    _client_executor.set(fn)
+
+
+def clear_client_executor() -> None:
+    """Remove the executor binding (best-effort; ContextVar resets on task exit)."""
+    try:
+        _client_executor.set(None)  # type: ignore[arg-type]
+    except Exception:
+        pass
+
+
+async def execute_on_client(
+    action: str,
+    table: str | None = None,
+    data: dict[str, Any] | None = None,
+    filters: dict[str, Any] | None = None,
+    vector: list[float] | None = None,
+    limit: int | None = None,
+) -> dict[str, Any]:
+    """Send a CRUD/vector operation to the Electron client and return the result.
+
+    Builds a ``tool_call`` payload, invokes the per-session WS callback,
+    and returns the ``tool_result`` dict from Electron.
+
+    Raises ``RuntimeError`` if no executor is set (i.e. called outside a WS session).
+    """
+    callback = _client_executor.get(None)
+    if callback is None:
+        raise RuntimeError(
+            "execute_on_client() called outside a WebSocket session — "
+            "no client executor is set."
+        )
+
+    payload: dict[str, Any] = {"id": str(uuid4()), "action": action}
+    if table is not None:
+        payload["table"] = table
+    if data is not None:
+        payload["data"] = data
+    if filters is not None:
+        payload["filters"] = {k: v for k, v in filters.items() if v is not None}
+    if vector is not None:
+        payload["vector"] = vector
+    if limit is not None:
+        payload["limit"] = limit
+
+    result = await callback(payload)
+    collector = _tool_result_collector.get(None)
+    if collector is not None:
+        collector.append({
+            "action": action,
+            "table": table,
+            "data": result,
+        })
+    return result

app/db.py

@@ -24,7 +24,7 @@ from app.config.settings import settings
 engine = create_async_engine(
     settings.DATABASE_URL,
     pool_pre_ping=True,
-    echo=settings.ENV == "dev",
+    echo=False,
 )
 
 async_session = async_sessionmaker(engine, expire_on_commit=False)

app/integrations/__init__.py

@@ -0,0 +1,164 @@
+"""Cloud provider integration utilities.
+
+Provides:
+  * Shared message dataclasses (``EmailMessage``, ``ChatMessage``) used by
+    both the Gmail and MS Graph clients and consumed by ``agent_runner``.
+  * ``get_provider()`` — factory that returns the correct client given a
+    provider name and decrypted OAuth credentials dict.
+  * ``encrypt_token()`` / ``decrypt_token()`` — Fernet-based at-rest
+    encryption for OAuth tokens stored in ``cloud_agent_configs``.
+
+Encryption rationale
+--------------------
+Unlike user content (which is E2E-encrypted client-side and **never**
+decrypted server-side), OAuth tokens *must* be decrypted server-side
+because the backend makes provider API calls on behalf of the user.
+The Fernet key lives solely in ``OAUTH_ENCRYPTION_KEY`` env var — it
+is never returned to clients.
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+from dataclasses import dataclass, field
+from datetime import datetime
+from typing import TYPE_CHECKING
+
+from cryptography.fernet import Fernet, InvalidToken
+
+from app.config.settings import settings
+
+if TYPE_CHECKING:
+    from app.integrations.gmail import GmailClient
+    from app.integrations.ms_graph import MSGraphClient
+
+logger = logging.getLogger(__name__)
+
+# ── Shared message types ──────────────────────────────────────────────────
+
+
+@dataclass
+class EmailMessage:
+    """A single email message fetched from Gmail or Outlook."""
+
+    id: str
+    subject: str
+    sender: str
+    body_text: str
+    date: datetime
+    labels: list[str] = field(default_factory=list)
+
+    @property
+    def as_text(self) -> str:
+        """Return a human-readable text representation for LLM extraction."""
+        date_str = self.date.strftime("%Y-%m-%d %H:%M")
+        labels_str = f" [{', '.join(self.labels)}]" if self.labels else ""
+        return (
+            f"From: {self.sender}\n"
+            f"Date: {date_str}{labels_str}\n"
+            f"Subject: {self.subject}\n\n"
+            f"{self.body_text}"
+        )
+
+
+@dataclass
+class ChatMessage:
+    """A single Teams chat or channel message fetched from MS Graph."""
+
+    id: str
+    content: str
+    sender: str
+    channel: str | None
+    date: datetime
+
+    @property
+    def as_text(self) -> str:
+        """Return a human-readable text representation for LLM extraction."""
+        date_str = self.date.strftime("%Y-%m-%d %H:%M")
+        channel_str = f" [channel: {self.channel}]" if self.channel else ""
+        return (
+            f"From: {self.sender}\n"
+            f"Date: {date_str}{channel_str}\n\n"
+            f"{self.content}"
+        )
+
+
+# ── Fernet helpers ────────────────────────────────────────────────────────
+
+
+def _get_fernet() -> Fernet:
+    """Return a ``Fernet`` instance using ``settings.OAUTH_ENCRYPTION_KEY``.
+
+    Raises ``RuntimeError`` if ``OAUTH_ENCRYPTION_KEY`` is not set — callers
+    must ensure this is configured before persisting OAuth tokens.
+    """
+    key = settings.OAUTH_ENCRYPTION_KEY
+    if not key:
+        raise RuntimeError(
+            "OAUTH_ENCRYPTION_KEY is not set. "
+            "Generate one with: python -c \"from cryptography.fernet import Fernet; print(Fernet.generate_key().decode())\""
+        )
+    return Fernet(key.encode() if isinstance(key, str) else key)
+
+
+def encrypt_token(token_info: dict) -> str:
+    """Fernet-encrypt an OAuth credential dict and return a base64 string.
+
+    Stores the full ``{access_token, refresh_token, token_uri, client_id,
+    client_secret, scopes, expiry}`` dict (or equivalent MSAL shape).
+
+    Raises:
+        RuntimeError: OAUTH_ENCRYPTION_KEY is not configured.
+        ValueError: ``token_info`` is not a non-empty dict.
+    """
+    if not isinstance(token_info, dict) or not token_info:
+        raise ValueError("token_info must be a non-empty dict")
+    plaintext = json.dumps(token_info).encode("utf-8")
+    return _get_fernet().encrypt(plaintext).decode("utf-8")
+
+
+def decrypt_token(encrypted: str) -> dict:
+    """Decrypt a Fernet-encrypted token string and return the credential dict.
+
+    Raises:
+        RuntimeError: OAUTH_ENCRYPTION_KEY is not configured.
+        ValueError: The encrypted string is invalid or was encrypted with a
+            different key.
+    """
+    try:
+        plaintext = _get_fernet().decrypt(encrypted.encode("utf-8"))
+        return json.loads(plaintext)
+    except (InvalidToken, json.JSONDecodeError) as exc:
+        raise ValueError(f"Failed to decrypt OAuth token: {exc}") from exc
+
+
+# ── Provider factory ──────────────────────────────────────────────────────
+
+
+def get_provider(
+    provider: str,
+    credentials_info: dict,
+) -> "GmailClient | MSGraphClient":
+    """Return the correct provider client for *provider*.
+
+    Parameters
+    ----------
+    provider:
+        One of ``"gmail"``, ``"outlook"``, ``"teams"``.
+    credentials_info:
+        Decrypted OAuth credential dict (Google or Microsoft shape).
+
+    Raises:
+        ValueError: Unknown provider name.
+    """
+    if provider == "gmail":
+        from app.integrations.gmail import GmailClient
+        return GmailClient(credentials_info)
+    if provider in {"outlook", "teams"}:
+        from app.integrations.ms_graph import MSGraphClient
+        return MSGraphClient(credentials_info)
+    raise ValueError(
+        f"Unknown cloud provider {provider!r}. "
+        "Supported: 'gmail', 'outlook', 'teams'."
+    )

app/integrations/gmail.py

@@ -0,0 +1,335 @@
+"""Gmail API client for cloud agent integration.
+
+Wraps the Google Gmail REST API to fetch email messages matching a
+``filter_config`` dict.  Uses the official ``google-api-python-client``
+library (synchronous) wrapped in ``asyncio.to_thread()`` to avoid
+blocking the event loop.
+
+Token refresh is handled transparently: when the stored access token has
+expired, ``google.auth.transport.requests.Request`` will use the refresh
+token to obtain a fresh one.  The caller is responsible for persisting
+any refreshed credentials back to ``CloudAgentConfig.oauth_token_encrypted``
+(see ``agent_runner.run_cloud_agent``).
+
+Credential dict shape (Google OAuth2):
+    {
+        "token": "<access_token>",
+        "refresh_token": "<refresh_token>",
+        "token_uri": "https://oauth2.googleapis.com/token",
+        "client_id": "<client_id>",
+        "client_secret": "<client_secret>",
+        "scopes": ["https://www.googleapis.com/auth/gmail.readonly"],
+        "expiry": "2025-01-01T00:00:00Z"  # optional ISO-8601
+    }
+"""
+
+from __future__ import annotations
+
+import asyncio
+import base64
+import email
+import html
+import logging
+import re
+from datetime import datetime, timezone
+from typing import Any
+
+from app.integrations import EmailMessage
+
+logger = logging.getLogger(__name__)
+
+# Gmail search date format — e.g. "after:2025/01/01"
+_GMAIL_DATE_FMT = "%Y/%m/%d"
+
+# Maximum characters of body text forwarded to the LLM.
+_BODY_TRUNCATE = 8_000
+
+# Maximum messages retrieved per run (prevents runaway quota usage).
+_MAX_MESSAGES = 200
+
+
+def _build_gmail_query(
+    filter_config: dict[str, Any] | None,
+    since: datetime | None,
+) -> str:
+    """Build a Gmail search query string from *filter_config* and *since*.
+
+    Supported ``filter_config`` keys:
+        labels (list[str]):  Gmail label names, e.g. ``["INBOX", "work"]``
+        senders (list[str]): Sender addresses or domains to include
+        date_range (dict):   ``{from: "<YYYY-MM-DD>", to: "<YYYY-MM-DD>"}``
+
+    A hard ``since`` date (from last run) always overrides ``date_range.from``
+    when it is earlier.
+    """
+    parts: list[str] = []
+    cfg = filter_config or {}
+
+    # Labels — joined with OR when multiple given.
+    labels: list[str] = cfg.get("labels", [])
+    if labels:
+        if len(labels) == 1:
+            parts.append(f"label:{labels[0]}")
+        else:
+            label_expr = " OR ".join(f"label:{lbl}" for lbl in labels)
+            parts.append(f"({label_expr})")
+
+    # Senders — each prefixed with "from:".
+    senders: list[str] = cfg.get("senders", [])
+    for sender in senders:
+        parts.append(f"from:{sender}")
+
+    # Date range.
+    date_range: dict = cfg.get("date_range", {})
+    from_str: str | None = date_range.get("from")
+    to_str: str | None = date_range.get("to")
+
+    # Determine effective "from" date: most recent of filter_config.date_range.from and since.
+    effective_since: datetime | None = since
+    if from_str:
+        try:
+            cfg_since = datetime.fromisoformat(from_str.replace("Z", "+00:00"))
+            if cfg_since.tzinfo is None:
+                cfg_since = cfg_since.replace(tzinfo=timezone.utc)
+            if effective_since is None or cfg_since > effective_since:
+                effective_since = cfg_since
+        except ValueError:
+            logger.warning("gmail: invalid date_range.from %r — ignoring", from_str)
+
+    if effective_since:
+        parts.append(f"after:{effective_since.strftime(_GMAIL_DATE_FMT)}")
+
+    if to_str:
+        try:
+            to_dt = datetime.fromisoformat(to_str.replace("Z", "+00:00"))
+            parts.append(f"before:{to_dt.strftime(_GMAIL_DATE_FMT)}")
+        except ValueError:
+            logger.warning("gmail: invalid date_range.to %r — ignoring", to_str)
+
+    return " ".join(parts)
+
+
+def _strip_html(raw_html: str) -> str:
+    """Remove HTML tags and decode entities to get plain text."""
+    no_tags = re.sub(r"<[^>]+>", " ", raw_html)
+    decoded = html.unescape(no_tags)
+    return re.sub(r"\s+", " ", decoded).strip()
+
+
+def _parse_body(payload: dict[str, Any]) -> str:
+    """Recursively extract the plain-text body from a Gmail message payload.
+
+    Prefers ``text/plain``; falls back to ``text/html`` (stripped of tags).
+    Returns an empty string if no body can be extracted.
+    """
+    mime_type: str = payload.get("mimeType", "")
+    body: dict = payload.get("body", {})
+    parts: list[dict] = payload.get("parts", [])
+
+    if mime_type == "text/plain":
+        data = body.get("data", "")
+        if data:
+            return base64.urlsafe_b64decode(data + "==").decode("utf-8", errors="replace")
+        return ""
+
+    if mime_type == "text/html":
+        data = body.get("data", "")
+        if data:
+            raw = base64.urlsafe_b64decode(data + "==").decode("utf-8", errors="replace")
+            return _strip_html(raw)
+        return ""
+
+    # Multipart — prefer text/plain part, fall back to text/html.
+    plain_fallback = ""
+    for part in parts:
+        part_mime = part.get("mimeType", "")
+        if part_mime == "text/plain":
+            return _parse_body(part)
+        if part_mime == "text/html" and not plain_fallback:
+            plain_fallback = _parse_body(part)
+        if part_mime.startswith("multipart/"):
+            nested = _parse_body(part)
+            if nested:
+                return nested
+    return plain_fallback
+
+
+def _parse_date(raw: str) -> datetime:
+    """Parse an RFC 2822 email date header into a UTC ``datetime``."""
+    try:
+        parsed = email.utils.parsedate_to_datetime(raw)
+        if parsed.tzinfo is None:
+            parsed = parsed.replace(tzinfo=timezone.utc)
+        return parsed.astimezone(timezone.utc)
+    except Exception:
+        return datetime.now(timezone.utc)
+
+
+class GmailClient:
+    """Fetch email messages from a Gmail account via the Gmail REST API.
+
+    Parameters
+    ----------
+    credentials_info:
+        Decrypted OAuth2 credential dict.  Must contain at minimum
+        ``token`` (access token) or ``refresh_token`` + ``token_uri`` +
+        ``client_id`` + ``client_secret``.
+    """
+
+    def __init__(self, credentials_info: dict[str, Any]) -> None:
+        from google.oauth2.credentials import Credentials
+
+        self._credentials_info = credentials_info
+        expiry_str: str | None = credentials_info.get("expiry")
+        expiry: datetime | None = None
+        if expiry_str:
+            try:
+                expiry = datetime.fromisoformat(
+                    expiry_str.replace("Z", "+00:00")
+                ).replace(tzinfo=timezone.utc)
+            except ValueError:
+                pass
+
+        self._credentials = Credentials(
+            token=credentials_info.get("token"),
+            refresh_token=credentials_info.get("refresh_token"),
+            token_uri=credentials_info.get("token_uri", "https://oauth2.googleapis.com/token"),
+            client_id=credentials_info.get("client_id"),
+            client_secret=credentials_info.get("client_secret"),
+            scopes=credentials_info.get("scopes"),
+            expiry=expiry,
+        )
+
+    # ── Public API ─────────────────────────────────────────────────────────
+
+    async def fetch_messages(
+        self,
+        filter_config: dict[str, Any] | None = None,
+        since: datetime | None = None,
+    ) -> list[EmailMessage]:
+        """Return up to ``_MAX_MESSAGES`` emails matching *filter_config*.
+
+        Runs the synchronous Google API calls inside ``asyncio.to_thread()``
+        to avoid blocking the async event loop.
+
+        Token refresh is performed automatically when the access token has
+        expired.  After the call, ``self.refreshed_credentials`` may be
+        consulted to detect whether new credentials should be persisted.
+        """
+        query = _build_gmail_query(filter_config, since)
+        logger.debug("gmail: executing search query %r", query)
+        return await asyncio.to_thread(self._fetch_sync, query)
+
+    @property
+    def refreshed_credentials(self) -> dict[str, Any] | None:
+        """Return updated credential dict if the access token was refreshed.
+
+        If the credentials were refreshed during ``fetch_messages()``, returns
+        a new dict that should be re-encrypted and written back to the DB.
+        Returns ``None`` if no refresh occurred.
+        """
+        creds = self._credentials
+        if not creds.valid and creds.expired:
+            return None
+        # Check whether the token changed from what was stored.
+        if creds.token != self._credentials_info.get("token"):
+            result = {
+                "token": creds.token,
+                "refresh_token": creds.refresh_token,
+                "token_uri": creds.token_uri,
+                "client_id": creds.client_id,
+                "client_secret": creds.client_secret,
+                "scopes": list(creds.scopes or []),
+            }
+            if creds.expiry:
+                result["expiry"] = creds.expiry.isoformat()
+            return result
+        return None
+
+    # ── Internal sync worker ───────────────────────────────────────────────
+
+    def _fetch_sync(self, query: str) -> list[EmailMessage]:
+        """Synchronous worker — called inside ``asyncio.to_thread()``."""
+        import googleapiclient.discovery
+        import googleapiclient.errors
+        from google.auth.transport.requests import Request
+
+        # Refresh token if needed before building the service.
+        if self._credentials.expired and self._credentials.refresh_token:
+            try:
+                self._credentials.refresh(Request())
+            except Exception as exc:
+                raise RuntimeError(f"Gmail token refresh failed: {exc}") from exc
+
+        service = googleapiclient.discovery.build(
+            "gmail", "v1", credentials=self._credentials, cache_discovery=False
+        )
+        user_api = service.users()  # type: ignore[attr-defined]
+
+        # ── List matching message IDs ──────────────────────────────────────
+        ids: list[str] = []
+        page_token: str | None = None
+        while len(ids) < _MAX_MESSAGES:
+            batch_size = min(100, _MAX_MESSAGES - len(ids))
+            kwargs: dict[str, Any] = {
+                "userId": "me",
+                "maxResults": batch_size,
+            }
+            if query:
+                kwargs["q"] = query
+            if page_token:
+                kwargs["pageToken"] = page_token
+
+            try:
+                resp = user_api.messages().list(**kwargs).execute()
+            except googleapiclient.errors.HttpError as exc:
+                raise RuntimeError(f"Gmail messages.list failed: {exc}") from exc
+
+            for msg in resp.get("messages", []):
+                ids.append(msg["id"])
+
+            page_token = resp.get("nextPageToken")
+            if not page_token:
+                break
+
+        if not ids:
+            logger.debug("gmail: no messages matched query %r", query)
+            return []
+
+        logger.info("gmail: fetching %d message(s)", len(ids))
+
+        # ── Fetch individual message details ──────────────────────────────
+        messages: list[EmailMessage] = []
+        for msg_id in ids:
+            try:
+                msg = user_api.messages().get(
+                    userId="me", id=msg_id, format="full"
+                ).execute()
+
+                headers: dict[str, str] = {
+                    h["name"].lower(): h["value"]
+                    for h in msg.get("payload", {}).get("headers", [])
+                }
+                subject = headers.get("subject", "(no subject)")
+                sender = headers.get("from", "unknown")
+                date_raw = headers.get("date", "")
+                date = _parse_date(date_raw) if date_raw else datetime.now(timezone.utc)
+
+                body_text = _parse_body(msg.get("payload", {}))[:_BODY_TRUNCATE]
+                labels = msg.get("labelIds", [])
+
+                messages.append(EmailMessage(
+                    id=msg_id,
+                    subject=subject,
+                    sender=sender,
+                    body_text=body_text,
+                    date=date,
+                    labels=labels,
+                ))
+            except googleapiclient.errors.HttpError as exc:
+                logger.warning("gmail: skipping message %s — HTTP error: %s", msg_id, exc)
+            except Exception as exc:
+                logger.warning("gmail: skipping message %s — unexpected error: %s", msg_id, exc)
+
+        logger.info("gmail: returned %d message(s)", len(messages))
+        return messages

app/integrations/ms_graph.py

@@ -0,0 +1,352 @@
+"""Microsoft Graph API client for Outlook and Teams cloud agent integration.
+
+Handles two data sources:
+
+* **Outlook email** (``provider="outlook"``) — ``fetch_emails()`` calls
+  ``/me/messages`` with an OData ``$filter`` built from ``filter_config``.
+* **Teams messages** (``provider="teams"``) — ``fetch_messages()`` calls
+  ``/me/chats/getAllMessages`` filtered by date.
+
+Authentication uses MSAL ``PublicClientApplication`` to acquire a token
+from a stored refresh token.  The ``httpx.AsyncClient`` (already a project
+dependency) is used for all API calls.
+
+Credential dict shape (Microsoft OAuth2 / MSAL):
+    {
+        "access_token":  "<access_token>",
+        "refresh_token": "<refresh_token>",
+        "token_type":    "Bearer",
+        "scope":         "Mail.Read ChannelMessage.Read.All offline_access",
+        "expires_in":    3600
+    }
+"""
+
+from __future__ import annotations
+
+import logging
+import re
+from datetime import datetime, timedelta, timezone
+from typing import Any
+
+import httpx
+
+from app.config.settings import settings
+from app.integrations import ChatMessage, EmailMessage
+
+logger = logging.getLogger(__name__)
+
+_GRAPH_BASE = "https://graph.microsoft.com/v1.0"
+
+# Max items fetched per run.
+_MAX_EMAILS = 200
+_MAX_MESSAGES = 200
+
+# Max characters of body forwarded to the LLM.
+_BODY_TRUNCATE = 8_000
+
+
+def _strip_html(raw: str) -> str:
+    """Strip HTML tags and collapse whitespace."""
+    no_tags = re.sub(r"<[^>]+>", " ", raw)
+    import html as _html
+    decoded = _html.unescape(no_tags)
+    return re.sub(r"\s+", " ", decoded).strip()
+
+
+def _odata_datetime(dt: datetime) -> str:
+    """Format a datetime as an OData datetime literal (UTC, ISO 8601)."""
+    utc = dt.astimezone(timezone.utc)
+    return utc.strftime("%Y-%m-%dT%H:%M:%SZ")
+
+
+def _build_email_filter(
+    filter_config: dict[str, Any] | None,
+    since: datetime | None,
+) -> str:
+    """Build an OData ``$filter`` expression for the ``/me/messages`` endpoint.
+
+    Supported ``filter_config`` keys:
+        senders (list[str]):  Sender email addresses.
+        date_range (dict):    ``{from: "<ISO-8601>", to: "<ISO-8601>"}``
+        folders (list[str]):  Folder display names (not directly filterable
+                              via OData, so ignored here — callers iterate
+                              folder IDs separately if needed; listed for
+                              completeness).
+
+    A hard ``since`` date always overrides ``date_range.from`` when it is
+    earlier.
+    """
+    clauses: list[str] = []
+    cfg = filter_config or {}
+
+    # Senders.
+    senders: list[str] = cfg.get("senders", [])
+    if senders:
+        sender_clauses = [f"from/emailAddress/address eq '{s}'" for s in senders]
+        clauses.append("(" + " or ".join(sender_clauses) + ")")
+
+    # Date range.
+    date_range: dict = cfg.get("date_range", {})
+    from_str: str | None = date_range.get("from")
+
+    effective_since: datetime | None = since
+    if from_str:
+        try:
+            cfg_since = datetime.fromisoformat(from_str.replace("Z", "+00:00"))
+            if cfg_since.tzinfo is None:
+                cfg_since = cfg_since.replace(tzinfo=timezone.utc)
+            if effective_since is None or cfg_since > effective_since:
+                effective_since = cfg_since
+        except ValueError:
+            logger.warning("ms_graph: invalid date_range.from %r — ignoring", from_str)
+
+    if effective_since:
+        clauses.append(f"receivedDateTime ge {_odata_datetime(effective_since)}")
+
+    to_str: str | None = date_range.get("to")
+    if to_str:
+        try:
+            to_dt = datetime.fromisoformat(to_str.replace("Z", "+00:00"))
+            if to_dt.tzinfo is None:
+                to_dt = to_dt.replace(tzinfo=timezone.utc)
+            clauses.append(f"receivedDateTime le {_odata_datetime(to_dt)}")
+        except ValueError:
+            logger.warning("ms_graph: invalid date_range.to %r — ignoring", to_str)
+
+    return " and ".join(clauses)
+
+
+class MSGraphClient:
+    """Fetch emails and Teams messages via the Microsoft Graph REST API.
+
+    Parameters
+    ----------
+    credentials_info:
+        Decrypted MSAL credential dict.
+    """
+
+    def __init__(self, credentials_info: dict[str, Any]) -> None:
+        self._credentials_info = credentials_info
+        self._access_token: str = credentials_info.get("access_token", "")
+        self._original_access_token: str = self._access_token
+        self._refresh_token: str | None = credentials_info.get("refresh_token")
+
+    # ── Token management ───────────────────────────────────────────────────
+
+    def _auth_headers(self) -> dict[str, str]:
+        return {"Authorization": f"Bearer {self._access_token}"}
+
+    async def _refresh_access_token(self) -> None:
+        """Use MSAL to exchange the refresh token for a fresh access token.
+
+        Updates ``self._access_token`` and ``self._credentials_info`` in-place.
+
+        Raises:
+            RuntimeError: MSAL reports an auth error.
+        """
+        import msal
+
+        app = msal.ConfidentialClientApplication(
+            client_id=settings.MS_CLIENT_ID,
+            client_credential=settings.MS_CLIENT_SECRET,
+            authority=f"https://login.microsoftonline.com/{settings.MS_TENANT_ID}",
+        )
+        scopes: list[str] = self._credentials_info.get("scope", "").split()
+        if not scopes:
+            scopes = ["https://graph.microsoft.com/.default"]
+
+        result = app.acquire_token_by_refresh_token(
+            self._refresh_token,
+            scopes=scopes,
+        )
+        if "access_token" not in result:
+            error = result.get("error_description", result.get("error", "unknown"))
+            raise RuntimeError(f"MS Graph token refresh failed: {error}")
+
+        self._access_token = result["access_token"]
+        # MSAL may issue a new refresh token.
+        if "refresh_token" in result:
+            self._refresh_token = result["refresh_token"]
+            self._credentials_info["refresh_token"] = result["refresh_token"]
+        self._credentials_info["access_token"] = self._access_token
+
+    @property
+    def refreshed_credentials(self) -> dict[str, Any] | None:
+        """Return updated credential dict if the access token was refreshed.
+
+        Returns ``None`` if no change was made.
+        """
+        if self._access_token != self._original_access_token:
+            return {**self._credentials_info, "access_token": self._access_token}
+        return None
+
+    # ── HTTP helpers ───────────────────────────────────────────────────────
+
+    async def _get(
+        self,
+        client: httpx.AsyncClient,
+        url: str,
+        params: dict[str, Any] | None = None,
+        *,
+        retry_on_401: bool = True,
+    ) -> dict[str, Any]:
+        """GET *url* with auth; refresh token on 401 and retry once."""
+        resp = await client.get(url, params=params, headers=self._auth_headers())
+        if resp.status_code == 401 and retry_on_401 and self._refresh_token:
+            logger.debug("ms_graph: 401 on %s — refreshing token", url)
+            await self._refresh_access_token()
+            resp = await client.get(url, params=params, headers=self._auth_headers())
+        if resp.status_code == 429:
+            raise RuntimeError("MS Graph rate limit hit (429). Try again later.")
+        resp.raise_for_status()
+        return resp.json()
+
+    # ── Public API ─────────────────────────────────────────────────────────
+
+    async def fetch_emails(
+        self,
+        filter_config: dict[str, Any] | None = None,
+        since: datetime | None = None,
+    ) -> list[EmailMessage]:
+        """Return up to ``_MAX_EMAILS`` Outlook messages matching *filter_config*.
+
+        Parameters
+        ----------
+        filter_config:
+            Optional dict with ``senders``, ``date_range``, ``folders`` keys.
+        since:
+            Hard lower-bound on email date (from last agent run).
+        """
+        odata_filter = _build_email_filter(filter_config, since)
+        params: dict[str, Any] = {
+            "$top": 50,
+            "$select": "id,subject,from,receivedDateTime,body,bodyPreview",
+            "$orderby": "receivedDateTime desc",
+        }
+        if odata_filter:
+            params["$filter"] = odata_filter
+
+        emails: list[EmailMessage] = []
+        url = f"{_GRAPH_BASE}/me/messages"
+
+        async with httpx.AsyncClient(timeout=30.0) as client:
+            while url and len(emails) < _MAX_EMAILS:
+                data = await self._get(client, url, params if url.startswith(_GRAPH_BASE) else None)
+                for item in data.get("value", []):
+                    emails.append(self._parse_email(item))
+                    if len(emails) >= _MAX_EMAILS:
+                        break
+                url = data.get("@odata.nextLink", "")
+                params = {}  # nextLink already contains encoded params.
+
+        logger.info("ms_graph: fetched %d Outlook email(s)", len(emails))
+        return emails
+
+    async def fetch_messages(
+        self,
+        filter_config: dict[str, Any] | None = None,
+        since: datetime | None = None,
+    ) -> list[ChatMessage]:
+        """Return up to ``_MAX_MESSAGES`` Teams messages matching *filter_config*.
+
+        Fetches from ``/me/chats/getAllMessages`` (personal + group chats).
+        The ``filter_config.channels`` key is checked as a text-filter on
+        the channel name post-fetch (the API doesn't support channel OData
+        filter directly on ``getAllMessages``).
+        """
+        cfg = filter_config or {}
+        channel_filter: list[str] = [c.lower() for c in cfg.get("channels", [])]
+        params: dict[str, Any] = {"$top": 50}
+        if since:
+            params["$filter"] = f"createdDateTime ge {_odata_datetime(since)}"
+
+        messages: list[ChatMessage] = []
+        url = f"{_GRAPH_BASE}/me/chats/getAllMessages"
+
+        async with httpx.AsyncClient(timeout=30.0) as client:
+            while url and len(messages) < _MAX_MESSAGES:
+                try:
+                    data = await self._get(client, url, params if url.startswith(_GRAPH_BASE) else None)
+                except httpx.HTTPStatusError as exc:
+                    # getAllMessages requires specific licensing; degrade gracefully.
+                    if exc.response.status_code in (403, 404):
+                        logger.warning(
+                            "ms_graph: /me/chats/getAllMessages not available (%d) — "
+                            "check Teams license or permissions",
+                            exc.response.status_code,
+                        )
+                        break
+                    raise
+
+                for item in data.get("value", []):
+                    msg = self._parse_teams_message(item)
+                    if channel_filter and msg.channel:
+                        if not any(c in msg.channel.lower() for c in channel_filter):
+                            continue
+                    messages.append(msg)
+                    if len(messages) >= _MAX_MESSAGES:
+                        break
+                url = data.get("@odata.nextLink", "")
+                params = {}
+
+        logger.info("ms_graph: fetched %d Teams message(s)", len(messages))
+        return messages
+
+    # ── Parsers ────────────────────────────────────────────────────────────
+
+    @staticmethod
+    def _parse_email(item: dict[str, Any]) -> EmailMessage:
+        subject: str = item.get("subject", "(no subject)") or "(no subject)"
+        sender_block = item.get("from", {}) or {}
+        sender_addr = (
+            (sender_block.get("emailAddress") or {}).get("address", "unknown")
+        )
+        date_str: str = item.get("receivedDateTime", "")
+        try:
+            date = datetime.fromisoformat(date_str.replace("Z", "+00:00"))
+        except Exception:
+            date = datetime.now(timezone.utc)
+
+        body_block = item.get("body", {}) or {}
+        content_type: str = body_block.get("contentType", "text")
+        raw_body: str = body_block.get("content", "")
+        if content_type == "html":
+            body_text = _strip_html(raw_body)
+        else:
+            body_text = raw_body or item.get("bodyPreview", "")
+        body_text = body_text[:_BODY_TRUNCATE]
+
+        return EmailMessage(
+            id=item.get("id", ""),
+            subject=subject,
+            sender=sender_addr,
+            body_text=body_text,
+            date=date,
+        )
+
+    @staticmethod
+    def _parse_teams_message(item: dict[str, Any]) -> ChatMessage:
+        msg_id: str = item.get("id", "")
+        sender_block = (item.get("from") or {}).get("user") or {}
+        sender: str = sender_block.get("displayName", "unknown")
+        channel: str | None = (item.get("channelIdentity") or {}).get("channelId")
+
+        date_str: str = item.get("createdDateTime", "")
+        try:
+            date = datetime.fromisoformat(date_str.replace("Z", "+00:00"))
+        except Exception:
+            date = datetime.now(timezone.utc)
+
+        body_block = item.get("body", {}) or {}
+        content_type: str = body_block.get("contentType", "text")
+        raw_content: str = body_block.get("content", "")
+        content = _strip_html(raw_content) if content_type == "html" else raw_content
+        content = content[:_BODY_TRUNCATE]
+
+        return ChatMessage(
+            id=msg_id,
+            content=content,
+            sender=sender,
+            channel=channel,
+            date=date,
+        )

app/main.py

@@ -1,8 +1,16 @@
 from contextlib import asynccontextmanager
+import logging
 
 from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
 
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s %(levelname)s %(name)s: %(message)s",
+)
+logging.getLogger("sqlalchemy.engine").setLevel(logging.WARNING)
+logging.getLogger("sqlalchemy.pool").setLevel(logging.WARNING)
+
 from app.api.middleware.rate_limit import TierRateLimitMiddleware
 from app.api.middleware.sanitizer import SanitizerMiddleware
 from app.config.settings import settings
@@ -10,9 +18,8 @@ from app.config.settings import settings
 
 @asynccontextmanager
 async def lifespan(app: FastAPI):
-    # Startup: initialise DB connection pool and agent registry
+    # Startup: ensure agent tool modules are loaded.
-    from app.core.agent_registry import registry  # noqa: F401 — triggers module load
+    import app.agents  # noqa: F401
-    import app.agents  # noqa: F401 — triggers @registry.register decorators
 
     yield
 
@@ -43,16 +50,17 @@ def create_app() -> FastAPI:
     app.add_middleware(SanitizerMiddleware)
     app.add_middleware(TierRateLimitMiddleware)
 
-    from app.api.routes import auth, backup, billing, chat, plans, plugins, storage, vectors
+    from app.api.routes import agents, auth, backup, billing, chat, device_ws, plugins, storage, vectors
 
     app.include_router(auth.router,       prefix="/api/v1")
     app.include_router(chat.router,       prefix="/api/v1")
-    app.include_router(plans.router,    prefix="/api/v1")
     app.include_router(storage.router,    prefix="/api/v1")
     app.include_router(vectors.router,    prefix="/api/v1")
     app.include_router(backup.router,     prefix="/api/v1")
     app.include_router(plugins.router,    prefix="/api/v1")
     app.include_router(billing.router,    prefix="/api/v1")
+    app.include_router(agents.router,     prefix="/api/v1")
+    app.include_router(device_ws.router,  prefix="/api/v1")
 
     @app.get("/api/v1/health", tags=["health"])
     async def health() -> dict:

app/marketplace/plugin_review.py

@@ -29,8 +29,8 @@ ALLOWED_PERMISSIONS: frozenset[str] = frozenset(
         "write:projects",
         "read:notes",
         "write:notes",
-        "read:checkpoints",
+        "read:timelines",
-        "write:checkpoints",
+        "write:timelines",
         "read:calendar",
         "write:calendar",
     }

app/models.py

@@ -14,6 +14,10 @@ Table inventory:
   plugin_installations — per-user install records
   plugin_reviews      — admin review decisions
   revenue_events      — Stripe Connect 70/30 split ledger
+  memory_core         — per-user persistent key/value preferences (encrypted)
+  memory_associative  — per-user semantic memory with embeddings (encrypted)
+  memory_episodic     — per-user session summaries (encrypted)
+  memory_proactive    — per-user behavioral patterns (encrypted)
 """
 
 from __future__ import annotations
@@ -23,11 +27,13 @@ from datetime import datetime, timezone
 
 from sqlalchemy import (
     BigInteger,
+    Boolean,
     DateTime,
     Enum,
     Float,
     ForeignKey,
     Integer,
+    JSON,
     String,
     Text,
     UniqueConstraint,
@@ -54,6 +60,9 @@ def _now() -> datetime:
 TierEnum = Enum("free", "pro", "power", "team", name="billing_tier")
 PluginStatusEnum = Enum("pending_review", "approved", "rejected", name="plugin_status")
 ReviewDecisionEnum = Enum("approved", "rejected", name="review_decision")
+AgentTypeEnum = Enum("local", "cloud", name="agent_type")
+AgentStatusEnum = Enum("running", "success", "error", "partial", name="agent_run_status")
+CloudProviderEnum = Enum("gmail", "teams", "outlook", name="cloud_provider")
 
 
 # ── Models ────────────────────────────────────────────────────────────────
@@ -66,9 +75,14 @@ class User(Base):
         Uuid(as_uuid=False), primary_key=True, default=_uuid
     )
     email: Mapped[str] = mapped_column(String(255), unique=True, nullable=False, index=True)
+    name: Mapped[str | None] = mapped_column(String(100), nullable=True)
+    surname: Mapped[str | None] = mapped_column(String(100), nullable=True)
     password_hash: Mapped[str] = mapped_column(String(255), nullable=False)
     tier: Mapped[str] = mapped_column(TierEnum, nullable=False, default="free")
     stripe_customer_id: Mapped[str | None] = mapped_column(String(255), nullable=True)
+    # Per-user Fernet key (base64-urlsafe, 44 chars). Generated on registration.
+    # Used to encrypt/decrypt all memory rows for this user.
+    encryption_key: Mapped[str | None] = mapped_column(String(64), nullable=True)
     created_at: Mapped[datetime] = mapped_column(
         DateTime(timezone=True), nullable=False, server_default=func.now()
     )
@@ -266,3 +280,197 @@ class RevenueEvent(Base):
     )
 
     plugin: Mapped[Plugin] = relationship(back_populates="revenue_events")
+
+
+class LocalAgentConfig(Base):
+    __tablename__ = "local_agent_configs"
+
+    id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), primary_key=True, default=_uuid
+    )
+    user_id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), ForeignKey("users.id", ondelete="CASCADE"), nullable=False, index=True
+    )
+    device_id: Mapped[str] = mapped_column(String(255), nullable=False)
+    name: Mapped[str] = mapped_column(String(255), nullable=False)
+    directory_paths: Mapped[list] = mapped_column(JSON, nullable=False, default=list)
+    data_types: Mapped[list] = mapped_column(JSON, nullable=False, default=list)
+    prompt_template: Mapped[str] = mapped_column(Text, nullable=False, default="")
+    file_extensions: Mapped[list] = mapped_column(JSON, nullable=False, default=list)
+    schedule_cron: Mapped[str] = mapped_column(String(100), nullable=False, default="0 */6 * * *")
+    enabled: Mapped[bool] = mapped_column(Boolean, nullable=False, default=True)
+    last_run_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)
+    created_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, server_default=func.now()
+    )
+    updated_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), onupdate=func.now()
+    )
+
+    run_logs: Mapped[list[AgentRunLog]] = relationship(
+        back_populates="local_agent",
+        primaryjoin="and_(AgentRunLog.agent_id == LocalAgentConfig.id, AgentRunLog.agent_type == 'local')",
+        foreign_keys="AgentRunLog.agent_id",
+        cascade="all, delete-orphan",
+        overlaps="run_logs,cloud_agent",
+    )
+
+
+class CloudAgentConfig(Base):
+    __tablename__ = "cloud_agent_configs"
+
+    id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), primary_key=True, default=_uuid
+    )
+    user_id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), ForeignKey("users.id", ondelete="CASCADE"), nullable=False, index=True
+    )
+    provider: Mapped[str] = mapped_column(CloudProviderEnum, nullable=False)
+    name: Mapped[str] = mapped_column(String(255), nullable=False)
+    data_types: Mapped[list] = mapped_column(JSON, nullable=False, default=list)
+    prompt_template: Mapped[str] = mapped_column(Text, nullable=False, default="")
+    oauth_token_encrypted: Mapped[str | None] = mapped_column(Text, nullable=True)
+    filter_config: Mapped[dict | None] = mapped_column(JSON, nullable=True)
+    schedule_cron: Mapped[str] = mapped_column(String(100), nullable=False, default="0 */6 * * *")
+    enabled: Mapped[bool] = mapped_column(Boolean, nullable=False, default=True)
+    last_run_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)
+    created_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, server_default=func.now()
+    )
+    updated_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), onupdate=func.now()
+    )
+
+    run_logs: Mapped[list[AgentRunLog]] = relationship(
+        back_populates="cloud_agent",
+        primaryjoin="and_(AgentRunLog.agent_id == CloudAgentConfig.id, AgentRunLog.agent_type == 'cloud')",
+        foreign_keys="AgentRunLog.agent_id",
+        cascade="all, delete-orphan",
+        overlaps="run_logs,local_agent",
+    )
+
+
+class AgentRunLog(Base):
+    __tablename__ = "agent_run_logs"
+
+    id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), primary_key=True, default=_uuid
+    )
+    # Plain string — not a FK because it references either local_agent_configs or cloud_agent_configs
+    # depending on agent_type. Query by (agent_id, agent_type) to locate the source config.
+    agent_id: Mapped[str] = mapped_column(String(255), nullable=False, index=True)
+    agent_type: Mapped[str] = mapped_column(AgentTypeEnum, nullable=False)
+    user_id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), ForeignKey("users.id", ondelete="CASCADE"), nullable=False, index=True
+    )
+    status: Mapped[str] = mapped_column(AgentStatusEnum, nullable=False, default="running")
+    items_processed: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
+    items_created: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
+    errors: Mapped[list | None] = mapped_column(JSON, nullable=True)
+    started_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, server_default=func.now()
+    )
+    completed_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)
+
+    local_agent: Mapped[LocalAgentConfig | None] = relationship(
+        back_populates="run_logs",
+        primaryjoin="and_(AgentRunLog.agent_id == LocalAgentConfig.id, AgentRunLog.agent_type == 'local')",
+        foreign_keys="AgentRunLog.agent_id",
+        overlaps="run_logs,cloud_agent",
+    )
+    cloud_agent: Mapped[CloudAgentConfig | None] = relationship(
+        back_populates="run_logs",
+        primaryjoin="and_(AgentRunLog.agent_id == CloudAgentConfig.id, AgentRunLog.agent_type == 'cloud')",
+        foreign_keys="AgentRunLog.agent_id",
+        overlaps="run_logs,local_agent",
+    )
+
+
+# ── Memory models ─────────────────────────────────────────────────────────────
+
+
+class MemoryCore(Base):
+    """Per-user persistent key/value preferences, encrypted at rest.
+
+    Examples: preferred_language, timezone, work_style.
+    Decrypted in-memory only using User.encryption_key.
+    """
+
+    __tablename__ = "memory_core"
+
+    id: Mapped[str] = mapped_column(Uuid(as_uuid=False), primary_key=True, default=_uuid)
+    user_id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), ForeignKey("users.id", ondelete="CASCADE"),
+        nullable=False, index=True,
+    )
+    key: Mapped[str] = mapped_column(String(255), nullable=False)
+    value_encrypted: Mapped[str] = mapped_column(Text, nullable=False)
+    updated_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), onupdate=func.now()
+    )
+
+
+class MemoryAssociative(Base):
+    """Per-user semantic memory: encrypted content + pgvector embedding for similarity search.
+
+    Production: ``embedding`` column is ``vector(1536)`` via pgvector.
+    Tests (SQLite): stored as JSON list.
+    """
+
+    __tablename__ = "memory_associative"
+
+    id: Mapped[str] = mapped_column(Uuid(as_uuid=False), primary_key=True, default=_uuid)
+    user_id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), ForeignKey("users.id", ondelete="CASCADE"),
+        nullable=False, index=True,
+    )
+    content_encrypted: Mapped[str] = mapped_column(Text, nullable=False)
+    # JSON-encoded float list in SQLite tests; vector(1536) in Postgres via migration.
+    embedding: Mapped[list | None] = mapped_column(JSON, nullable=True)
+    entity_type: Mapped[str | None] = mapped_column(String(100), nullable=True)
+    entity_id: Mapped[str | None] = mapped_column(String(255), nullable=True)
+    updated_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, server_default=func.now(), onupdate=func.now()
+    )
+
+
+class MemoryEpisodic(Base):
+    """Per-user session summaries, encrypted at rest.
+
+    One row per session interaction; used to recall recent conversations.
+    """
+
+    __tablename__ = "memory_episodic"
+
+    id: Mapped[str] = mapped_column(Uuid(as_uuid=False), primary_key=True, default=_uuid)
+    user_id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), ForeignKey("users.id", ondelete="CASCADE"),
+        nullable=False, index=True,
+    )
+    summary_encrypted: Mapped[str] = mapped_column(Text, nullable=False)
+    session_id: Mapped[str] = mapped_column(String(255), nullable=False, index=True)
+    created_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, server_default=func.now()
+    )
+
+
+class MemoryProactive(Base):
+    """Per-user inferred behavioral patterns, encrypted at rest.
+
+    Confidence in [0.0, 1.0]; only patterns above threshold are injected.
+    Source: 'inferred' (from episodes) or 'explicit' (user-stated).
+    """
+
+    __tablename__ = "memory_proactive"
+
+    id: Mapped[str] = mapped_column(Uuid(as_uuid=False), primary_key=True, default=_uuid)
+    user_id: Mapped[str] = mapped_column(
+        Uuid(as_uuid=False), ForeignKey("users.id", ondelete="CASCADE"),
+        nullable=False, index=True,
+    )
+    pattern_encrypted: Mapped[str] = mapped_column(Text, nullable=False)
+    confidence: Mapped[float] = mapped_column(Float, nullable=False, default=0.5)
+    source: Mapped[str] = mapped_column(String(50), nullable=False, default="inferred")
+    created_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, server_default=func.now()
+    )

app/schemas.py

@@ -5,6 +5,7 @@ Mirrors the TypeScript types from the Electron app (src/shared/api-types.ts).
 
 from __future__ import annotations
 
+from enum import Enum
 from typing import Any, Literal
 
 from pydantic import BaseModel, Field
@@ -26,6 +27,8 @@ class AuthTokens(BaseModel):
 class UserProfile(BaseModel):
     id: str
     email: str
+    name: str | None = None
+    surname: str | None = None
     tier: BillingTier
 
 
@@ -38,41 +41,13 @@ class ChatContext(BaseModel):
     conversation_history: list[dict[str, Any]] = Field(default_factory=list)
 
 
-class PlanAction(BaseModel):
-    type: Literal[
-        "create_record",
-        "update_record",
-        "delete_record",
-        "index_document",
-        "send_notification",
-    ]
-    table: str | None = None
-    data: dict[str, Any] | None = None
-
-
 class ChatRequest(BaseModel):
     message: str
     context: ChatContext = Field(default_factory=ChatContext)
-    execution_mode: Literal["direct", "plan"] = "direct"
 
 
 class ChatResponse(BaseModel):
     response: str
-    actions: list[PlanAction] = Field(default_factory=list)
-
-
-# ── Execution Plans ──────────────────────────────────────────────────
-
-class PlanStep(BaseModel):
-    action: str
-    prompt_template: str | None = None
-    variables: dict[str, Any] | None = None
-    data_from_step: int | None = None
-
-
-class ExecutionPlan(BaseModel):
-    agent: str
-    steps: list[PlanStep] = Field(default_factory=list)
 
 
 # ── Backup ───────────────────────────────────────────────────────────
@@ -155,3 +130,192 @@ class PluginListResponse(BaseModel):
 
 class PluginInstallRequest(BaseModel):
     plugin_id: str
+
+
+# ── WebSocket Frame Protocol ──────────────────────────────────────────
+
+class WsFrameType(str, Enum):
+    # ── v2 frame types (kept for backward compat) ──────────────────────
+    chat_request = "chat_request"
+    text_chunk = "text_chunk"
+    tool_call = "tool_call"
+    tool_result = "tool_result"
+    final = "final"
+    ping = "ping"
+    device_hello = "device_hello"
+    # ── v3 frame types ─────────────────────────────────────────────────
+    home_request = "home_request"
+    floating_request = "floating_request"
+    stream_start = "stream_start"
+    stream_text = "stream_text"
+    stream_end = "stream_end"
+    floating_domain = "floating_domain"
+    data_request = "data_request"
+    data_response = "data_response"
+    mutation = "mutation"
+    # ── v4 journey frame types ────────────────────────────────────────
+    journey_start = "journey_start"
+    journey_message = "journey_message"
+    journey_reply = "journey_reply"
+
+
+class WsToolCall(BaseModel):
+    """Server → Client: requests a CRUD/vector operation on the local DB."""
+
+    type: Literal[WsFrameType.tool_call] = WsFrameType.tool_call
+    id: str
+    action: str
+    table: str | None = None
+    data: dict[str, Any] | None = None
+    filters: dict[str, Any] | None = None
+    vector: list[float] | None = None
+    limit: int | None = None
+
+
+class WsToolResult(BaseModel):
+    """Client → Server: result of a CRUD/vector operation."""
+
+    type: Literal[WsFrameType.tool_result] = WsFrameType.tool_result
+    id: str
+    row: dict[str, Any] | None = None
+    rows: list[dict[str, Any]] | None = None
+    results: list[dict[str, Any]] | None = None
+    deleted: bool | None = None
+    ok: bool | None = None
+    error: str | None = None
+
+
+class WsTextChunk(BaseModel):
+    """Server → Client: incremental LLM response text."""
+
+    type: Literal[WsFrameType.text_chunk] = WsFrameType.text_chunk
+    text: str
+
+
+class WsFinal(BaseModel):
+    """Server → Client: signals end of response with the complete text."""
+
+    type: Literal[WsFrameType.final] = WsFrameType.final
+    response: str
+
+
+# ── WebSocket Agent Frame Protocol ────────────────────────────────────
+
+class WsDeviceHello(BaseModel):
+    """Client → Server: device identification on WS connect."""
+
+    type: Literal[WsFrameType.device_hello] = WsFrameType.device_hello
+    device_id: str
+    agent_ids: list[str] = Field(default_factory=list)
+
+
+
+# ── WebSocket v3 Frame Models ─────────────────────────────────────────
+
+class WsFloatingScope(BaseModel):
+    """Scope for a floating request — narrows the agent to a specific entity."""
+
+    type: Literal["task", "project", "note", "timeline"]
+    id: str | None = None
+
+
+class WsHomeRequest(BaseModel):
+    """Client → Server: Home chat message."""
+
+    type: Literal[WsFrameType.home_request] = WsFrameType.home_request
+    message: str
+    conversation_history: list[dict[str, Any]] = Field(default_factory=list)
+
+
+class WsFloatingRequest(BaseModel):
+    """Client → Server: Floating chat message scoped to an entity."""
+
+    type: Literal[WsFrameType.floating_request] = WsFrameType.floating_request
+    message: str
+    scope: WsFloatingScope
+
+
+class WsStreamStart(BaseModel):
+    """Server → Client: signals start of a streaming response."""
+
+    type: Literal[WsFrameType.stream_start] = WsFrameType.stream_start
+    request_id: str
+
+
+class WsStreamText(BaseModel):
+    """Server → Client: streamed text token."""
+
+    type: Literal[WsFrameType.stream_text] = WsFrameType.stream_text
+    request_id: str
+    chunk: str
+
+
+class WsStreamEnd(BaseModel):
+    """Server → Client: signals end of a streaming response."""
+
+    type: Literal[WsFrameType.stream_end] = WsFrameType.stream_end
+    request_id: str
+
+
+class WsDomain(BaseModel):
+    """Structured floating domain payload for UI routing decisions."""
+
+    type: Literal["task", "timeline", "project", "node"]
+    id: str | None = None
+    section: Literal["task", "timeline", "note"] | None = None
+
+
+class WsFloatingDomain(BaseModel):
+    """Server → Client: domain determined for a floating request."""
+
+    type: Literal[WsFrameType.floating_domain] = WsFrameType.floating_domain
+    request_id: str
+    domain: WsDomain
+
+
+# ── Agent Catalog ─────────────────────────────────────────────────────
+
+class AgentCatalogItem(BaseModel):
+    type: str
+    name: str
+    description: str
+
+
+class AgentCreationCheckRequest(BaseModel):
+    active_agents: int = Field(ge=0, default=0)
+
+
+class AgentCreationCheckResponse(BaseModel):
+    allowed: bool
+    tier: BillingTier
+    active_agents: int
+    limit: int
+
+
+class AgentTriggerRequest(BaseModel):
+    directory: str = Field(min_length=1)
+    device_id: str = Field(default="")
+    agent_id: str | None = None  # FE stable agent ID (electron-store UUID)
+    what_to_extract: list[str] = Field(min_length=1)
+    actions_by_type: dict[str, list[str]] | None = None
+    batch_interval: str = Field(min_length=1)
+    custom_agent_prompt: str = Field(min_length=1)
+    active_agents: int = Field(ge=0, default=0)
+
+
+# ── Agent Run Log ─────────────────────────────────────────────────────
+
+class AgentRunLogResponse(BaseModel):
+    id: str
+    agent_id: str
+    agent_type: Literal["local", "cloud"]
+    status: Literal["running", "success", "error", "partial"]
+    items_processed: int
+    items_created: int
+    errors: list[str]
+    started_at: int
+    completed_at: int | None
+
+
+# ── Chatbot Journey ───────────────────────────────────────────────────
+

docker-compose.yml

@@ -8,13 +8,16 @@ services:
         required: false
     environment:
       DATABASE_URL: postgresql+asyncpg://postgres:postgres@db:5432/adiuva
+      GITHUB_COPILOT_TOKEN_DIR: /root/.config/litellm/github_copilot
+    volumes:
+      - copilot_tokens:/root/.config/litellm/github_copilot
     depends_on:
       db:
         condition: service_healthy
     restart: unless-stopped
 
   db:
-    image: postgres:16-alpine
+    image: pgvector/pgvector:pg16
     environment:
       POSTGRES_USER: postgres
       POSTGRES_PASSWORD: postgres
@@ -66,3 +69,4 @@ volumes:
   postgres_data:
   minio_data:
   qdrant_data:
+  copilot_tokens:

docs/MICROSERVICES_ARCHITECTURE.md

@@ -0,0 +1,941 @@
+# Adiuva — Architettura Microservizi (MVP)
+
+## Panoramica
+
+Il monolite viene suddiviso in **4 servizi MVP** + un **API Gateway (Traefik)**, orchestrati con Docker Compose su un singolo VPS raggiungibile via Cloudflare.
+
+> **Fuori dall'MVP**: Storage Service (S3/backup CRUD) e Plugin Service (marketplace). Verranno aggiunti come servizi indipendenti in una fase successiva.
+
+```
+                          ┌──────────────┐
+                          │  Cloudflare  │
+                          │  (DNS + CDN) │
+                          └──────┬───────┘
+                                 │ HTTPS / WSS
+                          ┌──────▼───────┐
+                          │   Traefik    │
+                          │ API Gateway  │
+                          │  (routing,   │
+                          │   TLS, rate  │
+                          │   limiting)  │
+                          └──────┬───────┘
+                                 │
+          ┌──────────┬───────────┼───────────┐
+          │          │           │           │
+    ┌─────▼────┐ ┌───▼───┐ ┌────▼────┐ ┌────▼───┐
+    │  Auth    │ │  Chat │ │  Agent  │ │Billing │
+    │ Service  │ │Service│ │ Service │ │Service │
+    └─────┬────┘ └───┬───┘ └────┬────┘ └────┬───┘
+          │          │          │           │
+    ┌─────▼──────────▼──────────▼───────────▼────┐
+    │              Infrastruttura                 │
+    │  PostgreSQL  │  Redis  │  Qdrant            │
+    └─────────────────────────────────────────────┘
+```
+
+---
+
+## 1. Suddivisione dei Servizi
+
+### 1.1 Auth Service (`auth-service`)
+
+**Responsabilità**: Registrazione, login, refresh token, profilo utente, encryption key.
+
+| Endpoint originale | Metodo |
+|---|---|
+| `/api/v1/auth/register` | POST |
+| `/api/v1/auth/login` | POST |
+| `/api/v1/auth/refresh` | POST |
+| `/api/v1/auth/me` | GET / PUT |
+
+**Database**: Tabelle `users`, `refresh_tokens` (PostgreSQL condiviso, schema `auth`).
+
+**Modifica chiave — JWT con RS256**:
+Il monolite usa un `SECRET_KEY` simmetrico (HS256). Con i microservizi, passare a **RS256** (asimmetrico):
+- L'Auth Service firma i JWT con la **chiave privata**.
+- Tutti gli altri servizi verificano i JWT con la **chiave pubblica** senza mai contattare l'Auth Service.
+- La chiave pubblica viene esposta via `GET /api/v1/auth/.well-known/jwks.json` oppure montata come volume condiviso.
+
+```python
+# auth-service/app/auth/jwt.py
+from cryptography.hazmat.primitives.asymmetric import rsa
+from jose import jwt
+
+PRIVATE_KEY = ...  # Da env/secret
+PUBLIC_KEY = ...   # Derivata o da env
+
+def create_access_token(user_id: str, tier: str) -> str:
+    return jwt.encode(
+        {"sub": user_id, "tier": tier, "exp": ...},
+        PRIVATE_KEY,
+        algorithm="RS256",
+    )
+```
+
+```python
+# shared/auth.py  (usato da tutti gli altri servizi)
+from jose import jwt
+
+PUBLIC_KEY = ...  # Volume montato o fetched da JWKS endpoint
+
+def verify_token(token: str) -> dict:
+    return jwt.decode(token, PUBLIC_KEY, algorithms=["RS256"])
+```
+
+**Scaling**: 2 repliche sufficienti, stateless. Rate-limit dedicato su `/login` e `/register`.
+
+---
+
+### 1.2 Chat Service (`chat-service`) ⭐ Real-time
+
+**Responsabilità**: WebSocket device connection, home chat, floating chat, memory middleware, streaming LLM responses verso il client.
+
+Questo servizio gestisce la **connessione persistente** con l'app Electron e le interazioni **real-time** dell'utente (chat home, floating chat). È il proprietario della WebSocket.
+
+| Endpoint | Tipo |
+|---|---|
+| `/api/v1/ws/device` | WebSocket (connessione persistente) |
+| `/api/v1/chat` | POST (REST fallback) |
+
+**Moduli inclusi**: `deep_agent`, `memory_middleware`, `ws_context`, `device_manager` (Redis-backed), `output_formatter`, `llm`, tutti gli agent tools (`task_agent`, `project_agent`, `note_agent`, `timeline_agent`).
+
+**Perché separato dall'Agent Service**: Il Chat Service tiene la WebSocket aperta e risponde in tempo reale (streaming). Scalare aggiungendo repliche è semplice con sticky sessions + Redis pub/sub per il cross-instance routing dei tool_call.
+
+**Scaling**: 2–N repliche. Sticky cookies per le WS + Redis per cross-instance.
+
+---
+
+### 1.3 Agent Service (`agent-service`) ⭐ Batch
+
+**Responsabilità**: Batch agent processing (directory scanning, file classification, entity extraction), agent setup journeys, agent configuration CRUD.
+
+Questo servizio gestisce i processi **long-running** e **CPU-intensive**: scansione filesystem, classificazione file con LLM, estrazione entità in batch. Non possiede la WebSocket — comunica con il device dell'utente tramite **Redis pub/sub** passando per il Chat Service.
+
+| Endpoint | Tipo |
+|---|---|
+| `/api/v1/agents/catalog` | GET |
+| `/api/v1/agents/can-create` | POST |
+| `/api/v1/agents/trigger` | POST |
+| `/api/v1/agents/journey/start` | POST (o WS relay) |
+| `/api/v1/agents/journey/message` | POST (o WS relay) |
+
+**Moduli inclusi**: `agent_runner`, `agent_registry`, `filesystem_agent`, `llm`.
+
+**Flusso tool-call cross-service** (l'Agent Service non ha la WS):
+
+```
+┌──────────────┐            ┌──────────────┐            ┌──────────┐
+│ Agent Service│            │    Redis     │            │  Chat    │
+│ (batch run)  │            │              │            │ Service  │
+│              │            │              │            │ (ha WS)  │
+│ 1. Needs to  │  PUBLISH   │              │ SUBSCRIBE  │          │
+│    read file ├───────────►│tool_call:u123├───────────►│ 2. Invia │
+│    from      │            │              │            │    al    │
+│    device    │            │              │            │    device│
+│              │            │              │            │    via WS│
+│              │  SUBSCRIBE │              │  PUBLISH   │          │
+│ 4. Riceve   ◄────────────┤tool_result:id│◄───────────┤ 3. Device│
+│    risultato │            │              │            │    reply │
+└──────────────┘            └──────────────┘            └──────────┘
+```
+
+**Scaling**: 1–N repliche. Completamente stateless, scala indipendentemente dalla chat. Ogni replica processa batch job diversi. Può essere scalato a 0 se non ci sono agent attivi (risparmio risorse).
+
+**Vantaggio dello split**: Se 50 utenti triggerano agenti batch contemporaneamente, il Chat Service non ne risente — le risposte real-time rimangono veloci.
+
+---
+
+### 1.4 Billing Service (`billing-service`)
+
+**Responsabilità**: Stripe checkout, webhook, subscription management.
+
+| Endpoint originale | Metodo |
+|---|---|
+| `/api/v1/billing/checkout` | POST |
+| `/api/v1/billing/webhook` | POST |
+| `/api/v1/billing/subscription` | GET / DELETE |
+
+**Database**: Tabelle `subscriptions` (schema `billing`).
+
+**Comunicazione inter-servizio**: Quando Stripe invia un webhook e il tier cambia, il Billing Service pubblica un evento su **Redis pub/sub** channel `tier_changed:{user_id}`. L'Auth Service aggiorna il campo `tier` nella tabella users. Al prossimo token refresh il JWT conterrà il tier aggiornato.
+
+**Scaling**: 1 replica sufficiente. Basso traffico.
+
+---
+
+### 1.5 Servizi esclusi dall'MVP
+
+I seguenti servizi verranno aggiunti post-MVP come servizi indipendenti:
+
+| Servizio | Responsabilità | Note |
+|---|---|---|
+| **Storage Service** | S3 blobs CRUD, vector ops, backup | Le funzionalità vector/embed possono restare nel Chat Service per il MVP |
+| **Plugin Service** | Marketplace, install, revenue split | Feature non critica per il lancio |
+
+---
+
+## 2. Tier Check — Dove e Come
+
+Il tier dell'utente (free/pro/power/team) determina rate-limiting, quote e accesso a funzionalità. Con i microservizi, **ogni servizio controlla il tier autonomamente** senza chiamare l'Auth Service.
+
+### Strategia: Tier nel JWT
+
+L'Auth Service include il `tier` come claim nel JWT al momento del login/refresh:
+
+```json
+{
+  "sub": "user_123",
+  "tier": "pro",
+  "exp": 1742515200,
+  "iat": 1742511600
+}
+```
+
+Ogni servizio:
+1. Decodifica il JWT con la chiave pubblica (già lo fa per l'auth)
+2. Legge `payload["tier"]` — **zero chiamate extra**
+3. Applica le sue regole di enforcement localmente
+
+```python
+# shared/auth.py — dependency FastAPI condivisa
+from fastapi import Depends, HTTPException, Request
+from jose import jwt
+
+PUBLIC_KEY = ...
+
+class CurrentUser:
+    def __init__(self, user_id: str, tier: str):
+        self.user_id = user_id
+        self.tier = tier
+
+async def get_current_user(request: Request) -> CurrentUser:
+    token = request.headers.get("Authorization", "").removeprefix("Bearer ")
+    payload = jwt.decode(token, PUBLIC_KEY, algorithms=["RS256"])
+    return CurrentUser(user_id=payload["sub"], tier=payload["tier"])
+
+def require_tier(*allowed_tiers: str):
+    """Dependency che blocca se il tier non è tra quelli ammessi."""
+    async def check(user: CurrentUser = Depends(get_current_user)):
+        if user.tier not in allowed_tiers:
+            raise HTTPException(403, "Tier insufficient")
+        return user
+    return check
+```
+
+### Cosa succede quando il tier cambia (upgrade/downgrade)?
+
+```
+┌──────────┐  Stripe webhook   ┌──────────┐  tier_changed   ┌──────────┐
+│  Stripe  │ ─────────────────►│ Billing  │ ───────────────►│   Auth   │
+│          │                    │ Service  │  (Redis pub/sub) │ Service  │
+└──────────┘                    └──────────┘                  └────┬─────┘
+                                                                   │
+                                                          UPDATE users
+                                                          SET tier = 'power'
+                                                                   │
+                                                    Al prossimo /refresh
+                                                    il JWT conterrà tier='power'
+```
+
+**Latenza del cambio**: Il tier si propaga al prossimo token refresh (tipicamente 15–30 min, o il client può forzare un refresh immediato dopo il checkout). Per il billing webhook, il downgrade può essere forzato invalidando il refresh token su Redis → il client è obbligato a ri-autenticarsi.
+
+### Dove si applica in ciascun servizio
+
+| Servizio | Enforcement |
+|---|---|
+| **Auth Service** | Nessuno (è lui che scrive il tier) |
+| **Chat Service** | Rate-limit per tier (req/min), quota messaggi |
+| **Agent Service** | Max agent configs, max runs/day, max concurrent batches |
+| **Billing Service** | Nessuno (gestisce i tier, non li consuma) |
+
+### Rate-limit distribuito via Redis
+
+Poiché ogni servizio ha le sue repliche, il rate-limiting deve essere **condiviso** via Redis:
+
+```python
+# shared/middleware/rate_limit.py
+import redis.asyncio as aioredis
+
+class DistributedRateLimiter:
+    def __init__(self, redis: aioredis.Redis):
+        self._redis = redis
+
+    async def check(self, user_id: str, tier: str, service: str) -> bool:
+        limits = {"free": 20, "pro": 60, "power": 120, "team": 200}
+        max_req = limits.get(tier, 20)
+        key = f"rate:{service}:{user_id}"
+
+        pipe = self._redis.pipeline()
+        pipe.incr(key)
+        pipe.expire(key, 60)
+        count, _ = await pipe.execute()
+
+        return count <= max_req
+```
+
+---
+
+## 3. WebSocket con Scaling Orizzontale — Il Problema Chiave
+
+`DeviceConnectionManager` è un **singleton in-memory**:
+
+```python
+class DeviceConnectionManager:
+    def __init__(self):
+        self._connections: dict[str, DeviceConnection] = {}  # ← In-memory!
+```
+
+Con N istanze del Chat Service, il device si connette a **una sola** istanza. Quando un'altra istanza deve inviare un `tool_call` a quel device (es. un agent trigger da un'API call), non trova la connessione.
+
+### La soluzione: Redis Pub/Sub + Registry
+
+```
+┌──────────────────────────────────────────────────────────────┐
+│                     Redis                                    │
+│                                                              │
+│  Hash: ws:connections                                        │
+│    user_123 → instance_A                                     │
+│    user_456 → instance_B                                     │
+│                                                              │
+│  Pub/Sub channels:                                           │
+│    tool_call:{user_id}  → tool call payloads                 │
+│    tool_result:{call_id} → tool result payloads              │
+│    stream:{user_id}     → text_chunk streaming               │
+└──────────────────────────────────────────────────────────────┘
+
+ Instance A (ha WS di user_123)     Instance B (deve chiamare tool su user_123)
+ ┌───────────────────────┐          ┌───────────────────────┐
+ │  1. Sottoscrive a     │          │  1. Lookup Redis Hash │
+ │     tool_call:user_123│          │     → user_123 è su A │
+ │                       │          │                       │
+ │  2. Riceve tool_call  │◄─────────│  2. PUBLISH           │
+ │     da Redis channel  │          │    tool_call:user_123 │
+ │                       │          │    {id, action, ...}  │
+ │  3. Invia al device   │          │                       │
+ │     via WS            │          │  4. SUBSCRIBE         │
+ │                       │          │    tool_result:{id}   │
+ │  4. Device risponde   │          │                       │
+ │     tool_result       │──────────│► 5. Riceve risultato  │
+ │                       │          │                       │
+ │  5. PUBLISH           │          │                       │
+ │    tool_result:{id}   │          │                       │
+ └───────────────────────┘          └───────────────────────┘
+```
+
+### Implementazione: `RedisDeviceManager`
+
+```python
+# chat-service/app/core/device_manager.py
+
+import asyncio
+import json
+import os
+import redis.asyncio as aioredis
+from dataclasses import dataclass, field
+from fastapi import WebSocket
+
+INSTANCE_ID = os.environ.get("INSTANCE_ID", os.urandom(8).hex())
+
+@dataclass
+class LocalConnection:
+    ws: WebSocket
+    device_id: str
+    pending_calls: dict[str, asyncio.Future[dict]] = field(default_factory=dict)
+
+
+class RedisDeviceManager:
+    """Device manager backed by Redis for cross-instance communication."""
+
+    def __init__(self, redis_url: str = "redis://redis:6379"):
+        self._redis = aioredis.from_url(redis_url)
+        self._pubsub = self._redis.pubsub()
+        self._local: dict[str, LocalConnection] = {}  # Solo connessioni locali
+        self._remote_futures: dict[str, asyncio.Future[dict]] = {}
+
+    async def start(self):
+        """Avvia il listener Redis per tool_call in arrivo."""
+        asyncio.create_task(self._listen_tool_calls())
+
+    # ── Registrazione ──
+
+    async def register(self, user_id: str, device_id: str, ws: WebSocket):
+        # Registra localmente
+        self._local[user_id] = LocalConnection(ws=ws, device_id=device_id)
+        # Registra in Redis quale istanza ha la connessione
+        await self._redis.hset("ws:connections", user_id, INSTANCE_ID)
+        # Sottoscrivi ai tool_call per questo utente
+        await self._pubsub.subscribe(f"tool_call:{user_id}")
+
+    async def unregister(self, user_id: str):
+        conn = self._local.pop(user_id, None)
+        if conn:
+            for fut in conn.pending_calls.values():
+                if not fut.done():
+                    fut.cancel()
+        await self._redis.hdel("ws:connections", user_id)
+        await self._pubsub.unsubscribe(f"tool_call:{user_id}")
+
+    # ── Presenza ──
+
+    async def is_online(self, user_id: str) -> bool:
+        return await self._redis.hexists("ws:connections", user_id)
+
+    # ── Tool-call round-trip (cross-instance) ──
+
+    async def execute_tool_call(self, user_id: str, payload: dict) -> dict:
+        """
+        Invia un tool_call al device dell'utente.
+        Funziona sia che la WS sia locale che su un'altra istanza.
+        """
+        call_id = payload["id"]
+
+        # Caso 1: connessione locale → invio diretto
+        if user_id in self._local:
+            conn = self._local[user_id]
+            loop = asyncio.get_event_loop()
+            fut: asyncio.Future[dict] = loop.create_future()
+            conn.pending_calls[call_id] = fut
+            await conn.ws.send_text(json.dumps({"type": "tool_call", **payload}))
+            return await asyncio.wait_for(fut, timeout=30.0)
+
+        # Caso 2: connessione remota → Redis pub/sub
+        loop = asyncio.get_event_loop()
+        fut = loop.create_future()
+        self._remote_futures[call_id] = fut
+
+        # Sottoscrivi al canale di risposta
+        result_channel = f"tool_result:{call_id}"
+        await self._pubsub.subscribe(result_channel)
+
+        # Pubblica il tool_call
+        await self._redis.publish(
+            f"tool_call:{user_id}",
+            json.dumps(payload),
+        )
+
+        try:
+            return await asyncio.wait_for(fut, timeout=30.0)
+        finally:
+            self._remote_futures.pop(call_id, None)
+            await self._pubsub.unsubscribe(result_channel)
+
+    # ── Risoluzione tool_result (da WS locale) ──
+
+    def resolve_local(self, user_id: str, call_id: str, result: dict):
+        conn = self._local.get(user_id)
+        if conn:
+            fut = conn.pending_calls.pop(call_id, None)
+            if fut and not fut.done():
+                fut.set_result(result)
+
+    async def resolve_and_publish(self, user_id: str, call_id: str, result: dict):
+        """Chiamato quando il device locale invia un tool_result."""
+        self.resolve_local(user_id, call_id, result)
+        # Pubblica anche su Redis per l'istanza remota che aspetta
+        await self._redis.publish(
+            f"tool_result:{call_id}",
+            json.dumps(result),
+        )
+
+    # ── Listener Redis ──
+
+    async def _listen_tool_calls(self):
+        """Loop che ascolta i tool_call in arrivo da altre istanze."""
+        async for message in self._pubsub.listen():
+            if message["type"] != "message":
+                continue
+            channel = message["channel"]
+            if isinstance(channel, bytes):
+                channel = channel.decode()
+
+            data = json.loads(message["data"])
+
+            if channel.startswith("tool_call:"):
+                # Un'altra istanza vuole che inviamo un tool_call al nostro device
+                user_id = channel.split(":", 1)[1]
+                conn = self._local.get(user_id)
+                if conn:
+                    await conn.ws.send_text(json.dumps({"type": "tool_call", **data}))
+
+            elif channel.startswith("tool_result:"):
+                # Risposta a un tool_call che abbiamo inviato tramite Redis
+                call_id = channel.split(":", 1)[1]
+                fut = self._remote_futures.pop(call_id, None)
+                if fut and not fut.done():
+                    fut.set_result(data)
+
+    # ── Stream cross-instance ──
+
+    async def publish_stream_chunk(self, user_id: str, chunk: dict):
+        """Pubblica un chunk di streaming su Redis (per REST→WS relay)."""
+        await self._redis.publish(f"stream:{user_id}", json.dumps(chunk))
+```
+
+---
+
+## 4. Struttura Directory Proposta (MVP)
+
+```
+adiuva-api/
+├── docker-compose.yml          # Orchestrazione completa
+├── docker-compose.dev.yml      # Override per sviluppo locale
+├── shared/                     # Codice condiviso (montato come volume)
+│   ├── auth.py                 # JWT verification (chiave pubblica)
+│   ├── schemas.py              # Pydantic schemas condivisi
+│   ├── middleware/
+│   │   ├── rate_limit.py       # DistributedRateLimiter (Redis)
+│   │   └── sanitizer.py
+│   └── models/
+│       └── base.py             # SQLAlchemy base condivisa
+│
+├── auth-service/
+│   ├── Dockerfile
+│   ├── requirements.txt
+│   └── app/
+│       ├── main.py
+│       ├── config.py
+│       ├── db.py
+│       ├── models.py           # users, refresh_tokens
+│       ├── routes/
+│       │   └── auth.py
+│       └── services/
+│           ├── jwt_service.py  # RS256 signing
+│           └── user_service.py
+│
+├── chat-service/
+│   ├── Dockerfile
+│   ├── requirements.txt
+│   └── app/
+│       ├── main.py
+│       ├── config.py
+│       ├── db.py
+│       ├── models.py           # memory_*
+│       ├── routes/
+│       │   ├── device_ws.py    # WS connection owner
+│       │   └── chat.py         # REST fallback
+│       ├── core/
+│       │   ├── device_manager.py   # RedisDeviceManager
+│       │   ├── deep_agent.py       # Home + floating chat
+│       │   ├── memory_middleware.py
+│       │   ├── ws_context.py
+│       │   ├── output_formatter.py
+│       │   └── llm.py
+│       └── agents/                 # Tool definitions (used by deep_agent)
+│           ├── task_agent.py
+│           ├── project_agent.py
+│           ├── note_agent.py
+│           └── timeline_agent.py
+│
+├── agent-service/
+│   ├── Dockerfile
+│   ├── requirements.txt
+│   └── app/
+│       ├── main.py
+│       ├── config.py
+│       ├── db.py
+│       ├── models.py           # agent_run_logs, local/cloud_agent_configs
+│       ├── routes/
+│       │   ├── agents.py       # catalog, can-create, trigger
+│       │   └── agent_setup.py  # journey start/message
+│       ├── core/
+│       │   ├── agent_runner.py     # Batch classify → process
+│       │   ├── agent_registry.py
+│       │   ├── redis_executor.py   # execute_on_client via Redis pub/sub
+│       │   └── llm.py
+│       └── agents/
+│           ├── task_agent.py       # Tool definitions (batch context)
+│           ├── project_agent.py
+│           ├── note_agent.py
+│           ├── timeline_agent.py
+│           └── filesystem_agent.py
+│
+├── billing-service/
+│   ├── Dockerfile
+│   ├── requirements.txt
+│   └── app/
+│       ├── main.py
+│       ├── config.py
+│       ├── db.py
+│       ├── models.py           # subscriptions
+│       ├── routes/
+│       │   └── billing.py
+│       └── services/
+│           ├── stripe_service.py
+│           └── tier_manager.py
+│
+└── infra/
+    ├── traefik/
+    │   └── traefik.yml
+    ├── keys/
+    │   ├── jwt_private.pem     # Solo auth-service
+    │   └── jwt_public.pem      # Tutti i servizi
+    └── alembic/                # Migrazioni condivise o per-servizio
+```
+
+---
+
+## 5. Docker Compose — Configurazione MVP
+
+```yaml
+# docker-compose.yml
+
+services:
+
+  # ══════════════════════════════════════════════════════════
+  # API Gateway
+  # ══════════════════════════════════════════════════════════
+  traefik:
+    image: traefik:v3.2
+    command:
+      - "--api.insecure=true"
+      - "--providers.docker=true"
+      - "--providers.docker.exposedbydefault=false"
+      - "--entrypoints.web.address=:80"
+      - "--entrypoints.websecure.address=:443"
+      - "--entrypoints.web.http.redirections.entrypoint.to=websecure"
+    ports:
+      - "80:80"
+      - "443:443"
+      - "8080:8080"   # Dashboard Traefik (disabilitare in prod)
+    volumes:
+      - /var/run/docker.sock:/var/run/docker.sock:ro
+      - ./infra/certs:/certs:ro
+    restart: unless-stopped
+
+  # ══════════════════════════════════════════════════════════
+  # Auth Service (2 repliche)
+  # ══════════════════════════════════════════════════════════
+  auth-service:
+    build: ./auth-service
+    deploy:
+      replicas: 2
+    env_file: .env
+    environment:
+      DATABASE_URL: postgresql+asyncpg://postgres:postgres@db:5432/adiuva
+      REDIS_URL: redis://redis:6379
+      JWT_PRIVATE_KEY_FILE: /run/secrets/jwt_private_key
+      SERVICE_NAME: auth
+    secrets:
+      - jwt_private_key
+      - jwt_public_key
+    labels:
+      - "traefik.enable=true"
+      - "traefik.http.routers.auth.rule=PathPrefix(`/api/v1/auth`)"
+      - "traefik.http.services.auth.loadbalancer.server.port=8000"
+    depends_on:
+      db:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+
+  # ══════════════════════════════════════════════════════════
+  # Chat Service — Real-time WS + Chat (scalabile)
+  # ══════════════════════════════════════════════════════════
+  chat-service:
+    build: ./chat-service
+    deploy:
+      replicas: 2
+    env_file: .env
+    environment:
+      DATABASE_URL: postgresql+asyncpg://postgres:postgres@db:5432/adiuva
+      REDIS_URL: redis://redis:6379
+      JWT_PUBLIC_KEY_FILE: /run/secrets/jwt_public_key
+      SERVICE_NAME: chat
+    secrets:
+      - jwt_public_key
+    labels:
+      - "traefik.enable=true"
+      # REST chat endpoint
+      - "traefik.http.routers.chat.rule=PathPrefix(`/api/v1/chat`)"
+      - "traefik.http.services.chat.loadbalancer.server.port=8000"
+      # WebSocket route con sticky session
+      - "traefik.http.routers.ws.rule=PathPrefix(`/api/v1/ws`)"
+      - "traefik.http.routers.ws.service=chat-ws"
+      - "traefik.http.services.chat-ws.loadbalancer.server.port=8000"
+      - "traefik.http.services.chat-ws.loadbalancer.sticky.cookie.name=ws_affinity"
+      - "traefik.http.services.chat-ws.loadbalancer.sticky.cookie.httpOnly=true"
+    depends_on:
+      db:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+
+  # ══════════════════════════════════════════════════════════
+  # Agent Service — Batch processing (scalabile indipendentemente)
+  # ══════════════════════════════════════════════════════════
+  agent-service:
+    build: ./agent-service
+    deploy:
+      replicas: 2
+    env_file: .env
+    environment:
+      DATABASE_URL: postgresql+asyncpg://postgres:postgres@db:5432/adiuva
+      REDIS_URL: redis://redis:6379
+      JWT_PUBLIC_KEY_FILE: /run/secrets/jwt_public_key
+      SERVICE_NAME: agent
+    secrets:
+      - jwt_public_key
+    labels:
+      - "traefik.enable=true"
+      - "traefik.http.routers.agents.rule=PathPrefix(`/api/v1/agents`)"
+      - "traefik.http.services.agents.loadbalancer.server.port=8000"
+    depends_on:
+      db:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+
+  # ══════════════════════════════════════════════════════════
+  # Billing Service (1 replica)
+  # ══════════════════════════════════════════════════════════
+  billing-service:
+    build: ./billing-service
+    deploy:
+      replicas: 1
+    env_file: .env
+    environment:
+      DATABASE_URL: postgresql+asyncpg://postgres:postgres@db:5432/adiuva
+      REDIS_URL: redis://redis:6379
+      JWT_PUBLIC_KEY_FILE: /run/secrets/jwt_public_key
+      SERVICE_NAME: billing
+    secrets:
+      - jwt_public_key
+    labels:
+      - "traefik.enable=true"
+      - "traefik.http.routers.billing.rule=PathPrefix(`/api/v1/billing`)"
+      - "traefik.http.services.billing.loadbalancer.server.port=8000"
+    depends_on:
+      db:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+
+  # ══════════════════════════════════════════════════════════
+  # Infrastruttura
+  # ══════════════════════════════════════════════════════════
+  db:
+    image: pgvector/pgvector:pg16
+    environment:
+      POSTGRES_USER: postgres
+      POSTGRES_PASSWORD: postgres
+      POSTGRES_DB: adiuva
+    volumes:
+      - postgres_data:/var/lib/postgresql/data
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U postgres"]
+      interval: 5s
+      timeout: 5s
+      retries: 5
+    restart: unless-stopped
+
+  redis:
+    image: redis:7-alpine
+    command: redis-server --maxmemory 256mb --maxmemory-policy allkeys-lru
+    volumes:
+      - redis_data:/data
+    healthcheck:
+      test: ["CMD", "redis-cli", "ping"]
+      interval: 5s
+      timeout: 3s
+      retries: 5
+    restart: unless-stopped
+
+  qdrant:
+    image: qdrant/qdrant:latest
+    volumes:
+      - qdrant_data:/qdrant/storage
+    restart: unless-stopped
+
+secrets:
+  jwt_private_key:
+    file: ./infra/keys/jwt_private.pem
+  jwt_public_key:
+    file: ./infra/keys/jwt_public.pem
+
+volumes:
+  postgres_data:
+  redis_data:
+  qdrant_data:
+```
+
+---
+
+## 6. Configurazione Cloudflare + VPS
+
+### 6.1 DNS
+
+```
+api.tuodominio.com  →  A record  →  IP del VPS
+                    →  Proxy: ON (orange cloud)
+```
+
+### 6.2 Cloudflare Settings
+
+| Setting | Valore | Motivo |
+|---------|--------|--------|
+| SSL/TLS mode | **Full (Strict)** | Cloudflare ↔ VPS con certificato valido |
+| WebSocket | **ON** | Necessario per `/api/v1/ws/device` |
+| Proxy timeout | **100s** (Enterprise) o default | Le LLM calls possono durare 30s+ |
+| Under Attack Mode | Off (attivare se necessario) | |
+
+### 6.3 TLS sul VPS
+
+Due opzioni:
+- **Opzione A (consigliata)**: Cloudflare Origin Certificate → montato in Traefik
+- **Opzione B**: Let's Encrypt via Traefik (con DNS challenge Cloudflare)
+
+```yaml
+# traefik.yml — con Cloudflare Origin Certificate
+entryPoints:
+  websecure:
+    address: ":443"
+
+tls:
+  certificates:
+    - certFile: /certs/origin.pem
+      keyFile: /certs/origin-key.pem
+```
+
+### 6.4 Rete VPS
+
+```bash
+# UFW firewall — solo Cloudflare può raggiungere le porte 80/443
+# https://www.cloudflare.com/ips/
+ufw default deny incoming
+ufw allow from 173.245.48.0/20 to any port 443
+ufw allow from 103.21.244.0/22 to any port 443
+# ... (tutti gli IP range di Cloudflare)
+ufw allow ssh
+ufw enable
+```
+
+---
+
+## 7. Comunicazione Inter-Servizio
+
+### 7.1 Redis Pub/Sub — Event Bus
+
+```
+┌──────────┐  tier_changed:user_123   ┌──────────┐
+│ Billing  │ ────────────────────────► │   Auth   │
+│ Service  │                           │ Service  │
+└──────────┘                           └──────────┘
+
+┌──────────┐  tool_call:user_123      ┌──────────┐
+│  Agent   │ ────────────────────────► │   Chat   │
+│ Service  │                           │ Service  │
+│ (batch)  │ ◄────────────────────────│ (ha WS)  │
+└──────────┘  tool_result:{call_id}    └──────────┘
+```
+
+### 7.2 Health Checks e Service Discovery
+
+Traefik gestisce automaticamente il service discovery via Docker labels. I servizi non devono conoscersi tra loro — comunicano solo via:
+- **Redis pub/sub** (tool-call cross-instance, tier events)
+- **Redis hash** (stato condiviso: `ws:connections`, rate-limit counters)
+- **PostgreSQL** (dati persistenti condivisi)
+
+---
+
+## 8. Piano di Migrazione Incrementale (MVP)
+
+### Fase 1 — Preparazione (nel monolite attuale)
+1. Aggiungere Redis al `docker-compose.yml` attuale
+2. Migrare JWT da HS256 → RS256 (backward-compatible: accetta entrambi per un periodo)
+3. Implementare `RedisDeviceManager` come drop-in replacement del singleton in-memory
+4. Estrarre `shared/` con auth verification, schemas, middleware
+
+### Fase 2 — Auth Service (primo split)
+1. Estrarre `auth.py` routes + models in `auth-service/`
+2. Verificare che i JWT firmati da `auth-service` vengano validati dal monolite
+3. Aggiungere Traefik e routare `/api/v1/auth/*` al nuovo servizio
+4. Il monolite continua a servire tutto il resto
+
+### Fase 3 — Billing Service
+1. Estrarre billing routes, Stripe service, tier manager
+2. Configurare Redis pub/sub per `tier_changed` events
+3. Routare via Traefik
+
+### Fase 4 — Split Chat + Agent (il più delicato)
+1. Il monolite residuo contiene WS + chat + agents
+2. Separare Agent Service: estrarre `agent_runner`, `agent_registry`, `agent_setup`, route `/agents/*`
+3. Implementare `redis_executor.py` nell'Agent Service per tool-call via Redis
+4. Il Chat Service resta proprietario della WS e sottoscrive i canali `tool_call:{user_id}`
+5. Testare: trigger agent dall'Agent Service → tool_call via Redis → Chat Service → WS → device → risposta
+
+### Fase 5 — Scaling test
+1. Scalare Chat Service a 2 repliche, verificare sticky sessions
+2. Scalare Agent Service a 2 repliche, verificare batch processing distribuito
+3. Monitoring (Prometheus + Grafana) per ogni servizio
+
+---
+
+## 9. Monitoraggio e Logging
+
+```yaml
+# Aggiungere al docker-compose.yml
+
+  prometheus:
+    image: prom/prometheus:latest
+    volumes:
+      - ./infra/prometheus/prometheus.yml:/etc/prometheus/prometheus.yml
+    restart: unless-stopped
+
+  grafana:
+    image: grafana/grafana:latest
+    ports:
+      - "3000:3000"
+    volumes:
+      - grafana_data:/var/lib/grafana
+    restart: unless-stopped
+
+  loki:
+    image: grafana/loki:latest
+    restart: unless-stopped
+```
+
+Ogni servizio espone `/metrics` (Prometheus) e scrive log strutturati (JSON) raccolti da Loki.
+
+---
+
+## 10. Sizing VPS Minimo Consigliato (MVP)
+
+| Componente | CPU | RAM | Note |
+|---|---|---|---|
+| Traefik | 0.25 | 128MB | |
+| Auth Service ×2 | 0.25 ×2 | 128MB ×2 | Stateless, leggero |
+| Chat Service ×2 | 1.0 ×2 | 1GB ×2 | WS + streaming LLM |
+| Agent Service ×2 | 0.75 ×2 | 512MB ×2 | Batch LLM, CPU-bound |
+| Billing Service | 0.25 | 128MB | |
+| PostgreSQL | 1.0 | 1GB | |
+| Redis | 0.25 | 256MB | |
+| Qdrant | 0.5 | 512MB | |
+| **Totale MVP** | **~5.5 vCPU** | **~5 GB** | |
+
+**Raccomandazione**: VPS con **8 vCPU / 16 GB RAM** per avere margine. Hetzner CPX41 (~€30/mese) o equivalente. Senza Storage/Plugin si risparmia ~1 vCPU e 512MB rispetto alla versione completa.
+
+---
+
+## Riepilogo Architettura MVP
+
+| Servizio | Repliche | Proprietario di |
+|---|---|---|
+| **Traefik** | 1 | Routing, TLS, sticky sessions |
+| **Auth Service** | 2 | JWT RS256, registrazione, login, profilo |
+| **Chat Service** | 2–N | WebSocket, home/floating chat, streaming |
+| **Agent Service** | 2–N | Batch processing, directory scan, agent setup |
+| **Billing Service** | 1 | Stripe, subscriptions, tier management |
+
+| Decisione | Scelta | Motivazione |
+|---|---|---|
+| API Gateway | Traefik | Nativo Docker, WebSocket support, service discovery automatico |
+| JWT | RS256 (asimmetrico) | Verifica distribuita senza contattare Auth Service |
+| Tier check | Claim nel JWT | Ogni servizio verifica localmente, zero roundtrip |
+| WebSocket scaling | Redis pub/sub + sticky cookies | Cross-instance tool-call routing |
+| Chat ↔ Agent split | Servizi separati | Batch CPU-bound non impatta real-time chat |
+| Agent → Device comms | Redis pub/sub via Chat Service | Agent non possiede la WS, usa un relay |
+| Rate limiting | Redis contatori distribuiti | Sliding window condivisa tra repliche |
+| Database | PostgreSQL condiviso | Semplicità MVP; split DB futuro facile |
+| TLS | Cloudflare Origin Certificate | Zero maintenance |
+| Orchestrazione | Docker Compose | Sufficiente per un singolo VPS |
+| Storage / Plugin | Post-MVP | Non critici per il lancio |

logging.conf

@@ -0,0 +1,56 @@
+[loggers]
+keys=root,uvicorn,uvicorn.error,uvicorn.access,sqlalchemy,watchfiles
+
+[handlers]
+keys=console,file
+
+[formatters]
+keys=default
+
+[logger_root]
+level=INFO
+handlers=console,file
+
+[logger_uvicorn]
+level=INFO
+handlers=
+qualname=uvicorn
+propagate=1
+
+[logger_uvicorn.error]
+level=INFO
+handlers=
+qualname=uvicorn.error
+propagate=1
+
+[logger_uvicorn.access]
+level=INFO
+handlers=
+qualname=uvicorn.access
+propagate=1
+
+[logger_sqlalchemy]
+level=WARNING
+handlers=
+qualname=sqlalchemy
+propagate=1
+
+[logger_watchfiles]
+level=WARNING
+handlers=
+qualname=watchfiles
+propagate=1
+
+[handler_console]
+class=StreamHandler
+formatter=default
+args=(sys.stderr,)
+
+[handler_file]
+class=logging.handlers.RotatingFileHandler
+formatter=default
+args=('logs/app.log', 'a', 10485760, 5, 'utf-8')
+
+[formatter_default]
+format=%(asctime)s %(levelname)s %(name)s: %(message)s
+datefmt=%Y-%m-%d %H:%M:%S

requirements.txt

@@ -3,6 +3,7 @@ uvicorn[standard]>=0.34.0
 gunicorn>=22.0.0
 langchain>=0.3.0
 langchain-openai>=0.3.0
+langchain-litellm>=0.1.0
 litellm>=1.50.0
 pydantic>=2.10.0
 pydantic-settings>=2.7.0
@@ -24,4 +25,13 @@ aiosqlite>=0.20.0
 moto[s3]>=5.0.0
 pinecone>=5.0.0
 qdrant-client>=1.7.0
+croniter>=3.0.0
+google-api-python-client>=2.130.0
+google-auth>=2.29.0
+google-auth-oauthlib>=1.2.0
+google-auth-httplib2>=0.2.0
+msal>=1.28.0
+cryptography>=42.0.0
+redis>=5.0.0
+langfuse>=3.0.0
 ruff>=0.8.0

services/auth/.env.example

@@ -0,0 +1,19 @@
+# ── Auth Service ──────────────────────────────────────────────────────────────
+# This file contains env vars specific to the Auth Service.
+# Shared vars (DATABASE_URL, REDIS_URL, etc.) come from the root .env
+# or from docker-compose environment.
+
+# ── JWT RS256 Keys ────────────────────────────────────────────────────────────
+# Generate keypair:
+#   openssl genpkey -algorithm RSA -out private.pem -pkeyopt rsa_keygen_bits:2048
+#   openssl rsa -in private.pem -pubout -out public.pem
+#
+# Paste PEM content with literal \n for newlines:
+#   JWT_PRIVATE_KEY=-----BEGIN PRIVATE KEY-----\nMIIEvQ...
+#   JWT_PUBLIC_KEY=-----BEGIN PUBLIC KEY-----\nMIIBIj...
+
+# PRIVATE KEY — used to SIGN JWTs. NEVER share outside this service.
+JWT_PRIVATE_KEY=
+
+# PUBLIC KEY — used to VERIFY JWTs.
+JWT_PUBLIC_KEY=

services/auth/Dockerfile

@@ -0,0 +1,36 @@
+# ── builder ──────────────────────────────────────────────────────────────────
+FROM python:3.12-slim AS builder
+
+WORKDIR /build
+
+# Install shared + service deps in one layer
+COPY services/auth/requirements.txt ./requirements.txt
+RUN pip install --upgrade pip && \
+    pip install --no-cache-dir --prefix=/install -r requirements.txt
+
+# ── runtime ──────────────────────────────────────────────────────────────────
+FROM python:3.12-slim AS runtime
+
+RUN addgroup --system appgroup && adduser --system --ingroup appgroup appuser
+
+WORKDIR /app
+
+COPY --from=builder /install /usr/local
+
+# Copy shared module (available to all services)
+COPY shared/ shared/
+
+# Copy service source
+COPY services/auth/app/ app/
+
+RUN chown -R appuser:appgroup /app
+
+USER appuser
+
+EXPOSE 8000
+
+CMD ["gunicorn", "app.main:app", \
+     "-k", "uvicorn.workers.UvicornWorker", \
+     "--bind", "0.0.0.0:8000", \
+     "--workers", "2", \
+     "--timeout", "30"]

services/auth/README.md

@@ -0,0 +1,16 @@
+# Auth Service
+
+Owns: user registration, login, JWT RS256 issuance, token refresh, `/me` endpoint.
+
+## Tables owned
+- `users`
+- `refresh_tokens`
+- `subscriptions` (read; Billing Service writes)
+
+## Endpoints
+- `POST /auth/register`
+- `POST /auth/login`
+- `POST /auth/refresh`
+- `GET /auth/me`
+- `PUT /auth/me`
+- `GET /auth/verify` (ForwardAuth for Traefik)

services/auth/app/config.py

@@ -0,0 +1,34 @@
+"""Auth Service — local configuration.
+
+Contains secrets that ONLY the Auth Service needs (e.g., JWT private key).
+These are NOT in shared/config.py to prevent other services from accessing them.
+"""
+
+from pydantic import field_validator
+from pydantic_settings import BaseSettings, SettingsConfigDict
+
+
+class AuthSettings(BaseSettings):
+    # RS256 private key (PEM format). Used to SIGN JWTs.
+    # Only the Auth Service has this. Generate with:
+    #   openssl genpkey -algorithm RSA -out private.pem -pkeyopt rsa_keygen_bits:2048
+    # Then set the env var (newlines as \n):
+    #   JWT_PRIVATE_KEY="-----BEGIN PRIVATE KEY-----\nMIIEv..."
+    JWT_PRIVATE_KEY: str = ""
+
+    # RS256 public key (PEM format). Used to VERIFY JWTs.
+    # Derived from the private key:
+    #   openssl rsa -in private.pem -pubout -out public.pem
+    JWT_PUBLIC_KEY: str = ""
+
+    @field_validator("JWT_PRIVATE_KEY", "JWT_PUBLIC_KEY", mode="before")
+    @classmethod
+    def _expand_pem_newlines(cls, v: str) -> str:
+        if isinstance(v, str) and r"\n" in v:
+            return v.replace(r"\n", "\n")
+        return v
+
+    model_config = SettingsConfigDict(env_file=".env", env_file_encoding="utf-8")
+
+
+auth_settings = AuthSettings()

services/auth/app/deps.py

@@ -0,0 +1,69 @@
+"""Auth dependencies — JWT validation for the Auth Service.
+
+This is the canonical get_current_user used by protected endpoints
+within the Auth Service itself (/me, /me PUT).
+"""
+
+from __future__ import annotations
+
+from fastapi import Depends, HTTPException, status
+from fastapi.security import OAuth2PasswordBearer
+from jose import JWTError, jwt
+from sqlalchemy import select
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from shared.config import settings
+from shared.db import get_session
+from shared.models import Subscription, User
+from shared.schemas import UserProfile
+
+from app.config import auth_settings
+
+oauth2_scheme = OAuth2PasswordBearer(tokenUrl="/api/v1/auth/login")
+
+
+async def get_current_user(
+    token: str = Depends(oauth2_scheme),
+    db: AsyncSession = Depends(get_session),
+) -> UserProfile:
+    """Validate a Bearer JWT and return the authenticated user.
+
+    The JWT is used for identity and expiry.  Tier is fetched live from the
+    subscriptions table so upgrades/downgrades take effect immediately.
+    """
+    credentials_exc = HTTPException(
+        status_code=status.HTTP_401_UNAUTHORIZED,
+        detail="Could not validate credentials",
+        headers={"WWW-Authenticate": "Bearer"},
+    )
+    try:
+        payload = jwt.decode(
+            token, auth_settings.JWT_PUBLIC_KEY, algorithms=["RS256"]
+        )
+        user_id: str | None = payload.get("sub")
+        email: str | None = payload.get("email")
+        if not user_id or not email:
+            raise credentials_exc
+    except JWTError:
+        raise credentials_exc
+
+    # Live tier lookup
+    result = await db.execute(
+        select(Subscription.tier).where(Subscription.user_id == user_id)
+    )
+    default_tier = "power" if settings.ENV == "dev" else "free"
+    tier: str = result.scalar_one_or_none() or default_tier
+
+    # Fetch name/surname
+    user_result = await db.execute(
+        select(User.name, User.surname).where(User.id == user_id)
+    )
+    user_row = user_result.one_or_none()
+
+    return UserProfile(
+        id=user_id,
+        email=email,
+        name=user_row.name if user_row else None,
+        surname=user_row.surname if user_row else None,
+        tier=tier,
+    )  # type: ignore[arg-type]

services/auth/app/main.py

@@ -0,0 +1,62 @@
+"""Auth Service — JWT issuance, user management, ForwardAuth verification.
+
+Standalone FastAPI service extracted from the adiuva-api monolith.
+Owns: users, refresh_tokens, subscriptions (read).
+"""
+
+import sys
+from contextlib import asynccontextmanager
+from pathlib import Path
+
+# Ensure the repo root is on sys.path so "shared" is importable.
+# In Docker, COPY shared/ puts it at /app/shared/ (already importable).
+# In local dev, we need to add the repo root (two levels up from this file).
+_repo_root = str(Path(__file__).resolve().parents[3])
+if _repo_root not in sys.path:
+    sys.path.insert(0, _repo_root)
+
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+
+from shared.config import settings
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    yield
+    from shared.db import engine
+
+    await engine.dispose()
+
+
+def create_app() -> FastAPI:
+    app = FastAPI(
+        title="Adiuva Auth Service",
+        version="0.1.0",
+        docs_url="/docs" if settings.ENV == "dev" else None,
+        redoc_url=None,
+        lifespan=lifespan,
+    )
+
+    app.add_middleware(
+        CORSMiddleware,
+        allow_origins=settings.CORS_ORIGINS,
+        allow_credentials=True,
+        allow_methods=["*"],
+        allow_headers=["*"],
+    )
+
+    from app.routes import router
+    from app.verify import router as verify_router
+
+    app.include_router(router, prefix="/api/v1")
+    app.include_router(verify_router, prefix="/api/v1")
+
+    @app.get("/api/v1/health", tags=["health"])
+    async def health() -> dict:
+        return {"status": "ok", "service": "auth", "version": app.version}
+
+    return app
+
+
+app = create_app()

services/auth/app/routes.py

@@ -0,0 +1,249 @@
+"""Auth routes: register, login, refresh, me.
+
+Extracted from app/api/routes/auth.py — uses shared.* imports instead of app.*.
+"""
+
+from __future__ import annotations
+
+import hashlib
+import time
+import uuid
+from datetime import datetime, timedelta, timezone
+
+import bcrypt
+from cryptography.fernet import Fernet
+from fastapi import APIRouter, Depends, HTTPException, status
+from jose import jwt
+from pydantic import BaseModel
+from sqlalchemy import select
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from shared.config import settings
+from shared.db import get_session
+from shared.models import RefreshToken, Subscription, User
+from shared.schemas import AuthTokens, UserProfile
+
+from app.config import auth_settings
+from app.deps import get_current_user
+
+router = APIRouter(prefix="/auth", tags=["auth"])
+
+
+# ── Internal helpers ─────────────────────────────────────────────────
+
+
+def _hash_password(password: str) -> str:
+    return bcrypt.hashpw(password.encode(), bcrypt.gensalt()).decode()
+
+
+def _verify_password(password: str, hashed: str) -> bool:
+    return bcrypt.checkpw(password.encode(), hashed.encode())
+
+
+def _hash_token(plain_token: str) -> str:
+    """SHA-256 of the plain refresh token string."""
+    return hashlib.sha256(plain_token.encode()).hexdigest()
+
+
+def _make_access_token(user_id: str, email: str, tier: str) -> tuple[str, int]:
+    """Return (RS256-signed JWT, expires_at_ms)."""
+    now = int(time.time())
+    exp = now + settings.JWT_ACCESS_TOKEN_EXPIRE_MINUTES * 60
+    payload = {
+        "sub": user_id,
+        "email": email,
+        "tier": tier,
+        "exp": exp,
+        "iat": now,
+    }
+    token = jwt.encode(payload, auth_settings.JWT_PRIVATE_KEY, algorithm="RS256")
+    return token, exp * 1000  # ms for client
+
+
+async def _get_live_tier(db: AsyncSession, user_id: str) -> str:
+    """Fetch authoritative tier from subscriptions table."""
+    result = await db.execute(
+        select(Subscription.tier).where(Subscription.user_id == user_id)
+    )
+    default_tier = "power" if settings.ENV == "dev" else "free"
+    return result.scalar_one_or_none() or default_tier
+
+
+# ── Request bodies ────────────────────────────────────────────────────
+
+
+class _RegisterRequest(BaseModel):
+    email: str
+    password: str
+    name: str | None = None
+    surname: str | None = None
+
+
+class _LoginRequest(BaseModel):
+    email: str
+    password: str
+
+
+class _RefreshRequest(BaseModel):
+    refresh_token: str
+
+
+class _UpdateProfileRequest(BaseModel):
+    name: str | None = None
+    surname: str | None = None
+
+
+# ── Routes ────────────────────────────────────────────────────────────
+
+
+@router.post("/register", response_model=AuthTokens, status_code=status.HTTP_201_CREATED)
+async def register(
+    body: _RegisterRequest,
+    db: AsyncSession = Depends(get_session),
+) -> AuthTokens:
+    """Create a new account and return JWT tokens."""
+    existing = await db.execute(select(User).where(User.email == body.email))
+    if existing.scalar_one_or_none() is not None:
+        raise HTTPException(status.HTTP_409_CONFLICT, "Email already registered")
+
+    user = User(
+        id=str(uuid.uuid4()),
+        email=body.email,
+        name=body.name,
+        surname=body.surname,
+        password_hash=_hash_password(body.password),
+        tier="free",
+        encryption_key=Fernet.generate_key().decode(),
+    )
+    db.add(user)
+    await db.flush()
+
+    plain_token = str(uuid.uuid4())
+    expires_at = datetime.now(timezone.utc) + timedelta(
+        days=settings.JWT_REFRESH_TOKEN_EXPIRE_DAYS
+    )
+    rt = RefreshToken(
+        user_id=user.id,
+        token_hash=_hash_token(plain_token),
+        expires_at=expires_at,
+    )
+    db.add(rt)
+    await db.commit()
+
+    access_token, expires_at_ms = _make_access_token(user.id, user.email, user.tier)
+    return AuthTokens(
+        access_token=access_token,
+        refresh_token=plain_token,
+        expires_at=expires_at_ms,
+    )
+
+
+@router.post("/login", response_model=AuthTokens)
+async def login(
+    body: _LoginRequest,
+    db: AsyncSession = Depends(get_session),
+) -> AuthTokens:
+    """Validate credentials and return JWT tokens."""
+    result = await db.execute(select(User).where(User.email == body.email))
+    user = result.scalar_one_or_none()
+    if user is None or not _verify_password(body.password, user.password_hash):
+        raise HTTPException(status.HTTP_401_UNAUTHORIZED, "Invalid credentials")
+
+    # Fetch live tier for the JWT claim
+    tier = await _get_live_tier(db, user.id)
+
+    plain_token = str(uuid.uuid4())
+    expires_at = datetime.now(timezone.utc) + timedelta(
+        days=settings.JWT_REFRESH_TOKEN_EXPIRE_DAYS
+    )
+    rt = RefreshToken(
+        user_id=user.id,
+        token_hash=_hash_token(plain_token),
+        expires_at=expires_at,
+    )
+    db.add(rt)
+    await db.commit()
+
+    access_token, expires_at_ms = _make_access_token(user.id, user.email, tier)
+    return AuthTokens(
+        access_token=access_token,
+        refresh_token=plain_token,
+        expires_at=expires_at_ms,
+    )
+
+
+@router.post("/refresh", response_model=AuthTokens)
+async def refresh(
+    body: _RefreshRequest,
+    db: AsyncSession = Depends(get_session),
+) -> AuthTokens:
+    """Rotate a refresh token and return a new token pair."""
+    token_hash = _hash_token(body.refresh_token)
+    result = await db.execute(
+        select(RefreshToken).where(RefreshToken.token_hash == token_hash)
+    )
+    rt = result.scalar_one_or_none()
+
+    now = datetime.now(timezone.utc)
+    if rt is None or rt.expires_at.replace(tzinfo=timezone.utc) < now:
+        raise HTTPException(status.HTTP_401_UNAUTHORIZED, "Invalid or expired refresh token")
+
+    await db.delete(rt)
+
+    user_result = await db.execute(select(User).where(User.id == rt.user_id))
+    user = user_result.scalar_one_or_none()
+    if user is None:
+        raise HTTPException(status.HTTP_401_UNAUTHORIZED, "User not found")
+
+    # Fetch live tier for the new JWT
+    tier = await _get_live_tier(db, user.id)
+
+    plain_token = str(uuid.uuid4())
+    new_expires = now + timedelta(days=settings.JWT_REFRESH_TOKEN_EXPIRE_DAYS)
+    new_rt = RefreshToken(
+        user_id=user.id,
+        token_hash=_hash_token(plain_token),
+        expires_at=new_expires,
+    )
+    db.add(new_rt)
+    await db.commit()
+
+    access_token, expires_at_ms = _make_access_token(user.id, user.email, tier)
+    return AuthTokens(
+        access_token=access_token,
+        refresh_token=plain_token,
+        expires_at=expires_at_ms,
+    )
+
+
+@router.get("/me", response_model=UserProfile)
+async def me(current_user: UserProfile = Depends(get_current_user)) -> UserProfile:
+    """Return the profile for the authenticated user."""
+    return current_user
+
+
+@router.put("/me", response_model=UserProfile)
+async def update_profile(
+    body: _UpdateProfileRequest,
+    current_user: UserProfile = Depends(get_current_user),
+    db: AsyncSession = Depends(get_session),
+) -> UserProfile:
+    """Update the authenticated user's name and surname."""
+    result = await db.execute(select(User).where(User.id == current_user.id))
+    user = result.scalar_one()
+
+    if body.name is not None:
+        user.name = body.name
+    if body.surname is not None:
+        user.surname = body.surname
+
+    await db.commit()
+    await db.refresh(user)
+
+    return UserProfile(
+        id=user.id,
+        email=user.email,
+        name=user.name,
+        surname=user.surname,
+        tier=current_user.tier,
+    )

services/auth/app/verify.py

@@ -0,0 +1,66 @@
+"""ForwardAuth verification endpoint for Traefik.
+
+Traefik calls GET /api/v1/auth/verify on every request to a protected
+service.  This endpoint validates the JWT from the Authorization header
+and returns identity headers that Traefik injects into downstream requests.
+
+Downstream services NEVER validate JWTs themselves — they trust the
+X-User-Id, X-User-Email, X-User-Tier headers injected by Traefik.
+"""
+
+from __future__ import annotations
+
+from fastapi import APIRouter, Request, Response
+from fastapi import status as http_status
+from jose import JWTError, jwt
+from sqlalchemy import select
+
+from shared.config import settings
+from shared.db import async_session
+from shared.models import Subscription
+
+from app.config import auth_settings
+
+router = APIRouter(tags=["auth"])
+
+
+@router.get("/auth/verify")
+async def verify(request: Request) -> Response:
+    """Validate JWT and return identity headers for Traefik ForwardAuth.
+
+    Returns 200 with X-User-* headers on success, 401 on failure.
+    Traefik copies response headers to the downstream request.
+    """
+    auth_header = request.headers.get("Authorization", "")
+    if not auth_header.startswith("Bearer "):
+        return Response(status_code=http_status.HTTP_401_UNAUTHORIZED)
+
+    token = auth_header[7:]  # strip "Bearer "
+
+    try:
+        payload = jwt.decode(
+            token, auth_settings.JWT_PUBLIC_KEY, algorithms=["RS256"]
+        )
+        user_id: str | None = payload.get("sub")
+        email: str | None = payload.get("email")
+        if not user_id or not email:
+            return Response(status_code=http_status.HTTP_401_UNAUTHORIZED)
+    except JWTError:
+        return Response(status_code=http_status.HTTP_401_UNAUTHORIZED)
+
+    # Live tier lookup from subscriptions table
+    async with async_session() as db:
+        result = await db.execute(
+            select(Subscription.tier).where(Subscription.user_id == user_id)
+        )
+        default_tier = "power" if settings.ENV == "dev" else "free"
+        tier: str = result.scalar_one_or_none() or default_tier
+
+    return Response(
+        status_code=http_status.HTTP_200_OK,
+        headers={
+            "X-User-Id": user_id,
+            "X-User-Email": email,
+            "X-User-Tier": tier,
+        },
+    )

services/auth/requirements.txt

@@ -0,0 +1,11 @@
+fastapi>=0.115.0
+uvicorn[standard]>=0.34.0
+gunicorn>=22.0.0
+pydantic>=2.10.0
+pydantic-settings>=2.7.0
+python-jose[cryptography]>=3.3.0
+sqlalchemy>=2.0.0
+asyncpg>=0.30.0
+bcrypt>=4.2.0
+cryptography>=42.0.0
+python-dotenv>=1.0.0

services/batch-agent/Dockerfile

@@ -0,0 +1,36 @@
+# ── builder ──────────────────────────────────────────────────────────────────
+FROM python:3.12-slim AS builder
+
+WORKDIR /build
+
+COPY services/batch-agent/requirements.txt ./requirements.txt
+RUN pip install --upgrade pip && \
+    pip install --no-cache-dir --prefix=/install -r requirements.txt
+
+# ── runtime ──────────────────────────────────────────────────────────────────
+FROM python:3.12-slim AS runtime
+
+RUN addgroup --system appgroup && adduser --system --ingroup appgroup appuser
+
+WORKDIR /app
+
+COPY --from=builder /install /usr/local
+
+# Shared module
+COPY shared/ shared/
+
+# Service source
+COPY services/batch-agent/app/ app/
+
+RUN chown -R appuser:appgroup /app
+
+USER appuser
+
+EXPOSE 8000
+
+# Batch runs are long-lived — use a longer timeout than chat (300s vs 120s)
+CMD ["gunicorn", "app.main:app", \
+     "-k", "uvicorn.workers.UvicornWorker", \
+     "--bind", "0.0.0.0:8000", \
+     "--workers", "2", \
+     "--timeout", "300"]

services/batch-agent/README.md

@@ -0,0 +1,23 @@
+# Batch Agent Service
+
+Owns: agent_runner, journey builder, filesystem_agent, integrations (Gmail, MS Graph).
+
+## Tables owned
+- `local_agent_configs`
+- `cloud_agent_configs`
+- `agent_run_logs`
+
+## Endpoints
+- `GET /agents/catalog`
+- `POST /agents/can-create`
+- `POST /agents/trigger`
+- `GET /agents/{id}/history`
+
+## Redis channels
+- Subscribe: `batch:request:{user_id}`
+- Publish: `ws:out:{user_id}` (journey replies + tool calls)
+- BRPOP: `tool:result:{call_id}` (30s timeout)
+- SET+EX: `journey:{user_id}` (session state, TTL 1800s)
+
+## TODO
+- [ ] Integrate Langfuse tracing (reuse `services/chat/app/tracing.py` pattern — `trace_span()`, `get_langfuse_callback()`, prompt management). Each batch agent run should create a trace with input/output, link prompts, and pass the LangChain `CallbackHandler` to LLM calls.

services/batch-agent/app/agent_runner.py

@@ -0,0 +1,884 @@
+"""Agent run orchestrator — adapted for Batch Agent Service.
+
+Key changes from monolith app/core/agent_runner.py:
+  - No DeviceConnectionManager — tool calls go through Redis ws_context.
+  - set_current_user / clear_current_user replace set_client_executor.
+  - run_local_agent accepts a serialized dict (from Redis / REST) instead
+    of SQLAlchemy model objects.
+  - _finalize_run writes to PostgreSQL via shared.db.async_session.
+  - Cloud agent import path changed to app.integrations.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import logging
+import uuid
+from datetime import datetime, timedelta, timezone
+from typing import Any
+
+from langchain_core.messages import AIMessage, HumanMessage, SystemMessage, ToolMessage
+from sqlalchemy import select
+
+from app.agents.filesystem_agent import FILESYSTEM_TOOLS
+from app.agents.note_agent import NOTE_TOOLS
+from app.agents.project_agent import PROJECT_TOOLS
+from app.agents.task_agent import TASK_TOOLS
+from app.agents.timeline_agent import TIMELINE_TOOLS
+from app.llm import get_llm
+from app.ws_context import execute_on_client, set_current_user, clear_current_user
+from shared.db import async_session
+from shared.models import AgentRunLog, CloudAgentConfig, LocalAgentConfig
+from shared.redis import redis_client, ws_out_channel
+
+logger = logging.getLogger(__name__)
+
+# ── Concurrency guard ─────────────────────────────────────────────────────
+_running_agents: set[str] = set()
+
+
+def is_agent_running(agent_id: str) -> bool:
+    return agent_id in _running_agents
+
+
+# ── Timeouts ───────────────────────────────────────────────────────────────
+_TOOL_CALL_TIMEOUT: int = 30
+_MAX_PROCESSING_STEPS: int = 12
+_MAX_SCAN_DEPTH: int = 5
+
+# ── Data-type to tool mapping ─────────────────────────────────────────────
+_DATA_TYPE_TOOLS: dict[str, list[Any]] = {
+    "tasks": TASK_TOOLS,
+    "notes": NOTE_TOOLS,
+    "timelines": TIMELINE_TOOLS,
+}
+
+# ── Step 1: Classification prompt ─────────────────────────────────────────
+
+_DOMAIN_DESCRIPTIONS: dict[str, str] = {
+    "tasks": (
+        "Action items, to-dos, deliverables — anything that describes work to be done, "
+        "assigned to someone, or tracked with a due date or status."
+    ),
+    "notes": (
+        "Documentation, meeting notes, summaries, reference material — "
+        "written content meant to be read and referenced rather than acted on."
+    ),
+    "timelines": (
+        "Project milestones, deadlines, scheduled events — "
+        "specific dates that mark a point in the progress of a project."
+    ),
+    "projects": (
+        "High-level project entities — only relevant if the file clearly introduces "
+        "a new project or updates the scope of an existing one."
+    ),
+}
+
+_STEP1_SYSTEM_PROMPT = """\
+You are a file classifier for a freelance project management tool.
+
+Your job is to match a file to an existing project and identify which data domains to extract.
+
+## Project matching rules (STRICT — follow in order)
+
+1. Search the file content for any mention of a project name, client name, acronym, or topic
+   that overlaps with the existing projects listed below.
+2. The match does NOT need to be exact — partial name, abbreviation, or topic similarity is enough.
+3. STRONGLY PREFER matching an existing project. Only return "new" as an absolute last resort
+   when the file has zero meaningful connection to any listed project.
+4. When in doubt, pick the closest match from the list.
+
+## Response format
+
+Respond ONLY with a JSON object — no markdown, no explanation:
+
+{{"project_id": "<exact id from the list below, or new>", "new_project_name": "<concise 2-5 word name, only when project_id is new>", "domains": ["tasks", "notes"]}}
+
+## Domain definitions (only consider domains in the allowed list)
+
+{domain_definitions}
+
+## Existing projects
+
+{projects_list}
+"""
+
+# ── Step 2: Processing prompt ─────────────────────────────────────────────
+
+_PROCESSING_SYSTEM_PROMPT = """\
+You are a data extraction assistant for a freelance project management tool.
+
+Your task: extract structured data from the file content and persist it using the available tools.
+
+## Mandatory process — follow this order for EVERY item you extract
+
+1. READ the existing records listed below for the relevant domain.
+2. SEARCH for a match by title, topic, or semantic similarity.
+3. If a match exists → call the update_* tool with the existing record's id.
+4. If no match exists → call the create_* tool and set isAiSuggested=1.
+
+NEVER call create_* without first checking the existing records.
+NEVER duplicate a record that already exists under a different wording.
+
+## Existing records (source of truth)
+
+{existing_context}
+
+## Context
+
+Project: {project_context}
+Domains to extract: {data_types}
+
+{custom_prompt_section}
+"""
+
+# ── Cloud processing prompt ───────────────────────────────────────────────
+
+_CLOUD_PROCESSING_PROMPT = """\
+You are a data extraction and management assistant for a freelance project
+management tool.
+
+Available tools:
+  Filesystem : read_file_content, list_directory, get_file_metadata
+  Tasks      : list_tasks, create_task, update_task, add_task_comment
+  Notes      : list_notes, get_note, create_note, update_note
+  Timelines  : list_timelines, create_timeline, update_timeline
+  Projects   : list_all_projects, get_project, create_project, update_project
+
+Your task:
+1. Read the full content of each file below using read_file_content.
+2. For each piece of information found, ALWAYS try to match and update an
+   existing record before creating a new one.
+3. ONLY act on these entity types: {data_types}.
+4. Do NOT invent data. Only extract what is clearly present in the files.
+5. If a file contains no relevant data for the target entity types, skip it.
+
+{project_context}
+
+Files to process:
+{file_list}
+
+{custom_prompt_section}
+
+After processing all files, respond with a brief summary of what you updated
+and what you created.
+"""
+
+
+# ── LLM tool-calling loop ─────────────────────────────────────────────────
+
+
+def _as_text(content: Any) -> str:
+    if content is None:
+        return ""
+    if isinstance(content, str):
+        return content
+    if isinstance(content, list):
+        parts: list[str] = []
+        for item in content:
+            if isinstance(item, str):
+                parts.append(item)
+            elif isinstance(item, dict):
+                text = item.get("text")
+                if isinstance(text, str):
+                    parts.append(text)
+        return "".join(parts)
+    return str(content)
+
+
+async def _run_agent_with_tools(
+    *,
+    system_prompt: str,
+    user_message: str,
+    tools: list[Any],
+    max_steps: int,
+) -> str:
+    """Run an LLM agent with tool-calling, returning the final text response."""
+    llm = get_llm()
+    llm_with_tools = llm.bind_tools(tools)
+    messages: list[Any] = [
+        SystemMessage(content=system_prompt),
+        HumanMessage(content=user_message),
+    ]
+
+    tool_map = {tool_def.name: tool_def for tool_def in tools}
+
+    for _ in range(max_steps):
+        response: AIMessage = await llm_with_tools.ainvoke(messages)
+        messages.append(response)
+
+        if not response.tool_calls:
+            return _as_text(response.content)
+
+        for call in response.tool_calls:
+            call_id = str(call.get("id", ""))
+            call_name = str(call.get("name", ""))
+            call_args = call.get("args", {})
+            logger.info(
+                "agent_runner: tool_call name=%s args=%s",
+                call_name,
+                json.dumps(call_args, ensure_ascii=True)[:800],
+            )
+
+            tool_fn = tool_map.get(call_name)
+            if tool_fn is None:
+                tool_output = f"Unknown tool: {call_name}"
+            else:
+                tool_output = await tool_fn.ainvoke(call_args)
+
+            logger.info(
+                "agent_runner: tool_result name=%s output=%s",
+                call_name,
+                str(tool_output)[:200],
+            )
+            messages.append(ToolMessage(content=str(tool_output), tool_call_id=call["id"]))
+
+    final = await llm.ainvoke(messages)
+    return _as_text(final.content)
+
+
+# ── Tool list builder ─────────────────────────────────────────────────────
+
+
+def _build_processing_tools(data_types: list[str]) -> list[Any]:
+    tools: list[Any] = list(FILESYSTEM_TOOLS)
+    for dt in data_types:
+        dt_tools = _DATA_TYPE_TOOLS.get(dt)
+        if dt_tools:
+            tools.extend(dt_tools)
+    return tools
+
+
+# ── Code-based directory scanner ─────────────────────────────────────────
+
+
+async def _scan_directories(
+    paths: list[str],
+    extensions: list[str],
+    last_run_at: datetime | None,
+) -> list[str]:
+    all_files: list[str] = []
+    ext_set = {e.lstrip(".").lower() for e in extensions} if extensions else set()
+
+    async def _walk(path: str, depth: int) -> None:
+        if depth > _MAX_SCAN_DEPTH:
+            return
+        try:
+            result = await execute_on_client(action="list_directory", data={"path": path})
+        except Exception as exc:
+            logger.warning("agent_runner: list_directory failed %r: %s", path, exc)
+            return
+        for entry in result.get("entries", []):
+            entry_path = entry.get("path", "")
+            if not entry_path:
+                continue
+            if entry.get("type") == "directory":
+                await _walk(entry_path, depth + 1)
+            elif entry.get("type") == "file":
+                if ext_set:
+                    dot_pos = entry_path.rfind(".")
+                    file_ext = entry_path[dot_pos + 1:].lower() if dot_pos != -1 else ""
+                    if file_ext not in ext_set:
+                        continue
+                all_files.append(entry_path)
+
+    for root in paths:
+        await _walk(root, depth=0)
+
+    if last_run_at is None:
+        return all_files
+
+    last_run_ms = int(last_run_at.timestamp() * 1000)
+    filtered: list[str] = []
+    for file_path in all_files:
+        try:
+            meta = await execute_on_client(action="get_file_metadata", data={"path": file_path})
+            modified_at = meta.get("modifiedAt")
+            if modified_at is None:
+                filtered.append(file_path)
+                continue
+            if isinstance(modified_at, (int, float)):
+                mod_ms = int(modified_at)
+            else:
+                mod_ms = int(datetime.fromisoformat(str(modified_at)).timestamp() * 1000)
+            if mod_ms > last_run_ms:
+                filtered.append(file_path)
+        except Exception:
+            filtered.append(file_path)
+
+    return filtered
+
+
+# ── Code-based entity fetchers ────────────────────────────────────────────
+
+
+async def _fetch_projects() -> list[dict]:
+    try:
+        result = await execute_on_client(action="select", table="projects")
+        return result.get("rows", [])
+    except Exception as exc:
+        logger.warning("agent_runner: failed to fetch projects: %s", exc)
+        return []
+
+
+_DOMAIN_TABLE: dict[str, str] = {
+    "tasks": "tasks",
+    "notes": "notes",
+    "timelines": "timelines",
+    "projects": "projects",
+}
+
+
+async def _fetch_domain_entities(domain: str, project_id: str) -> list[dict]:
+    table = _DOMAIN_TABLE.get(domain)
+    if not table:
+        return []
+    filters: dict[str, Any] = {}
+    if project_id != "standalone" and domain != "projects":
+        filters["projectId"] = project_id
+    try:
+        result = await execute_on_client(
+            action="select",
+            table=table,
+            filters=filters if filters else None,
+        )
+        return result.get("rows", [])
+    except Exception as exc:
+        logger.warning("agent_runner: failed to fetch %s: %s", domain, exc)
+        return []
+
+
+def _format_entities_for_context(domain: str, rows: list[dict]) -> str:
+    if not rows:
+        return f"No existing {domain}."
+    lines: list[str] = []
+    for r in rows:
+        if domain == "tasks":
+            desc = r.get("description") or ""
+            desc_part = f" — {desc[:120]}" if desc else ""
+            assignee = r.get("assignee") or r.get("assignees") or ""
+            due = r.get("dueDate") or r.get("due_date") or ""
+            meta = ", ".join(filter(None, [
+                f"priority: {r.get('priority', '')}" if r.get("priority") else "",
+                f"assignee: {assignee}" if assignee else "",
+                f"due: {due}" if due else "",
+            ]))
+            lines.append(
+                f"  - [{r.get('status', '?')}] {r.get('title', '')}{desc_part}"
+                f" ({meta}, id: {r['id']})"
+            )
+        elif domain == "notes":
+            snippet = (r.get("content") or "")[:200].replace("\n", " ")
+            snippet_part = f"\n      Preview: {snippet}" if snippet else ""
+            lines.append(
+                f"  - {r.get('title', '')} (id: {r['id']}){snippet_part}"
+            )
+        elif domain == "timelines":
+            lines.append(
+                f"  - {r.get('title', '')} date={r.get('date', '')} (id: {r['id']})"
+            )
+        elif domain == "projects":
+            summary = (r.get("aiSummary") or r.get("ai_summary") or "")[:120]
+            summary_part = f" — {summary}" if summary else ""
+            lines.append(
+                f"  - {r.get('name', '')} [{r.get('status', '')}]{summary_part}"
+                f" (id: {r['id']})"
+            )
+    return f"Existing {domain}:\n" + "\n".join(lines)
+
+
+# ── Step 1: LLM file classifier ───────────────────────────────────────────
+
+
+async def _classify_file(
+    file_path: str,
+    file_content: str,
+    projects: list[dict],
+    config_data_types: list[str],
+) -> tuple[str, list[str], str | None]:
+    fallback: tuple[str, list[str], str | None] = ("new", list(config_data_types), None)
+
+    if not file_content.strip():
+        return fallback
+
+    valid_project_ids = {p["id"] for p in projects}
+
+    def _fmt_project(p: dict) -> str:
+        summary = (p.get("aiSummary") or p.get("ai_summary") or "").strip()
+        summary_part = f" — {summary[:100]}" if summary else ""
+        return f"  - id={p['id']} | name={p.get('name', '')} | status={p.get('status', '')}{summary_part}"
+
+    projects_list = "\n".join(_fmt_project(p) for p in projects) or "  (none yet)"
+
+    domain_definitions = "\n".join(
+        f"  - {d}: {_DOMAIN_DESCRIPTIONS[d]}"
+        for d in config_data_types
+        if d in _DOMAIN_DESCRIPTIONS
+    )
+
+    system = _STEP1_SYSTEM_PROMPT.format(
+        domain_definitions=domain_definitions,
+        projects_list=projects_list,
+    )
+
+    llm = get_llm()
+    try:
+        response = await llm.ainvoke([
+            SystemMessage(content=system),
+            HumanMessage(content=f"File: {file_path}\n\nContent:\n{file_content[:4000]}"),
+        ])
+        raw = _as_text(response.content).strip()
+        if raw.startswith("```"):
+            raw = raw.split("```")[1]
+            if raw.startswith("json"):
+                raw = raw[4:]
+        parsed = json.loads(raw.strip())
+        raw_project_id: str = str(parsed.get("project_id") or "new")
+        project_id = raw_project_id if raw_project_id in valid_project_ids else "new"
+        new_project_name: str | None = (
+            str(parsed["new_project_name"]).strip() or None
+            if project_id == "new" and parsed.get("new_project_name")
+            else None
+        )
+        domains: list[str] = [
+            d for d in parsed.get("domains", [])
+            if d in config_data_types
+        ]
+        if not domains:
+            domains = list(config_data_types)
+        return project_id, domains, new_project_name
+    except Exception as exc:
+        logger.warning(
+            "agent_runner: step1 classification failed for %r: %s", file_path, exc
+        )
+        return fallback
+
+
+# ── Local agent runner (two-step per file) ────────────────────────────────
+
+
+async def run_local_agent(user_id: str, trigger_data: dict[str, Any]) -> None:
+    """Execute a local directory agent run.
+
+    In the microservice world, trigger_data is a serialized dict from
+    the REST route (forwarded via Redis), containing the agent config
+    fields and run_context.
+
+    set_current_user() must be called BEFORE this function.
+    """
+    run_context: dict = trigger_data.get("run_context", {})
+    agent_id = run_context.get("agent_id", str(uuid.uuid4()))
+    run_id = run_context.get("run_id")
+
+    _running_agents.add(agent_id)
+
+    # Extract config from trigger payload
+    directory_paths: list[str] = trigger_data.get("directory_paths", [])
+    if not directory_paths:
+        directory = trigger_data.get("directory", "")
+        if directory:
+            directory_paths = [directory]
+
+    data_types: list[str] = trigger_data.get("data_types", [])
+    file_extensions: list[str] = trigger_data.get("file_extensions", [])
+    prompt_template: str = trigger_data.get("prompt_template", "")
+    last_run_at_raw = trigger_data.get("last_run_at")
+    last_run_at: datetime | None = None
+    if last_run_at_raw:
+        if isinstance(last_run_at_raw, str):
+            last_run_at = datetime.fromisoformat(last_run_at_raw)
+        elif isinstance(last_run_at_raw, (int, float)):
+            last_run_at = datetime.fromtimestamp(last_run_at_raw / 1000, tz=timezone.utc)
+
+    errors: list[str] = []
+    items_processed = 0
+    items_created = 0
+
+    custom_section = (
+        f"User instructions:\n{prompt_template}"
+        if prompt_template
+        else ""
+    )
+
+    # Create or load run log
+    run_log_id = run_id
+    if not run_log_id:
+        async with async_session() as db:
+            run_log = AgentRunLog(
+                agent_id=agent_id,
+                agent_type="local",
+                user_id=user_id,
+                status="running",
+            )
+            db.add(run_log)
+            await db.commit()
+            await db.refresh(run_log)
+            run_log_id = run_log.id
+
+    try:
+        # ── Scan directories ─────────────────────────────────────────
+        logger.info("agent_runner: run=%s scanning directories user=%s", run_log_id, user_id)
+        file_paths = await _scan_directories(
+            paths=directory_paths,
+            extensions=file_extensions,
+            last_run_at=last_run_at,
+        )
+        logger.info(
+            "agent_runner: run=%s found %d file(s) after filtering", run_log_id, len(file_paths)
+        )
+
+        if not file_paths:
+            await _finalize_run(run_log_id, status="success", items_processed=0, items_created=0)
+            return
+
+        # ── Fetch all projects once ──────────────────────────────────
+        projects = await _fetch_projects()
+
+        for file_path in file_paths:
+            try:
+                file_result = await execute_on_client(
+                    action="read_file_content", data={"path": file_path}
+                )
+                file_content: str = file_result.get("content", "")
+                if not file_content:
+                    continue
+
+                items_processed += 1
+
+                # Step 1 — classify file
+                project_id, domains, new_project_name = await _classify_file(
+                    file_path=file_path,
+                    file_content=file_content,
+                    projects=projects,
+                    config_data_types=data_types,
+                )
+
+                # Step 2 — resolve project_id, fetch entities, process
+                if project_id == "new":
+                    proj_name = new_project_name or "Untitled Project"
+                    try:
+                        proj_result = await execute_on_client(
+                            action="insert",
+                            table="projects",
+                            data={"name": proj_name, "clientId": None},
+                        )
+                        created = proj_result.get("row", {})
+                        effective_project_id = created.get("id", "standalone")
+                        if "id" in created:
+                            projects.append(created)
+                    except Exception as exc:
+                        logger.warning("agent_runner: run=%s create project failed: %s", run_log_id, exc)
+                        effective_project_id = "standalone"
+                        proj_name = "unknown"
+                    project_context = (
+                        f"Project: {proj_name} (id: {effective_project_id}). "
+                        "Always set projectId to this id on every record you create."
+                    )
+                else:
+                    effective_project_id = project_id
+                    proj = next((p for p in projects if p["id"] == project_id), None)
+                    proj_name = proj.get("name", project_id) if proj else project_id
+                    project_context = (
+                        f"Project: {proj_name} (id: {project_id}). "
+                        "Always set projectId to this id on every record you create."
+                    )
+
+                domains = [d for d in domains if d != "projects"]
+
+                existing_blocks: list[str] = []
+                for domain in domains:
+                    rows = await _fetch_domain_entities(domain, effective_project_id)
+                    existing_blocks.append(_format_entities_for_context(domain, rows))
+
+                existing_context = "\n\n".join(existing_blocks)
+
+                system_prompt = _PROCESSING_SYSTEM_PROMPT.format(
+                    existing_context=existing_context,
+                    project_context=project_context,
+                    data_types=", ".join(domains),
+                    custom_prompt_section=custom_section,
+                )
+
+                processing_tools = _build_processing_tools(domains)
+
+                result_text = await _run_agent_with_tools(
+                    system_prompt=system_prompt,
+                    user_message=(
+                        f"Process this file and extract relevant information.\n\n"
+                        f"File: {file_path}\n\nContent:\n{file_content}"
+                    ),
+                    tools=processing_tools,
+                    max_steps=_MAX_PROCESSING_STEPS,
+                )
+                logger.info(
+                    "agent_runner: run=%s file=%r result=%s",
+                    run_log_id, file_path, result_text[:200],
+                )
+
+            except Exception as exc:
+                errors.append(f"Error processing '{file_path}': {exc}")
+                logger.error("agent_runner: run=%s file=%r failed: %s", run_log_id, file_path, exc)
+
+    except Exception as exc:
+        errors.append(f"Agent run failed: {exc}")
+        logger.error("agent_runner: run=%s failed: %s", run_log_id, exc)
+    finally:
+        _running_agents.discard(agent_id)
+
+    # ── Finalise ────────────────────────────────────────────────────
+    if errors and items_processed == 0:
+        final_status = "error"
+    elif errors:
+        final_status = "partial"
+    else:
+        final_status = "success"
+
+    await _finalize_run(
+        run_log_id,
+        status=final_status,
+        items_processed=items_processed,
+        items_created=items_created,
+        errors=errors,
+    )
+
+    # Notify Electron that the run is complete via Redis
+    if run_context:
+        try:
+            channel = ws_out_channel(user_id)
+            await redis_client.publish(channel, json.dumps({
+                "type": "run_complete",
+                "run_context": run_context,
+                "status": final_status,
+            }))
+        except Exception as exc:
+            logger.warning("agent_runner: run=%s failed to send run_complete: %s", run_log_id, exc)
+
+
+# ── Cloud agent runner ─────────────────────────────────────────────────────
+
+_CLOUD_DEFAULT_LOOKBACK_DAYS: int = 7
+
+
+async def run_cloud_agent(user_id: str, config_id: str) -> None:
+    """Execute a cloud connector agent run.
+
+    Loads the CloudAgentConfig from DB, decrypts OAuth tokens, fetches
+    messages from the provider, and runs LLM extraction.
+
+    set_current_user() must be called BEFORE this function.
+    """
+    from app.integrations import decrypt_token, encrypt_token, get_provider
+
+    async with async_session() as db:
+        result = await db.execute(
+            select(CloudAgentConfig).where(CloudAgentConfig.id == config_id)
+        )
+        config = result.scalar_one_or_none()
+        if config is None:
+            logger.error("agent_runner: cloud config %s not found", config_id)
+            return
+
+        # Create run log
+        run_log = AgentRunLog(
+            agent_id=config.id,
+            agent_type="cloud",
+            user_id=user_id,
+            status="running",
+        )
+        db.add(run_log)
+        await db.commit()
+        await db.refresh(run_log)
+        run_log_id = run_log.id
+
+    # ── Decrypt OAuth token ────────────────────────────────────────
+    if not config.oauth_token_encrypted:
+        await _finalize_run(
+            run_log_id,
+            status="error",
+            errors=[f"No OAuth token stored for cloud agent '{config.name}'"],
+        )
+        return
+
+    try:
+        credentials_info = decrypt_token(config.oauth_token_encrypted)
+    except ValueError as exc:
+        await _finalize_run(
+            run_log_id,
+            status="error",
+            errors=[f"Failed to decrypt OAuth token: {exc}"],
+        )
+        return
+
+    # ── Instantiate provider ──────────────────────────────────────
+    try:
+        provider = get_provider(config.provider, credentials_info)
+    except ValueError as exc:
+        await _finalize_run(run_log_id, status="error", errors=[str(exc)])
+        return
+
+    # ── Fetch messages ────────────────────────────────────────────
+    since: datetime | None = config.last_run_at
+    if since is None:
+        since = datetime.now(timezone.utc) - timedelta(days=_CLOUD_DEFAULT_LOOKBACK_DAYS)
+    if since.tzinfo is None:
+        since = since.replace(tzinfo=timezone.utc)
+
+    errors: list[str] = []
+    items_processed = 0
+
+    try:
+        if config.provider == "gmail":
+            raw_messages = await provider.fetch_messages(
+                filter_config=config.filter_config,
+                since=since,
+            )
+        elif config.provider == "outlook":
+            raw_messages = await provider.fetch_emails(
+                filter_config=config.filter_config,
+                since=since,
+            )
+        elif config.provider == "teams":
+            raw_messages = await provider.fetch_messages(
+                filter_config=config.filter_config,
+                since=since,
+            )
+        else:
+            raw_messages = []
+    except RuntimeError as exc:
+        await _finalize_run(
+            run_log_id,
+            status="error",
+            errors=[f"Provider fetch failed: {exc}"],
+            update_config_last_run=True,
+            config_id=config.id,
+            config_type="cloud",
+        )
+        return
+
+    logger.info(
+        "agent_runner: cloud agent %s fetched %d item(s) from %s",
+        config.id, len(raw_messages), config.provider,
+    )
+
+    # ── Extract + insert via LLM ─────────────────────────────────
+    try:
+        processing_tools = _build_processing_tools(config.data_types)
+        custom_section = (
+            f"User instructions:\n{config.prompt_template}"
+            if config.prompt_template
+            else ""
+        )
+
+        for msg in raw_messages:
+            content_text = msg.as_text
+            if not content_text:
+                continue
+            items_processed += 1
+
+            processing_prompt = _CLOUD_PROCESSING_PROMPT.format(
+                data_types=", ".join(config.data_types),
+                project_context="Determine the appropriate project from the message context.",
+                file_list=f"Message from {config.provider} (id: {msg.id})",
+                custom_prompt_section=custom_section,
+            )
+
+            try:
+                await _run_agent_with_tools(
+                    system_prompt=processing_prompt,
+                    user_message=f"Process this message content:\n\n{content_text[:8000]}",
+                    tools=processing_tools,
+                    max_steps=_MAX_PROCESSING_STEPS,
+                )
+            except Exception as exc:
+                errors.append(f"LLM processing error for message {msg.id!r}: {exc}")
+    except Exception as exc:
+        errors.append(f"Agent run failed: {exc}")
+
+    # ── Persist refreshed token ───────────────────────────────────
+    refreshed = getattr(provider, "refreshed_credentials", None)
+    if refreshed:
+        try:
+            new_encrypted = encrypt_token(refreshed)
+            async with async_session() as db:
+                cfg_result = await db.execute(
+                    select(CloudAgentConfig).where(CloudAgentConfig.id == config.id)
+                )
+                cfg_row = cfg_result.scalar_one_or_none()
+                if cfg_row:
+                    cfg_row.oauth_token_encrypted = new_encrypted
+                    await db.commit()
+        except Exception as exc:
+            logger.warning("agent_runner: failed to persist refreshed token: %s", exc)
+
+    # ── Finalise ──────────────────────────────────────────────────
+    if errors and items_processed == 0:
+        final_status = "error"
+    elif errors:
+        final_status = "partial"
+    else:
+        final_status = "success"
+
+    await _finalize_run(
+        run_log_id,
+        status=final_status,
+        items_processed=items_processed,
+        items_created=0,
+        errors=errors,
+        update_config_last_run=True,
+        config_id=config.id,
+        config_type="cloud",
+    )
+
+
+# ── Internal helper ─────────────────────────────────────────────────────────
+
+
+async def _finalize_run(
+    run_log_id: int | str,
+    *,
+    status: str,
+    items_processed: int = 0,
+    items_created: int = 0,
+    errors: list[str] | None = None,
+    update_config_last_run: bool = False,
+    config_id: str | None = None,
+    config_type: str | None = None,
+) -> None:
+    """Persist the run outcome and optionally update last_run_at on the config."""
+    now = datetime.now(timezone.utc)
+    try:
+        async with async_session() as db:
+            result = await db.execute(
+                select(AgentRunLog).where(AgentRunLog.id == run_log_id)
+            )
+            managed = result.scalar_one_or_none()
+            if managed is None:
+                logger.warning("agent_runner: run_log %s not found for finalization", run_log_id)
+                return
+
+            managed.status = status
+            managed.items_processed = items_processed
+            managed.items_created = items_created
+            managed.errors = errors or []
+            managed.completed_at = now
+
+            if update_config_last_run and config_id:
+                if config_type == "local":
+                    cfg_result = await db.execute(
+                        select(LocalAgentConfig).where(LocalAgentConfig.id == config_id)
+                    )
+                    cfg = cfg_result.scalar_one_or_none()
+                    if cfg:
+                        cfg.last_run_at = now
+                elif config_type == "cloud":
+                    cfg_result = await db.execute(
+                        select(CloudAgentConfig).where(CloudAgentConfig.id == config_id)
+                    )
+                    cfg = cfg_result.scalar_one_or_none()
+                    if cfg:
+                        cfg.last_run_at = now
+
+            await db.commit()
+    except Exception as exc:
+        logger.error("agent_runner: failed to finalize run_log=%s: %s", run_log_id, exc)

services/batch-agent/app/agents/__init__.py

@@ -0,0 +1 @@
+"""Batch Agent Service domain agents and filesystem tools."""

services/batch-agent/app/agents/filesystem_agent.py

@@ -0,0 +1,83 @@
+"""Filesystem agent — tools for reading local directories and files on Electron.
+
+Adapted for Batch Agent Service: import from app.ws_context.
+"""
+
+from __future__ import annotations
+
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.ws_context import execute_on_client
+
+
+@tool
+async def list_directory(path: str) -> str:
+    """List files and folders in a local directory on the user's device.
+
+    Returns a formatted listing of entries with name, type (file/directory),
+    and full path.
+    """
+    result = await execute_on_client(
+        action="list_directory",
+        data={"path": path},
+    )
+    entries: list[dict[str, Any]] = result.get("entries", [])
+    if not entries:
+        return f"Directory '{path}' is empty or does not exist."
+    lines: list[str] = []
+    for entry in entries:
+        entry_type = entry.get("type", "unknown")
+        entry_name = entry.get("name", "")
+        entry_path = entry.get("path", "")
+        lines.append(f"- [{entry_type}] {entry_name}  ({entry_path})")
+    return f"Directory listing for '{path}' ({len(entries)} entries):\n" + "\n".join(lines)
+
+
+@tool
+async def read_file_content(path: str) -> str:
+    """Read the text content of a local file on the user's device.
+
+    Returns the file content as a string.  Large files may be truncated
+    by the Electron client.
+    """
+    result = await execute_on_client(
+        action="read_file_content",
+        data={"path": path},
+    )
+    content: str = result.get("content", "")
+    if not content:
+        return f"File '{path}' is empty or could not be read."
+    return content
+
+
+@tool
+async def get_file_metadata(path: str) -> str:
+    """Get metadata for a local file: size, creation date, modification date, extension.
+
+    Returns a formatted summary of the file's metadata.
+    """
+    result = await execute_on_client(
+        action="get_file_metadata",
+        data={"path": path},
+    )
+    size = result.get("size", "unknown")
+    created = result.get("createdAt", "unknown")
+    modified = result.get("modifiedAt", "unknown")
+    extension = result.get("extension", "unknown")
+    name = result.get("name", path)
+    return (
+        f"File: {name}\n"
+        f"  Extension: {extension}\n"
+        f"  Size: {size} bytes\n"
+        f"  Created: {created}\n"
+        f"  Modified: {modified}"
+    )
+
+
+FILESYSTEM_TOOLS: list[Any] = [
+    list_directory,
+    read_file_content,
+    get_file_metadata,
+]

services/batch-agent/app/agents/note_agent.py

@@ -0,0 +1,110 @@
+"""Note agent — Markdown note management.
+
+Adapted for Batch Agent Service: import from app.ws_context and app.llm.
+"""
+
+from __future__ import annotations
+
+import re
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.llm import embed
+from app.ws_context import execute_on_client
+
+_UUID_RE = re.compile(
+    r"^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$"
+)
+
+
+def _is_uuid(value: str) -> bool:
+    return bool(_UUID_RE.match(value))
+
+
+@tool
+async def list_notes(project_id: str = "") -> str:
+    """List notes, optionally scoped to a project by project_id."""
+    normalized_project_id = project_id if (project_id and _is_uuid(project_id)) else ""
+    result = await execute_on_client(
+        action="select",
+        table="notes",
+        filters={"projectId": normalized_project_id or None},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No notes found."
+    lines = [f"- {r['title']} (id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} note(s):\n" + "\n".join(lines)
+
+
+@tool
+async def get_note(note_id: str) -> str:
+    """Fetch a single note by its UUID to read its full Markdown content."""
+    result = await execute_on_client(action="get", table="notes", data={"id": note_id})
+    row = result.get("row")
+    if not row:
+        return f"Note {note_id} not found."
+    return f"Note '{row['title']}' (id: {row['id']}):\n\n{row['content']}"
+
+
+@tool
+async def create_note(title: str, content: str, project_id: str = "") -> str:
+    """Create a new note."""
+    result = await execute_on_client(
+        action="insert",
+        table="notes",
+        data={
+            "title": title,
+            "content": content,
+            "projectId": project_id or None,
+        },
+    )
+    row = result["row"]
+    vector = await embed(content)
+    await execute_on_client(
+        action="vector_upsert",
+        data={"id": row["id"], "projectId": row.get("projectId"), "content": content},
+        vector=vector,
+    )
+    return f"Note created: '{row['title']}' (id: {row['id']})."
+
+
+@tool
+async def update_note(note_id: str, title: str = "", content: str = "") -> str:
+    """Update an existing note. Only pass fields that should change."""
+    updates: dict[str, Any] = {}
+    if title:
+        updates["title"] = title
+    if content:
+        updates["content"] = content
+    result = await execute_on_client(
+        action="update",
+        table="notes",
+        data={"id": note_id, "updates": updates},
+    )
+    row = result["row"]
+    if content:
+        vector = await embed(content)
+        await execute_on_client(
+            action="vector_upsert",
+            data={"id": note_id, "projectId": row.get("projectId"), "content": content},
+            vector=vector,
+        )
+    return f"Note updated: '{row['title']}' (id: {row['id']})."
+
+
+@tool
+async def delete_note(note_id: str) -> str:
+    """Delete a note permanently by its UUID."""
+    await execute_on_client(action="delete", table="notes", data={"id": note_id})
+    return f"Note {note_id} deleted."
+
+
+NOTE_TOOLS: list[Any] = [
+    list_notes,
+    get_note,
+    create_note,
+    update_note,
+    delete_note,
+]

services/batch-agent/app/agents/project_agent.py

@@ -0,0 +1,110 @@
+"""Project agent — full lifecycle management.
+
+Adapted for Batch Agent Service: import from app.ws_context.
+"""
+
+from __future__ import annotations
+
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.ws_context import execute_on_client
+
+
+@tool
+async def list_projects(client_id: str = "", include_archived: int = 0) -> str:
+    """List projects, optionally filtered by client_id."""
+    result = await execute_on_client(
+        action="select",
+        table="projects",
+        filters={
+            "clientId": client_id or None,
+            "includeArchived": bool(include_archived),
+        },
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No projects found."
+    lines = [f"- {r['name']} (status: {r['status']}, id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} project(s):\n" + "\n".join(lines)
+
+
+@tool
+async def list_all_projects() -> str:
+    """List every project regardless of client or status."""
+    result = await execute_on_client(action="select", table="projects")
+    rows = result.get("rows", [])
+    if not rows:
+        return "No projects found."
+    lines = [f"- {r['name']} (status: {r['status']}, id: {r['id']})" for r in rows]
+    return f"All projects ({len(rows)}):\n" + "\n".join(lines)
+
+
+@tool
+async def get_project(project_id: str) -> str:
+    """Fetch a single project by its UUID."""
+    result = await execute_on_client(action="get", table="projects", data={"id": project_id})
+    row = result.get("row")
+    if not row:
+        return f"Project {project_id} not found."
+    return (
+        f"Project: '{row['name']}' (id: {row['id']}, status: {row['status']}, "
+        f"clientId: {row.get('clientId', 'none')})"
+    )
+
+
+@tool
+async def create_project(name: str, client_id: str = "") -> str:
+    """Create a new project."""
+    result = await execute_on_client(
+        action="insert",
+        table="projects",
+        data={"name": name, "clientId": client_id or None},
+    )
+    row = result["row"]
+    return f"Project created: '{row['name']}' (id: {row['id']})"
+
+
+@tool
+async def update_project(
+    project_id: str,
+    name: str = "",
+    client_id: str = "",
+    status: str = "",
+    ai_summary: str = "",
+) -> str:
+    """Update a project. Only pass fields that should change."""
+    updates: dict[str, Any] = {}
+    if name:
+        updates["name"] = name
+    if client_id:
+        updates["clientId"] = client_id
+    if status:
+        updates["status"] = status
+    if ai_summary:
+        updates["aiSummary"] = ai_summary
+    result = await execute_on_client(
+        action="update",
+        table="projects",
+        data={"id": project_id, "updates": updates},
+    )
+    row = result["row"]
+    return f"Project updated: '{row['name']}' (id: {row['id']}, status: {row['status']})"
+
+
+@tool
+async def delete_project(project_id: str) -> str:
+    """Permanently delete a project."""
+    await execute_on_client(action="delete", table="projects", data={"id": project_id})
+    return f"Project {project_id} permanently deleted."
+
+
+PROJECT_TOOLS: list[Any] = [
+    list_projects,
+    list_all_projects,
+    get_project,
+    create_project,
+    update_project,
+    delete_project,
+]

services/batch-agent/app/agents/task_agent.py

@@ -0,0 +1,197 @@
+"""Task agent — full CRUD for tasks and task comments.
+
+Adapted for Batch Agent Service: import from app.ws_context.
+"""
+
+from __future__ import annotations
+
+from datetime import datetime, timezone
+import re
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.ws_context import execute_on_client
+
+_UUID_RE = re.compile(
+    r"^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$"
+)
+
+
+def _is_uuid(value: str) -> bool:
+    return bool(_UUID_RE.match(value))
+
+
+@tool
+async def list_tasks(
+    project_id: str = "",
+    status: str = "",
+    search: str = "",
+    order_by: str = "",
+) -> str:
+    """List tasks, optionally filtered by project_id, status, search, or order_by."""
+    normalized_project_id = project_id if (project_id and _is_uuid(project_id)) else ""
+    result = await execute_on_client(
+        action="select",
+        table="tasks",
+        filters={
+            "projectId": normalized_project_id or None,
+            "status": status or None,
+            "search": search or None,
+            "orderBy": order_by or None,
+        },
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No tasks found matching the given filters."
+    lines = [
+        f"- {r['title']} (status: {r['status']}, priority: {r['priority']}, id: {r['id']})"
+        for r in rows
+    ]
+    return f"Found {len(rows)} task(s):\n" + "\n".join(lines)
+
+
+@tool
+async def create_task(
+    title: str,
+    description: str = "",
+    status: str = "todo",
+    priority: str = "medium",
+    assignees: str = "[]",
+    due_date: int = 0,
+    project_id: str = "",
+    is_ai_suggested: int = 0,
+) -> str:
+    """Create a new task."""
+    result = await execute_on_client(
+        action="insert",
+        table="tasks",
+        data={
+            "title": title,
+            "description": description or None,
+            "status": status,
+            "priority": priority,
+            "assignee": assignees,
+            "dueDate": due_date or None,
+            "projectId": project_id or None,
+            "isAiSuggested": is_ai_suggested,
+        },
+    )
+    row = result["row"]
+    return (
+        f"Task created: '{row['title']}' "
+        f"(id: {row['id']}, status: {row['status']}, priority: {row['priority']})"
+    )
+
+
+@tool
+async def update_task(
+    task_id: str,
+    title: str = "",
+    description: str = "",
+    status: str = "",
+    priority: str = "",
+    assignees: str = "",
+    due_date: int = -1,
+    project_id: str = "",
+) -> str:
+    """Update fields on an existing task. Only pass fields you want to change."""
+    updates: dict[str, Any] = {}
+    if title:
+        updates["title"] = title
+    if description:
+        updates["description"] = description
+    if status:
+        updates["status"] = status
+    if priority:
+        updates["priority"] = priority
+    if assignees:
+        updates["assignee"] = assignees
+    if due_date != -1:
+        updates["dueDate"] = due_date or None
+    if project_id:
+        updates["projectId"] = project_id
+    result = await execute_on_client(
+        action="update",
+        table="tasks",
+        data={"id": task_id, "updates": updates},
+    )
+    row = result["row"]
+    return f"Task updated: '{row['title']}' (id: {row['id']}, status: {row['status']})"
+
+
+@tool
+async def delete_task(task_id: str) -> str:
+    """Delete a task permanently by its UUID."""
+    await execute_on_client(action="delete", table="tasks", data={"id": task_id})
+    return f"Task {task_id} deleted."
+
+
+@tool
+async def list_tasks_due_today() -> str:
+    """List all tasks whose due date falls on today's date."""
+    now = datetime.now(tz=timezone.utc)
+    start_ms = int(datetime(now.year, now.month, now.day, tzinfo=timezone.utc).timestamp() * 1000)
+    end_ms = start_ms + 86_400_000 - 1
+    result = await execute_on_client(
+        action="select",
+        table="tasks",
+        filters={"dueDateFrom": start_ms, "dueDateTo": end_ms},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No tasks are due today."
+    lines = [
+        f"- {r['title']} (priority: {r['priority']}, status: {r['status']}, id: {r['id']})"
+        for r in rows
+    ]
+    return f"Tasks due today ({len(rows)}):\n" + "\n".join(lines)
+
+
+@tool
+async def list_task_comments(task_id: str) -> str:
+    """List all comments on a task by its UUID."""
+    result = await execute_on_client(
+        action="select",
+        table="taskComments",
+        filters={"taskId": task_id},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return f"No comments found for task {task_id}."
+    lines = [f"- [{r['author']}]: {r['content']} (id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} comment(s):\n" + "\n".join(lines)
+
+
+@tool
+async def add_task_comment(task_id: str, author: str, content: str) -> str:
+    """Add a comment to a task."""
+    result = await execute_on_client(
+        action="insert",
+        table="taskComments",
+        data={"taskId": task_id, "author": author, "content": content},
+    )
+    row = result.get("row", {})
+    row_author = row.get("author", author)
+    row_task_id = row.get("taskId") or row.get("task_id") or task_id
+    row_comment_id = row.get("id", "unknown")
+    return f"Comment added by {row_author} on task {row_task_id} (comment id: {row_comment_id})."
+
+
+@tool
+async def delete_task_comment(comment_id: str) -> str:
+    """Delete a task comment by its UUID."""
+    await execute_on_client(action="delete", table="taskComments", data={"id": comment_id})
+    return f"Comment {comment_id} deleted."
+
+
+TASK_TOOLS: list[Any] = [
+    list_tasks,
+    create_task,
+    update_task,
+    delete_task,
+    list_tasks_due_today,
+    list_task_comments,
+    add_task_comment,
+    delete_task_comment,
+]

services/batch-agent/app/agents/timeline_agent.py

@@ -0,0 +1,88 @@
+"""Timeline agent — project milestone management.
+
+Adapted for Batch Agent Service: import from app.ws_context.
+"""
+
+from __future__ import annotations
+
+import re
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.ws_context import execute_on_client
+
+_UUID_RE = re.compile(
+    r"^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$"
+)
+
+
+def _is_uuid(value: str) -> bool:
+    return bool(_UUID_RE.match(value))
+
+
+@tool
+async def list_timelines(project_id: str = "") -> str:
+    """List timelines. Provide project_id to scope to a specific project."""
+    normalized_project_id = project_id if (project_id and _is_uuid(project_id)) else ""
+    result = await execute_on_client(
+        action="select",
+        table="timelines",
+        filters={"projectId": normalized_project_id or None},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No timelines found."
+    lines = [f"- {r['title']} (date: {r['date']}, id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} timeline(s):\n" + "\n".join(lines)
+
+
+@tool
+async def create_timeline(
+    project_id: str, title: str, date: int, is_ai_suggested: int = 0,
+) -> str:
+    """Create a project timeline (milestone)."""
+    result = await execute_on_client(
+        action="insert",
+        table="timelines",
+        data={
+            "projectId": project_id,
+            "title": title,
+            "date": date,
+            "isAiSuggested": is_ai_suggested,
+        },
+    )
+    row = result["row"]
+    return f"Timeline created: '{row['title']}' (id: {row['id']}, date: {row['date']})"
+
+
+@tool
+async def update_timeline(timeline_id: str, title: str = "", date: int = -1) -> str:
+    """Update a timeline. Only pass fields that should change."""
+    updates: dict[str, Any] = {}
+    if title:
+        updates["title"] = title
+    if date != -1:
+        updates["date"] = date
+    result = await execute_on_client(
+        action="update",
+        table="timelines",
+        data={"id": timeline_id, "updates": updates},
+    )
+    row = result["row"]
+    return f"Timeline updated: '{row['title']}' (id: {row['id']})"
+
+
+@tool
+async def delete_timeline(timeline_id: str) -> str:
+    """Delete a timeline permanently by its UUID."""
+    await execute_on_client(action="delete", table="timelines", data={"id": timeline_id})
+    return f"Timeline {timeline_id} deleted."
+
+
+TIMELINE_TOOLS: list[Any] = [
+    list_timelines,
+    create_timeline,
+    update_timeline,
+    delete_timeline,
+]

services/batch-agent/app/integrations/__init__.py

@@ -0,0 +1,108 @@
+"""Cloud provider integration utilities.
+
+Adapted for Batch Agent Service: import from shared.config instead of app.config.
+
+Provides:
+  * Shared message dataclasses (EmailMessage, ChatMessage)
+  * get_provider() — factory for Gmail/MS Graph clients
+  * encrypt_token() / decrypt_token() — Fernet-based OAuth token encryption
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+from dataclasses import dataclass, field
+from datetime import datetime
+from typing import TYPE_CHECKING
+
+from cryptography.fernet import Fernet, InvalidToken
+
+from shared.config import settings
+
+if TYPE_CHECKING:
+    from app.integrations.gmail import GmailClient
+    from app.integrations.ms_graph import MSGraphClient
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class EmailMessage:
+    id: str
+    subject: str
+    sender: str
+    body_text: str
+    date: datetime
+    labels: list[str] = field(default_factory=list)
+
+    @property
+    def as_text(self) -> str:
+        date_str = self.date.strftime("%Y-%m-%d %H:%M")
+        labels_str = f" [{', '.join(self.labels)}]" if self.labels else ""
+        return (
+            f"From: {self.sender}\n"
+            f"Date: {date_str}{labels_str}\n"
+            f"Subject: {self.subject}\n\n"
+            f"{self.body_text}"
+        )
+
+
+@dataclass
+class ChatMessage:
+    id: str
+    content: str
+    sender: str
+    channel: str | None
+    date: datetime
+
+    @property
+    def as_text(self) -> str:
+        date_str = self.date.strftime("%Y-%m-%d %H:%M")
+        channel_str = f" [channel: {self.channel}]" if self.channel else ""
+        return (
+            f"From: {self.sender}\n"
+            f"Date: {date_str}{channel_str}\n\n"
+            f"{self.content}"
+        )
+
+
+def _get_fernet() -> Fernet:
+    key = settings.OAUTH_ENCRYPTION_KEY
+    if not key:
+        raise RuntimeError(
+            "OAUTH_ENCRYPTION_KEY is not set. "
+            "Generate one with: python -c \"from cryptography.fernet import Fernet; print(Fernet.generate_key().decode())\""
+        )
+    return Fernet(key.encode() if isinstance(key, str) else key)
+
+
+def encrypt_token(token_info: dict) -> str:
+    if not isinstance(token_info, dict) or not token_info:
+        raise ValueError("token_info must be a non-empty dict")
+    plaintext = json.dumps(token_info).encode("utf-8")
+    return _get_fernet().encrypt(plaintext).decode("utf-8")
+
+
+def decrypt_token(encrypted: str) -> dict:
+    try:
+        plaintext = _get_fernet().decrypt(encrypted.encode("utf-8"))
+        return json.loads(plaintext)
+    except (InvalidToken, json.JSONDecodeError) as exc:
+        raise ValueError(f"Failed to decrypt OAuth token: {exc}") from exc
+
+
+def get_provider(
+    provider: str,
+    credentials_info: dict,
+) -> "GmailClient | MSGraphClient":
+    if provider == "gmail":
+        from app.integrations.gmail import GmailClient
+        return GmailClient(credentials_info)
+    if provider in {"outlook", "teams"}:
+        from app.integrations.ms_graph import MSGraphClient
+        return MSGraphClient(credentials_info)
+    raise ValueError(
+        f"Unknown cloud provider {provider!r}. "
+        "Supported: 'gmail', 'outlook', 'teams'."
+    )

services/batch-agent/app/integrations/gmail.py

@@ -0,0 +1,252 @@
+"""Gmail API client for cloud agent integration.
+
+Adapted for Batch Agent Service: import from app.integrations instead of
+app.integrations (same relative path within the service).
+"""
+
+from __future__ import annotations
+
+import asyncio
+import base64
+import email
+import html
+import logging
+import re
+from datetime import datetime, timezone
+from typing import Any
+
+from app.integrations import EmailMessage
+
+logger = logging.getLogger(__name__)
+
+_GMAIL_DATE_FMT = "%Y/%m/%d"
+_BODY_TRUNCATE = 8_000
+_MAX_MESSAGES = 200
+
+
+def _build_gmail_query(
+    filter_config: dict[str, Any] | None,
+    since: datetime | None,
+) -> str:
+    parts: list[str] = []
+    cfg = filter_config or {}
+
+    labels: list[str] = cfg.get("labels", [])
+    if labels:
+        if len(labels) == 1:
+            parts.append(f"label:{labels[0]}")
+        else:
+            label_expr = " OR ".join(f"label:{lbl}" for lbl in labels)
+            parts.append(f"({label_expr})")
+
+    senders: list[str] = cfg.get("senders", [])
+    for sender in senders:
+        parts.append(f"from:{sender}")
+
+    date_range: dict = cfg.get("date_range", {})
+    from_str: str | None = date_range.get("from")
+    to_str: str | None = date_range.get("to")
+
+    effective_since: datetime | None = since
+    if from_str:
+        try:
+            cfg_since = datetime.fromisoformat(from_str.replace("Z", "+00:00"))
+            if cfg_since.tzinfo is None:
+                cfg_since = cfg_since.replace(tzinfo=timezone.utc)
+            if effective_since is None or cfg_since > effective_since:
+                effective_since = cfg_since
+        except ValueError:
+            logger.warning("gmail: invalid date_range.from %r — ignoring", from_str)
+
+    if effective_since:
+        parts.append(f"after:{effective_since.strftime(_GMAIL_DATE_FMT)}")
+
+    if to_str:
+        try:
+            to_dt = datetime.fromisoformat(to_str.replace("Z", "+00:00"))
+            parts.append(f"before:{to_dt.strftime(_GMAIL_DATE_FMT)}")
+        except ValueError:
+            logger.warning("gmail: invalid date_range.to %r — ignoring", to_str)
+
+    return " ".join(parts)
+
+
+def _strip_html(raw_html: str) -> str:
+    no_tags = re.sub(r"<[^>]+>", " ", raw_html)
+    decoded = html.unescape(no_tags)
+    return re.sub(r"\s+", " ", decoded).strip()
+
+
+def _parse_body(payload: dict[str, Any]) -> str:
+    mime_type: str = payload.get("mimeType", "")
+    body: dict = payload.get("body", {})
+    parts: list[dict] = payload.get("parts", [])
+
+    if mime_type == "text/plain":
+        data = body.get("data", "")
+        if data:
+            return base64.urlsafe_b64decode(data + "==").decode("utf-8", errors="replace")
+        return ""
+
+    if mime_type == "text/html":
+        data = body.get("data", "")
+        if data:
+            raw = base64.urlsafe_b64decode(data + "==").decode("utf-8", errors="replace")
+            return _strip_html(raw)
+        return ""
+
+    plain_fallback = ""
+    for part in parts:
+        part_mime = part.get("mimeType", "")
+        if part_mime == "text/plain":
+            return _parse_body(part)
+        if part_mime == "text/html" and not plain_fallback:
+            plain_fallback = _parse_body(part)
+        if part_mime.startswith("multipart/"):
+            nested = _parse_body(part)
+            if nested:
+                return nested
+    return plain_fallback
+
+
+def _parse_date(raw: str) -> datetime:
+    try:
+        parsed = email.utils.parsedate_to_datetime(raw)
+        if parsed.tzinfo is None:
+            parsed = parsed.replace(tzinfo=timezone.utc)
+        return parsed.astimezone(timezone.utc)
+    except Exception:
+        return datetime.now(timezone.utc)
+
+
+class GmailClient:
+    def __init__(self, credentials_info: dict[str, Any]) -> None:
+        from google.oauth2.credentials import Credentials
+
+        self._credentials_info = credentials_info
+        expiry_str: str | None = credentials_info.get("expiry")
+        expiry: datetime | None = None
+        if expiry_str:
+            try:
+                expiry = datetime.fromisoformat(
+                    expiry_str.replace("Z", "+00:00")
+                ).replace(tzinfo=timezone.utc)
+            except ValueError:
+                pass
+
+        self._credentials = Credentials(
+            token=credentials_info.get("token"),
+            refresh_token=credentials_info.get("refresh_token"),
+            token_uri=credentials_info.get("token_uri", "https://oauth2.googleapis.com/token"),
+            client_id=credentials_info.get("client_id"),
+            client_secret=credentials_info.get("client_secret"),
+            scopes=credentials_info.get("scopes"),
+            expiry=expiry,
+        )
+
+    async def fetch_messages(
+        self,
+        filter_config: dict[str, Any] | None = None,
+        since: datetime | None = None,
+    ) -> list[EmailMessage]:
+        query = _build_gmail_query(filter_config, since)
+        logger.debug("gmail: executing search query %r", query)
+        return await asyncio.to_thread(self._fetch_sync, query)
+
+    @property
+    def refreshed_credentials(self) -> dict[str, Any] | None:
+        creds = self._credentials
+        if not creds.valid and creds.expired:
+            return None
+        if creds.token != self._credentials_info.get("token"):
+            result = {
+                "token": creds.token,
+                "refresh_token": creds.refresh_token,
+                "token_uri": creds.token_uri,
+                "client_id": creds.client_id,
+                "client_secret": creds.client_secret,
+                "scopes": list(creds.scopes or []),
+            }
+            if creds.expiry:
+                result["expiry"] = creds.expiry.isoformat()
+            return result
+        return None
+
+    def _fetch_sync(self, query: str) -> list[EmailMessage]:
+        import googleapiclient.discovery
+        import googleapiclient.errors
+        from google.auth.transport.requests import Request
+
+        if self._credentials.expired and self._credentials.refresh_token:
+            try:
+                self._credentials.refresh(Request())
+            except Exception as exc:
+                raise RuntimeError(f"Gmail token refresh failed: {exc}") from exc
+
+        service = googleapiclient.discovery.build(
+            "gmail", "v1", credentials=self._credentials, cache_discovery=False
+        )
+        user_api = service.users()
+
+        ids: list[str] = []
+        page_token: str | None = None
+        while len(ids) < _MAX_MESSAGES:
+            batch_size = min(100, _MAX_MESSAGES - len(ids))
+            kwargs: dict[str, Any] = {
+                "userId": "me",
+                "maxResults": batch_size,
+            }
+            if query:
+                kwargs["q"] = query
+            if page_token:
+                kwargs["pageToken"] = page_token
+
+            try:
+                resp = user_api.messages().list(**kwargs).execute()
+            except googleapiclient.errors.HttpError as exc:
+                raise RuntimeError(f"Gmail messages.list failed: {exc}") from exc
+
+            for msg in resp.get("messages", []):
+                ids.append(msg["id"])
+
+            page_token = resp.get("nextPageToken")
+            if not page_token:
+                break
+
+        if not ids:
+            return []
+
+        logger.info("gmail: fetching %d message(s)", len(ids))
+
+        messages: list[EmailMessage] = []
+        for msg_id in ids:
+            try:
+                msg = user_api.messages().get(
+                    userId="me", id=msg_id, format="full"
+                ).execute()
+
+                headers: dict[str, str] = {
+                    h["name"].lower(): h["value"]
+                    for h in msg.get("payload", {}).get("headers", [])
+                }
+                subject = headers.get("subject", "(no subject)")
+                sender = headers.get("from", "unknown")
+                date_raw = headers.get("date", "")
+                date = _parse_date(date_raw) if date_raw else datetime.now(timezone.utc)
+
+                body_text = _parse_body(msg.get("payload", {}))[:_BODY_TRUNCATE]
+                labels = msg.get("labelIds", [])
+
+                messages.append(EmailMessage(
+                    id=msg_id,
+                    subject=subject,
+                    sender=sender,
+                    body_text=body_text,
+                    date=date,
+                    labels=labels,
+                ))
+            except Exception as exc:
+                logger.warning("gmail: skipping message %s: %s", msg_id, exc)
+
+        logger.info("gmail: returned %d message(s)", len(messages))
+        return messages

services/batch-agent/app/integrations/ms_graph.py

@@ -0,0 +1,266 @@
+"""Microsoft Graph API client for Outlook and Teams.
+
+Adapted for Batch Agent Service: import settings from shared.config.
+"""
+
+from __future__ import annotations
+
+import logging
+import re
+from datetime import datetime, timezone
+from typing import Any
+
+import httpx
+
+from shared.config import settings
+from app.integrations import ChatMessage, EmailMessage
+
+logger = logging.getLogger(__name__)
+
+_GRAPH_BASE = "https://graph.microsoft.com/v1.0"
+
+_MAX_EMAILS = 200
+_MAX_MESSAGES = 200
+_BODY_TRUNCATE = 8_000
+
+
+def _strip_html(raw: str) -> str:
+    no_tags = re.sub(r"<[^>]+>", " ", raw)
+    import html as _html
+    decoded = _html.unescape(no_tags)
+    return re.sub(r"\s+", " ", decoded).strip()
+
+
+def _odata_datetime(dt: datetime) -> str:
+    utc = dt.astimezone(timezone.utc)
+    return utc.strftime("%Y-%m-%dT%H:%M:%SZ")
+
+
+def _build_email_filter(
+    filter_config: dict[str, Any] | None,
+    since: datetime | None,
+) -> str:
+    clauses: list[str] = []
+    cfg = filter_config or {}
+
+    senders: list[str] = cfg.get("senders", [])
+    if senders:
+        sender_clauses = [f"from/emailAddress/address eq '{s}'" for s in senders]
+        clauses.append("(" + " or ".join(sender_clauses) + ")")
+
+    date_range: dict = cfg.get("date_range", {})
+    from_str: str | None = date_range.get("from")
+
+    effective_since: datetime | None = since
+    if from_str:
+        try:
+            cfg_since = datetime.fromisoformat(from_str.replace("Z", "+00:00"))
+            if cfg_since.tzinfo is None:
+                cfg_since = cfg_since.replace(tzinfo=timezone.utc)
+            if effective_since is None or cfg_since > effective_since:
+                effective_since = cfg_since
+        except ValueError:
+            logger.warning("ms_graph: invalid date_range.from %r — ignoring", from_str)
+
+    if effective_since:
+        clauses.append(f"receivedDateTime ge {_odata_datetime(effective_since)}")
+
+    to_str: str | None = date_range.get("to")
+    if to_str:
+        try:
+            to_dt = datetime.fromisoformat(to_str.replace("Z", "+00:00"))
+            if to_dt.tzinfo is None:
+                to_dt = to_dt.replace(tzinfo=timezone.utc)
+            clauses.append(f"receivedDateTime le {_odata_datetime(to_dt)}")
+        except ValueError:
+            logger.warning("ms_graph: invalid date_range.to %r — ignoring", to_str)
+
+    return " and ".join(clauses)
+
+
+class MSGraphClient:
+    def __init__(self, credentials_info: dict[str, Any]) -> None:
+        self._credentials_info = credentials_info
+        self._access_token: str = credentials_info.get("access_token", "")
+        self._original_access_token: str = self._access_token
+        self._refresh_token: str | None = credentials_info.get("refresh_token")
+
+    def _auth_headers(self) -> dict[str, str]:
+        return {"Authorization": f"Bearer {self._access_token}"}
+
+    async def _refresh_access_token(self) -> None:
+        import msal
+
+        app = msal.ConfidentialClientApplication(
+            client_id=settings.MS_CLIENT_ID,
+            client_credential=settings.MS_CLIENT_SECRET,
+            authority=f"https://login.microsoftonline.com/{settings.MS_TENANT_ID}",
+        )
+        scopes: list[str] = self._credentials_info.get("scope", "").split()
+        if not scopes:
+            scopes = ["https://graph.microsoft.com/.default"]
+
+        result = app.acquire_token_by_refresh_token(
+            self._refresh_token,
+            scopes=scopes,
+        )
+        if "access_token" not in result:
+            error = result.get("error_description", result.get("error", "unknown"))
+            raise RuntimeError(f"MS Graph token refresh failed: {error}")
+
+        self._access_token = result["access_token"]
+        if "refresh_token" in result:
+            self._refresh_token = result["refresh_token"]
+            self._credentials_info["refresh_token"] = result["refresh_token"]
+        self._credentials_info["access_token"] = self._access_token
+
+    @property
+    def refreshed_credentials(self) -> dict[str, Any] | None:
+        if self._access_token != self._original_access_token:
+            return {**self._credentials_info, "access_token": self._access_token}
+        return None
+
+    async def _get(
+        self,
+        client: httpx.AsyncClient,
+        url: str,
+        params: dict[str, Any] | None = None,
+        *,
+        retry_on_401: bool = True,
+    ) -> dict[str, Any]:
+        resp = await client.get(url, params=params, headers=self._auth_headers())
+        if resp.status_code == 401 and retry_on_401 and self._refresh_token:
+            await self._refresh_access_token()
+            resp = await client.get(url, params=params, headers=self._auth_headers())
+        if resp.status_code == 429:
+            raise RuntimeError("MS Graph rate limit hit (429). Try again later.")
+        resp.raise_for_status()
+        return resp.json()
+
+    async def fetch_emails(
+        self,
+        filter_config: dict[str, Any] | None = None,
+        since: datetime | None = None,
+    ) -> list[EmailMessage]:
+        odata_filter = _build_email_filter(filter_config, since)
+        params: dict[str, Any] = {
+            "$top": 50,
+            "$select": "id,subject,from,receivedDateTime,body,bodyPreview",
+            "$orderby": "receivedDateTime desc",
+        }
+        if odata_filter:
+            params["$filter"] = odata_filter
+
+        emails: list[EmailMessage] = []
+        url = f"{_GRAPH_BASE}/me/messages"
+
+        async with httpx.AsyncClient(timeout=30.0) as client:
+            while url and len(emails) < _MAX_EMAILS:
+                data = await self._get(client, url, params if url.startswith(_GRAPH_BASE) else None)
+                for item in data.get("value", []):
+                    emails.append(self._parse_email(item))
+                    if len(emails) >= _MAX_EMAILS:
+                        break
+                url = data.get("@odata.nextLink", "")
+                params = {}
+
+        logger.info("ms_graph: fetched %d Outlook email(s)", len(emails))
+        return emails
+
+    async def fetch_messages(
+        self,
+        filter_config: dict[str, Any] | None = None,
+        since: datetime | None = None,
+    ) -> list[ChatMessage]:
+        cfg = filter_config or {}
+        channel_filter: list[str] = [c.lower() for c in cfg.get("channels", [])]
+        params: dict[str, Any] = {"$top": 50}
+        if since:
+            params["$filter"] = f"createdDateTime ge {_odata_datetime(since)}"
+
+        messages: list[ChatMessage] = []
+        url = f"{_GRAPH_BASE}/me/chats/getAllMessages"
+
+        async with httpx.AsyncClient(timeout=30.0) as client:
+            while url and len(messages) < _MAX_MESSAGES:
+                try:
+                    data = await self._get(client, url, params if url.startswith(_GRAPH_BASE) else None)
+                except httpx.HTTPStatusError as exc:
+                    if exc.response.status_code in (403, 404):
+                        logger.warning(
+                            "ms_graph: /me/chats/getAllMessages not available (%d)",
+                            exc.response.status_code,
+                        )
+                        break
+                    raise
+
+                for item in data.get("value", []):
+                    msg = self._parse_teams_message(item)
+                    if channel_filter and msg.channel:
+                        if not any(c in msg.channel.lower() for c in channel_filter):
+                            continue
+                    messages.append(msg)
+                    if len(messages) >= _MAX_MESSAGES:
+                        break
+                url = data.get("@odata.nextLink", "")
+                params = {}
+
+        logger.info("ms_graph: fetched %d Teams message(s)", len(messages))
+        return messages
+
+    @staticmethod
+    def _parse_email(item: dict[str, Any]) -> EmailMessage:
+        subject: str = item.get("subject", "(no subject)") or "(no subject)"
+        sender_block = item.get("from", {}) or {}
+        sender_addr = (
+            (sender_block.get("emailAddress") or {}).get("address", "unknown")
+        )
+        date_str: str = item.get("receivedDateTime", "")
+        try:
+            date = datetime.fromisoformat(date_str.replace("Z", "+00:00"))
+        except Exception:
+            date = datetime.now(timezone.utc)
+
+        body_block = item.get("body", {}) or {}
+        content_type: str = body_block.get("contentType", "text")
+        raw_body: str = body_block.get("content", "")
+        if content_type == "html":
+            body_text = _strip_html(raw_body)
+        else:
+            body_text = raw_body or item.get("bodyPreview", "")
+        body_text = body_text[:_BODY_TRUNCATE]
+
+        return EmailMessage(
+            id=item.get("id", ""),
+            subject=subject,
+            sender=sender_addr,
+            body_text=body_text,
+            date=date,
+        )
+
+    @staticmethod
+    def _parse_teams_message(item: dict[str, Any]) -> ChatMessage:
+        msg_id: str = item.get("id", "")
+        sender_block = (item.get("from") or {}).get("user") or {}
+        sender: str = sender_block.get("displayName", "unknown")
+        channel: str | None = (item.get("channelIdentity") or {}).get("channelId")
+
+        date_str: str = item.get("createdDateTime", "")
+        try:
+            date = datetime.fromisoformat(date_str.replace("Z", "+00:00"))
+        except Exception:
+            date = datetime.now(timezone.utc)
+
+        body_block = item.get("body", {}) or {}
+        content_type: str = body_block.get("contentType", "text")
+        raw_content: str = body_block.get("content", "")
+        content = _strip_html(raw_content) if content_type == "html" else raw_content
+        content = content[:_BODY_TRUNCATE]
+
+        return ChatMessage(
+            id=msg_id,
+            content=content,
+            sender=sender,
+            channel=channel,
+            date=date,
+        )

services/batch-agent/app/journey.py

@@ -0,0 +1,385 @@
+"""Chatbot Journey — guided conversation to build an agent prompt_template.
+
+Adapted for Batch Agent Service: imports from app.agents.filesystem_agent
+and app.llm instead of monolith paths.  Session state is in-memory (could
+be moved to Redis for horizontal scaling in the future).
+
+Journey flow:
+  1. Redis consumer dispatches ``journey_start`` with basic agent config.
+  2. Server creates an in-memory session, runs the setup LLM with
+     file-system tools to explore the directory, returns first question.
+  3. ``journey_message`` frames drive the conversation.
+  4. After 3-5 turns the LLM emits PROMPT_TEMPLATE_START / _END block.
+  5. Server parses the block and returns ``journey_reply`` with ``done=True``.
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+import time
+import uuid
+from dataclasses import dataclass, field
+from typing import Any
+
+from langchain_core.messages import AIMessage, HumanMessage, SystemMessage, ToolMessage
+
+from app.agents.filesystem_agent import FILESYSTEM_TOOLS
+from app.llm import get_llm
+
+logger = logging.getLogger(__name__)
+
+# ── Session TTL ───────────────────────────────────────────────────────────
+
+_SESSION_TTL_SECONDS: int = 1800  # 30 minutes
+
+# Sentinel strings used to delimit the LLM-produced prompt_template.
+_TEMPLATE_START = "PROMPT_TEMPLATE_START"
+_TEMPLATE_END = "PROMPT_TEMPLATE_END"
+
+_MIN_TURNS_BEFORE_NUDGE: int = 3
+_MAX_TURNS: int = 15
+_MAX_TOOL_STEPS: int = 6
+
+# ── In-memory session store ───────────────────────────────────────────────
+
+
+@dataclass
+class JourneySession:
+    session_id: str
+    user_id: str
+    agent_type: str  # "local" | "cloud"
+    directory: str
+    data_types: list[str]
+    history: list[dict[str, Any]] = field(default_factory=list)
+    system_prompt: str = ""
+    created_at: float = field(default_factory=time.monotonic)
+
+    def is_expired(self) -> bool:
+        return (time.monotonic() - self.created_at) > _SESSION_TTL_SECONDS
+
+
+# session_id → session
+_sessions: dict[str, JourneySession] = {}
+
+
+def get_journey_session(session_id: str, user_id: str) -> JourneySession | None:
+    """Retrieve session; return None on missing, expired, or wrong owner."""
+    s = _sessions.get(session_id)
+    if s is None or s.is_expired():
+        _sessions.pop(session_id, None)
+        return None
+    if s.user_id != user_id:
+        return None
+    return s
+
+
+# ── System prompt builder ─────────────────────────────────────────────────
+
+_SYSTEM_PROMPT_TEMPLATE = """\
+You are a friendly assistant helping a freelancer configure a data-extraction agent.
+Your job is to understand exactly what data the user wants to extract from their
+local directory and produce a detailed prompt_template that a separate AI will use
+as its instruction set.
+
+The extraction agent already has this base behaviour built in:
+  - Reads each file using file-system tools.
+  - Creates records (tasks, notes, timelines, projects) via CRUD tools.
+  - Sets isAiSuggested=1 on every new record.
+  - Only extracts data explicitly present in the files — it never invents information.
+The user's custom prompt is appended AFTER this base behaviour, so focus on
+what to look for and how to map it — not on the general extraction mechanics.
+
+You have access to file-system tools to explore the user's directory:
+- list_directory: to see folder structure
+- read_file_content: to peek at file contents
+- get_file_metadata: to check file info
+
+The user's configured directory is: {directory}
+Target data types: {data_types}
+
+IMPORTANT — project assignment is handled automatically by the main agent runner
+before the custom prompt is ever used.  You MUST NOT ask the user about projects,
+projectId, or how to link records to projects.  Never include projectId logic or
+project creation instructions in the generated prompt_template.
+
+Start by exploring the directory to understand its structure.  Then ask concise,
+focused questions one at a time.  Cover these topics (not necessarily in this order):
+  1. The type and format of the source content (confirmed by your exploration).
+  2. How fields should be mapped (e.g. filename → task title).
+  3. Priority or status rules (e.g. "urgent" keyword → high priority).
+  4. Any special handling, date extraction, or exclusions.
+
+Once you reach 90% confidence, output the final prompt_template between these exact
+markers on their own lines:
+
+{template_start}
+<the complete extraction prompt here>
+{template_end}
+
+The prompt_template must be a self-contained instruction for an AI that reads files
+and must perform CRUD operations using tools to create records.  It should specify:
+  - What entity types to create (tasks, notes, timelines) — never projects.
+  - How to map file content to record fields (camelCase: title, status, priority,
+    dueDate, content, etc.) — never include projectId.
+  - That isAiSuggested must be set to 1 on every new record.
+  - Concrete examples of mappings based on what you discovered in the directory.
+
+{existing_section}\
+Keep asking clarifying questions until you are at least 90% confident you have
+enough information to generate an accurate prompt_template.  Once you reach that
+confidence level, stop asking and produce the final template immediately.
+Begin by exploring the directory, then ask your first question.\
+"""
+
+
+def _build_system_prompt(
+    directory: str,
+    data_types: list[str],
+    existing_template: str | None = None,
+) -> str:
+    existing_section = (
+        f"\nThe user already has the following prompt_template — refine it based on their answers:\n"
+        f"---\n{existing_template}\n---\n"
+        if existing_template
+        else ""
+    )
+    return _SYSTEM_PROMPT_TEMPLATE.format(
+        directory=directory,
+        data_types=", ".join(data_types),
+        template_start=_TEMPLATE_START,
+        template_end=_TEMPLATE_END,
+        existing_section=existing_section,
+    )
+
+
+# ── Template extraction ───────────────────────────────────────────────────
+
+
+def _extract_template(text: str) -> str | None:
+    """Return the text between PROMPT_TEMPLATE_START and PROMPT_TEMPLATE_END, or None."""
+    if _TEMPLATE_START not in text or _TEMPLATE_END not in text:
+        return None
+    start_idx = text.index(_TEMPLATE_START) + len(_TEMPLATE_START)
+    end_idx = text.index(_TEMPLATE_END)
+    return text[start_idx:end_idx].strip() or None
+
+
+# ── LLM call with tool support ───────────────────────────────────────────
+
+
+def _as_text(content: Any) -> str:
+    if content is None:
+        return ""
+    if isinstance(content, str):
+        return content
+    if isinstance(content, list):
+        parts: list[str] = []
+        for item in content:
+            if isinstance(item, str):
+                parts.append(item)
+            elif isinstance(item, dict):
+                text = item.get("text")
+                if isinstance(text, str):
+                    parts.append(text)
+        return "".join(parts)
+    return str(content)
+
+
+async def _call_llm_with_tools(
+    system_prompt: str,
+    history: list[dict[str, Any]],
+    tools: list[Any],
+) -> str:
+    """Build LangChain messages from history and invoke the LLM with tools.
+
+    Handles tool-calling loops: if the LLM calls tools, execute them and
+    continue until a final text response is produced.
+    """
+    messages: list[Any] = [SystemMessage(content=system_prompt)]
+    for turn in history:
+        if turn["role"] == "user":
+            messages.append(HumanMessage(content=turn["content"]))
+        else:
+            messages.append(AIMessage(content=turn["content"]))
+
+    llm = get_llm(model=None, temperature=0.4)
+    llm_with_tools = llm.bind_tools(tools)
+    tool_map = {tool_def.name: tool_def for tool_def in tools}
+
+    for _ in range(_MAX_TOOL_STEPS):
+        response: AIMessage = await llm_with_tools.ainvoke(messages)
+        messages.append(response)
+
+        if not response.tool_calls:
+            return _as_text(response.content)
+
+        for call in response.tool_calls:
+            call_name = str(call.get("name", ""))
+            call_args = call.get("args", {})
+            logger.info(
+                "journey: tool_call name=%s args=%s",
+                call_name,
+                json.dumps(call_args, ensure_ascii=True)[:500],
+            )
+
+            tool_fn = tool_map.get(call_name)
+            if tool_fn is None:
+                tool_output = f"Unknown tool: {call_name}"
+            else:
+                tool_output = await tool_fn.ainvoke(call_args)
+
+            logger.info(
+                "journey: tool_result name=%s output=%s",
+                call_name,
+                str(tool_output)[:800],
+            )
+            messages.append(ToolMessage(content=str(tool_output), tool_call_id=call["id"]))
+
+    # Fallback: exceeded max tool steps.
+    final = await llm.ainvoke(messages)
+    return _as_text(final.content)
+
+
+# ── Journey handlers (called from redis_consumer) ────────────────────────
+
+
+async def handle_journey_start(
+    user_id: str,
+    frame: dict[str, Any],
+) -> dict[str, Any]:
+    """Handle a ``journey_start`` request.
+
+    Creates a session, runs the setup LLM with directory exploration,
+    and returns the ``journey_reply`` payload.
+    """
+    agent_type = frame.get("agent_type", "local")
+    directory = frame.get("directory", "")
+    data_types = frame.get("data_types", [])
+    existing_template = frame.get("existing_template")
+
+    session_id = frame.get("session_id") or str(uuid.uuid4())
+    system_prompt = _build_system_prompt(directory, data_types, existing_template)
+
+    session = JourneySession(
+        session_id=session_id,
+        user_id=user_id,
+        agent_type=agent_type,
+        directory=directory,
+        data_types=data_types,
+        system_prompt=system_prompt,
+    )
+
+    seed_history: list[dict[str, Any]] = [
+        {"role": "user", "content": "Hi, I'm ready to set up my agent. Please explore my directory and ask me your first question."},
+    ]
+    ai_reply = await _call_llm_with_tools(
+        system_prompt=system_prompt,
+        history=seed_history,
+        tools=list(FILESYSTEM_TOOLS),
+    )
+
+    session.history.extend(seed_history)
+    session.history.append({"role": "assistant", "content": ai_reply})
+    _sessions[session_id] = session
+
+    logger.info(
+        "journey: session %s started for user %s (directory=%s)",
+        session_id,
+        user_id,
+        directory,
+    )
+
+    prompt_template = _extract_template(ai_reply)
+    done = prompt_template is not None
+
+    display_message = ai_reply
+    if done:
+        display_message = (
+            ai_reply[: ai_reply.index(_TEMPLATE_START)].strip()
+            or "Here is your agent configuration. You can save it or continue refining."
+        )
+        _sessions.pop(session_id, None)
+
+    return {
+        "type": "journey_reply",
+        "session_id": session_id,
+        "message": display_message,
+        "done": done,
+        "prompt_template": prompt_template,
+    }
+
+
+async def handle_journey_message(
+    user_id: str,
+    frame: dict[str, Any],
+) -> dict[str, Any]:
+    """Handle a ``journey_message`` request.
+
+    Appends the user message, calls the LLM, and returns the
+    ``journey_reply`` payload.
+    """
+    session_id = frame.get("session_id", "")
+    message = frame.get("message", "")
+
+    session = get_journey_session(session_id, user_id)
+    if session is None:
+        return {
+            "type": "journey_reply",
+            "session_id": session_id,
+            "message": "Journey session not found or expired. Please start a new setup.",
+            "done": True,
+            "prompt_template": None,
+        }
+
+    session.history.append({"role": "user", "content": message})
+
+    ai_reply = await _call_llm_with_tools(
+        system_prompt=session.system_prompt,
+        history=session.history,
+        tools=list(FILESYSTEM_TOOLS),
+    )
+
+    session.history.append({"role": "assistant", "content": ai_reply})
+
+    prompt_template = _extract_template(ai_reply)
+    done = prompt_template is not None
+
+    if not done:
+        turns = sum(1 for t in session.history if t["role"] == "user")
+        if turns >= _MAX_TURNS:
+            nudge_content = (
+                "[System: You have enough information. Please generate the final "
+                f"prompt_template now, wrapped in {_TEMPLATE_START} / {_TEMPLATE_END} markers.]"
+            )
+            session.history.append({"role": "user", "content": nudge_content})
+
+            nudge_reply = await _call_llm_with_tools(
+                system_prompt=session.system_prompt,
+                history=session.history,
+                tools=list(FILESYSTEM_TOOLS),
+            )
+            session.history.append({"role": "assistant", "content": nudge_reply})
+
+            prompt_template = _extract_template(nudge_reply)
+            if prompt_template is not None:
+                done = True
+                ai_reply = nudge_reply
+
+    display_message = ai_reply
+    if done:
+        display_message = (
+            ai_reply[: ai_reply.index(_TEMPLATE_START)].strip()
+            if _TEMPLATE_START in ai_reply
+            else "Here is your agent configuration. You can save it or continue refining."
+        )
+        _sessions.pop(session_id, None)
+        logger.info("journey: session %s completed for user %s", session_id, user_id)
+
+    return {
+        "type": "journey_reply",
+        "session_id": session_id,
+        "message": display_message,
+        "done": done,
+        "prompt_template": prompt_template,
+    }

services/batch-agent/app/llm.py

@@ -0,0 +1,76 @@
+"""LLM factory — centralised model instantiation via LiteLLM.
+
+Identical to services/chat/app/llm.py. Uses shared.config.settings.
+"""
+
+from __future__ import annotations
+
+import os
+import warnings
+
+from openai import AsyncOpenAI
+import litellm
+
+from langchain_openai import ChatOpenAI
+from langchain_litellm import ChatLiteLLM
+
+from shared.config import settings
+
+litellm.drop_params = True
+
+warnings.filterwarnings(
+    "ignore",
+    message=r"PydanticSerializationUnexpectedValue\(Expected `ResponseAPIUsage`",
+    category=UserWarning,
+)
+
+
+def _api_key_for_model(model: str) -> str | None:
+    if model.startswith("anthropic/"):
+        return settings.ANTHROPIC_API_KEY or None
+    if model.startswith("gemini/") or model.startswith("google/"):
+        return settings.GOOGLE_API_KEY or None
+    if model.startswith("cerebras/"):
+        return settings.CEREBRAS_API_KEY or None
+    if model.startswith("github_copilot/"):
+        return None
+    return settings.OPENAI_API_KEY or None
+
+
+def get_llm(
+    *,
+    model: str | None = None,
+    temperature: float = 0,
+) -> ChatOpenAI | ChatLiteLLM:
+    model = model or settings.LLM_MODEL
+
+    if settings.GITHUB_COPILOT_TOKEN_DIR:
+        os.environ.setdefault("GITHUB_COPILOT_TOKEN_DIR", settings.GITHUB_COPILOT_TOKEN_DIR)
+
+    if "/" in model:
+        return ChatLiteLLM(model=model, temperature=temperature)
+
+    return ChatOpenAI(
+        model=model,
+        temperature=temperature,
+        api_key=_api_key_for_model(model),
+    )
+
+
+def get_router_llm(
+    *,
+    temperature: float = 0,
+) -> ChatOpenAI | ChatLiteLLM:
+    return get_llm(model=settings.LLM_ROUTER_MODEL, temperature=temperature)
+
+
+async def embed(text: str) -> list[float]:
+    model = settings.LLM_EMBED_MODEL
+
+    if model.startswith("github_copilot/") or "/" in model:
+        response = await litellm.aembedding(model=model, input=[text])
+        return response.data[0]["embedding"]
+
+    client = AsyncOpenAI(api_key=settings.OPENAI_API_KEY)
+    response = await client.embeddings.create(model=model, input=text)
+    return response.data[0].embedding

services/batch-agent/app/main.py

@@ -0,0 +1,57 @@
+"""Batch Agent Service — FastAPI application.
+
+Owns: agent_runner (local directory + cloud connectors), journey builder,
+filesystem_agent, integrations (Gmail, MS Graph).
+
+Communicates with WS Gateway via Redis:
+  - Subscribes to  batch:request:{user_id}  (journey_start, journey_message)
+  - Publishes to   ws:out:{user_id}         (journey replies + tool calls)
+  - BRPOP on       tool:result:{call_id}    (tool-call round-trip, 30s timeout)
+  - SET+EX on      journey:{user_id}        (journey session state, TTL 1800s)
+"""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+
+from contextlib import asynccontextmanager
+from typing import AsyncGenerator
+
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+
+from app.redis_consumer import start_consumer
+from app.routes import router
+
+logger = logging.getLogger(__name__)
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI) -> AsyncGenerator[None, None]:
+    logger.info("batch-agent: starting Redis consumer")
+    task = asyncio.create_task(start_consumer())
+    yield
+    task.cancel()
+    try:
+        await task
+    except asyncio.CancelledError:
+        pass
+    logger.info("batch-agent: Redis consumer stopped")
+
+
+app = FastAPI(title="Adiuva Batch Agent Service", lifespan=lifespan)
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_methods=["GET", "POST"],
+    allow_headers=["*"],
+)
+
+app.include_router(router)
+
+
+@app.get("/health")
+async def health() -> dict[str, str]:
+    return {"status": "ok", "service": "batch-agent"}

services/batch-agent/app/redis_consumer.py

@@ -0,0 +1,141 @@
+"""Redis consumer for the Batch Agent Service.
+
+Subscribes to  batch:request:*  (pattern) and dispatches:
+  - journey_start   → handle_journey_start
+  - journey_message → handle_journey_message
+  - agent_trigger   → run_local_agent / run_cloud_agent
+
+Results are published back to  ws:out:{user_id}  via Redis.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import logging
+from typing import Any
+
+from shared.redis import redis_client, batch_request_channel, ws_out_channel
+
+from app.ws_context import set_current_user, clear_current_user
+
+logger = logging.getLogger(__name__)
+
+
+async def _publish_to_user(user_id: str, payload: dict[str, Any]) -> None:
+    """Publish a frame to the user's WS outbound channel."""
+    channel = ws_out_channel(user_id)
+    await redis_client.publish(channel, json.dumps(payload))
+
+
+async def _handle_journey_start(user_id: str, data: dict[str, Any]) -> None:
+    """Handle a journey_start request from WS Gateway."""
+    from app.journey import handle_journey_start
+
+    set_current_user(user_id)
+    try:
+        reply = await handle_journey_start(user_id, data)
+        await _publish_to_user(user_id, reply)
+    except Exception as exc:
+        logger.error("batch-agent: journey_start failed user=%s: %s", user_id, exc)
+        await _publish_to_user(user_id, {
+            "type": "journey_reply",
+            "session_id": data.get("session_id", ""),
+            "message": f"Journey setup failed: {exc}",
+            "done": True,
+            "prompt_template": None,
+        })
+    finally:
+        clear_current_user()
+
+
+async def _handle_journey_message(user_id: str, data: dict[str, Any]) -> None:
+    """Handle a journey_message from WS Gateway."""
+    from app.journey import handle_journey_message
+
+    set_current_user(user_id)
+    try:
+        reply = await handle_journey_message(user_id, data)
+        await _publish_to_user(user_id, reply)
+    except Exception as exc:
+        logger.error("batch-agent: journey_message failed user=%s: %s", user_id, exc)
+        await _publish_to_user(user_id, {
+            "type": "journey_reply",
+            "session_id": data.get("session_id", ""),
+            "message": f"Journey processing failed: {exc}",
+            "done": True,
+            "prompt_template": None,
+        })
+    finally:
+        clear_current_user()
+
+
+async def _handle_agent_trigger(user_id: str, data: dict[str, Any]) -> None:
+    """Handle an agent_trigger request from the REST route (forwarded via Redis)."""
+    from app.agent_runner import run_local_agent
+
+    set_current_user(user_id)
+    try:
+        await run_local_agent(user_id, data)
+    except Exception as exc:
+        logger.error("batch-agent: agent_trigger failed user=%s: %s", user_id, exc)
+        await _publish_to_user(user_id, {
+            "type": "run_complete",
+            "status": "error",
+            "run_context": data.get("run_context", {}),
+        })
+    finally:
+        clear_current_user()
+
+
+async def _dispatch(user_id: str, message_data: dict[str, Any]) -> None:
+    """Route a batch request to the correct handler."""
+    msg_type = message_data.get("type", "")
+
+    if msg_type == "journey_start":
+        await _handle_journey_start(user_id, message_data)
+    elif msg_type == "journey_message":
+        await _handle_journey_message(user_id, message_data)
+    elif msg_type == "agent_trigger":
+        await _handle_agent_trigger(user_id, message_data)
+    else:
+        logger.warning("batch-agent: unknown message type %r from user=%s", msg_type, user_id)
+
+
+async def start_consumer() -> None:
+    """Subscribe to batch:request:* and dispatch incoming frames."""
+    pubsub = redis_client.pubsub()
+    await pubsub.psubscribe("batch:request:*")
+    logger.info("batch-agent: subscribed to batch:request:*")
+
+    try:
+        async for message in pubsub.listen():
+            if message["type"] != "pmessage":
+                continue
+
+            channel: str = message["channel"]
+            if isinstance(channel, bytes):
+                channel = channel.decode()
+
+            # Extract user_id from channel: batch:request:{user_id}
+            parts = channel.split(":", 2)
+            if len(parts) < 3:
+                continue
+            user_id = parts[2]
+
+            raw = message["data"]
+            if isinstance(raw, bytes):
+                raw = raw.decode()
+
+            try:
+                data = json.loads(raw)
+            except json.JSONDecodeError:
+                logger.warning("batch-agent: invalid JSON on channel %s", channel)
+                continue
+
+            # Dispatch in a separate task to avoid blocking the consumer
+            asyncio.create_task(_dispatch(user_id, data))
+    except asyncio.CancelledError:
+        logger.info("batch-agent: consumer shutting down")
+    finally:
+        await pubsub.punsubscribe("batch:request:*")

services/batch-agent/app/routes.py

@@ -0,0 +1,208 @@
+"""Agent REST routes — catalog, billing checks, trigger.
+
+Adapted for Batch Agent Service: uses shared.db, shared.models, shared.schemas.
+Agent trigger dispatches via Redis to the consumer instead of spawning
+an in-process background task.
+"""
+
+from __future__ import annotations
+
+import json
+import uuid
+from datetime import datetime, timezone
+
+from fastapi import APIRouter, Header, HTTPException, status
+from sqlalchemy import func, select
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from shared.db import async_session
+from shared.models import AgentRunLog
+from shared.redis import redis_client, batch_request_channel
+
+from app.agent_runner import is_agent_running
+
+router = APIRouter(prefix="/agents", tags=["agents"])
+
+# ── Tier feature limits ───────────────────────────────────────────────
+# Mirrors app/billing/tier_manager.py FEATURES dict.
+FEATURES: dict[str, dict] = {
+    "free":  {"batch_active": 1,  "batch_runs_per_day": 3},
+    "pro":   {"batch_active": 5,  "batch_runs_per_day": 20},
+    "power": {"batch_active": 20, "batch_runs_per_day": 100},
+    "team":  {"batch_active": -1, "batch_runs_per_day": -1},
+}
+
+
+def _dt_ms(dt: datetime) -> int:
+    return int(dt.timestamp() * 1000)
+
+
+def _dt_ms_opt(dt: datetime | None) -> int | None:
+    return int(dt.timestamp() * 1000) if dt else None
+
+
+def _to_data_types(values: list[str]) -> list[str]:
+    normalize = {
+        "task": "tasks",           "tasks": "tasks",
+        "note": "notes",           "notes": "notes",
+        "timeline": "timelines",   "timelines": "timelines",   "timelineEvents": "timelines",
+        "project": "projects",     "projects": "projects",
+    }
+    seen: set[str] = set()
+    result: list[str] = []
+    for v in values:
+        mapped = normalize.get(v)
+        if mapped and mapped not in seen:
+            seen.add(mapped)
+            result.append(mapped)
+    return result
+
+
+def _enforce_agent_limit(tier: str, current_count: int) -> int:
+    limit: int = FEATURES.get(tier, FEATURES["free"])["batch_active"]
+    if limit != -1 and current_count >= limit:
+        raise HTTPException(
+            status_code=status.HTTP_403_FORBIDDEN,
+            detail=f"Agent limit ({limit}) reached for your tier. Upgrade to create more.",
+        )
+    return limit
+
+
+async def _enforce_run_frequency(tier: str, user_id: str) -> None:
+    limit: int = FEATURES.get(tier, FEATURES["free"])["batch_runs_per_day"]
+    if limit == -1:
+        return
+    today_start = datetime.now(timezone.utc).replace(
+        hour=0, minute=0, second=0, microsecond=0
+    )
+    async with async_session() as db:
+        result = await db.execute(
+            select(func.count(AgentRunLog.id)).where(
+                AgentRunLog.user_id == user_id,
+                AgentRunLog.started_at >= today_start,
+            )
+        )
+        runs_today: int = result.scalar_one()
+
+    if runs_today >= limit:
+        raise HTTPException(
+            status_code=status.HTTP_402_PAYMENT_REQUIRED,
+            detail=f"Daily batch run limit ({limit}) reached for your tier.",
+        )
+
+
+# ── Catalog ───────────────────────────────────────────────────────────
+
+@router.get("/catalog")
+async def get_agent_catalog(
+    x_user_id: str = Header(..., alias="X-User-Id"),
+) -> list[dict]:
+    return [
+        {
+            "type": "local_directory",
+            "name": "Local Directory Monitor",
+            "description": "Watches local directories, extracts data from files using AI",
+        },
+        {
+            "type": "gmail",
+            "name": "Gmail Connector",
+            "description": "Scans Gmail inbox, extracts tasks/notes from emails",
+        },
+        {
+            "type": "teams",
+            "name": "Microsoft Teams Connector",
+            "description": "Monitors Teams messages, extracts action items",
+        },
+        {
+            "type": "outlook",
+            "name": "Outlook Connector",
+            "description": "Scans Outlook inbox, extracts tasks/notes",
+        },
+    ]
+
+
+# ── Can-create check ─────────────────────────────────────────────────
+
+@router.post("/can-create")
+async def can_create_agent(
+    body: dict,
+    x_user_id: str = Header(..., alias="X-User-Id"),
+    x_user_tier: str = Header("free", alias="X-User-Tier"),
+) -> dict:
+    active_agents = body.get("active_agents", 0)
+    limit: int = FEATURES.get(x_user_tier, FEATURES["free"])["batch_active"]
+    allowed = limit == -1 or active_agents < limit
+    return {
+        "allowed": allowed,
+        "tier": x_user_tier,
+        "active_agents": active_agents,
+        "limit": limit,
+    }
+
+
+# ── Trigger ──────────────────────────────────────────────────────────
+
+@router.post("/trigger", status_code=status.HTTP_202_ACCEPTED)
+async def trigger_agent_run(
+    body: dict,
+    x_user_id: str = Header(..., alias="X-User-Id"),
+    x_user_tier: str = Header("free", alias="X-User-Tier"),
+) -> dict:
+    """Trigger a local agent run — creates run log and dispatches via Redis."""
+    active_agents = body.get("active_agents", 0)
+    _enforce_agent_limit(x_user_tier, active_agents)
+    await _enforce_run_frequency(x_user_tier, x_user_id)
+
+    stable_agent_id = body.get("agent_id") or str(uuid.uuid4())
+
+    if is_agent_running(stable_agent_id):
+        raise HTTPException(
+            status_code=status.HTTP_409_CONFLICT,
+            detail="Agent is already running.",
+        )
+
+    # Create run log in DB
+    async with async_session() as db:
+        run_log = AgentRunLog(
+            agent_id=stable_agent_id,
+            agent_type="local",
+            user_id=x_user_id,
+            status="running",
+        )
+        db.add(run_log)
+        await db.commit()
+        await db.refresh(run_log)
+        run_log_id = run_log.id
+
+    run_context = {
+        "type": "agent_batch",
+        "run_id": run_log_id,
+        "agent_id": stable_agent_id,
+    }
+
+    # Dispatch to the Redis consumer for processing
+    trigger_data = {
+        "type": "agent_trigger",
+        "directory": body.get("directory", ""),
+        "directory_paths": [body.get("directory", "")] if body.get("directory") else [],
+        "data_types": _to_data_types(body.get("what_to_extract", [])),
+        "file_extensions": body.get("file_extensions", []),
+        "prompt_template": body.get("custom_agent_prompt", ""),
+        "device_id": body.get("device_id", ""),
+        "run_context": run_context,
+    }
+
+    channel = batch_request_channel(x_user_id)
+    await redis_client.publish(channel, json.dumps(trigger_data))
+
+    return {
+        "id": run_log_id,
+        "agent_id": stable_agent_id,
+        "agent_type": "local",
+        "status": "running",
+        "items_processed": 0,
+        "items_created": 0,
+        "errors": [],
+        "started_at": _dt_ms(run_log.started_at),
+        "completed_at": None,
+    }

services/batch-agent/app/ws_context.py

@@ -0,0 +1,135 @@
+"""WebSocket context for Batch Agent Service — Redis-based tool call round-trip.
+
+Same pattern as services/chat/app/ws_context.py: publishes tool_call frames
+to Redis ws:out:{user_id} and awaits BRPOP on tool:result:{call_id}.
+
+Additionally provides set_client_executor / clear_client_executor stubs
+for backward compatibility with the agent_runner code (which originally
+used a DeviceConnectionManager callback).  In the microservice world these
+are no-ops — execute_on_client() always uses the Redis path.
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+from contextvars import ContextVar
+from typing import Any, Callable, Coroutine
+from uuid import uuid4
+
+from shared.redis import redis_client, tool_result_key, ws_out_channel
+
+logger = logging.getLogger(__name__)
+
+_TOOL_CALL_TIMEOUT = 30  # seconds — BRPOP timeout
+
+# Per-request user_id context var (set before agent run)
+_current_user_id: ContextVar[str | None] = ContextVar("_current_user_id", default=None)
+
+# Optional collector for debug / logging
+_tool_result_collector: ContextVar[list[dict] | None] = ContextVar(
+    "_tool_result_collector", default=None
+)
+
+
+def set_current_user(user_id: str) -> None:
+    _current_user_id.set(user_id)
+
+
+def clear_current_user() -> None:
+    _current_user_id.set(None)
+
+
+def set_tool_result_collector(lst: list[dict]) -> None:
+    _tool_result_collector.set(lst)
+
+
+def clear_tool_result_collector() -> None:
+    _tool_result_collector.set(None)
+
+
+# ── Compatibility shims ──────────────────────────────────────────────────
+# agent_runner.py originally called set_client_executor / clear_client_executor
+# with a DeviceConnectionManager callback.  In the microservice world the
+# Redis-based execute_on_client replaces this, so these are no-ops that
+# keep the agent_runner code unchanged.
+
+def set_client_executor(fn: Callable[[dict], Coroutine[Any, Any, dict]] | None) -> None:
+    """No-op — kept for agent_runner compatibility."""
+    pass
+
+
+def clear_client_executor() -> None:
+    """No-op — kept for agent_runner compatibility."""
+    pass
+
+
+async def execute_on_client(
+    action: str,
+    table: str | None = None,
+    data: dict[str, Any] | None = None,
+    filters: dict[str, Any] | None = None,
+    vector: list[float] | None = None,
+    limit: int | None = None,
+) -> dict[str, Any]:
+    """Send a tool_call to Electron via Redis and await the result.
+
+    1. Build tool_call payload
+    2. Publish to ws:out:{user_id} (WS Gateway forwards to Electron)
+    3. BRPOP on tool:result:{call_id} (WS Gateway pushes when Electron replies)
+    4. Return result dict
+
+    Raises RuntimeError if no user_id is set or if the call times out.
+    """
+    user_id = _current_user_id.get()
+    if not user_id:
+        raise RuntimeError(
+            "execute_on_client() called without a user_id — "
+            "set_current_user() must be called first."
+        )
+
+    call_id = str(uuid4())
+    payload: dict[str, Any] = {
+        "type": "tool_call",
+        "id": call_id,
+        "action": action,
+    }
+    if table is not None:
+        payload["table"] = table
+    if data is not None:
+        payload["data"] = data
+    if filters is not None:
+        payload["filters"] = {k: v for k, v in filters.items() if v is not None}
+    if vector is not None:
+        payload["vector"] = vector
+    if limit is not None:
+        payload["limit"] = limit
+
+    # Publish tool_call to WS Gateway → Electron
+    channel = ws_out_channel(user_id)
+    await redis_client.publish(channel, json.dumps(payload))
+
+    # Wait for Electron's tool_result
+    result_key = tool_result_key(call_id)
+    response = await redis_client.brpop(result_key, timeout=_TOOL_CALL_TIMEOUT)
+
+    if response is None:
+        raise RuntimeError(
+            f"Tool call {call_id} timed out after {_TOOL_CALL_TIMEOUT}s — "
+            f"device may be offline or unresponsive."
+        )
+
+    # response is (key, value) tuple
+    _, raw = response
+    result = json.loads(raw)
+
+    # Collect for debug if requested
+    collector = _tool_result_collector.get(None)
+    if collector is not None:
+        collector.append({
+            "action": action,
+            "table": table,
+            "data": result,
+        })
+
+    return result

services/batch-agent/requirements.txt

@@ -0,0 +1,20 @@
+fastapi>=0.115.0
+uvicorn[standard]>=0.34.0
+gunicorn>=22.0.0
+pydantic>=2.10.0
+pydantic-settings>=2.7.0
+sqlalchemy>=2.0.0
+asyncpg>=0.30.0
+redis>=5.0.0
+cryptography>=42.0.0
+python-dotenv>=1.0.0
+langchain-core>=0.3.0
+langchain-openai>=0.3.0
+langchain-litellm>=0.3.0
+litellm>=1.50.0
+openai>=1.50.0
+httpx>=0.27.0
+croniter>=2.0.0
+google-api-python-client>=2.130.0
+google-auth>=2.30.0
+msal>=1.28.0

services/billing/README.md

@@ -0,0 +1,15 @@
+# Billing Service
+
+Owns: Stripe integration, tier management, subscription CRUD.
+
+## Tables owned (write)
+- `subscriptions`
+
+## Endpoints
+- `POST /billing/checkout`
+- `POST /billing/webhook` (Stripe, no JWT auth)
+- `GET /billing/subscription`
+- `DELETE /billing/subscription`
+
+## Redis channels
+- Publish: `tier:changed:{user_id}` on tier change

services/chat/Dockerfile

@@ -0,0 +1,36 @@
+# ── builder ──────────────────────────────────────────────────────────────────
+FROM python:3.12-slim AS builder
+
+WORKDIR /build
+
+COPY services/chat/requirements.txt ./requirements.txt
+RUN pip install --upgrade pip && \
+    pip install --no-cache-dir --prefix=/install -r requirements.txt
+
+# ── runtime ──────────────────────────────────────────────────────────────────
+FROM python:3.12-slim AS runtime
+
+RUN addgroup --system appgroup && adduser --system --ingroup appgroup appuser
+
+WORKDIR /app
+
+COPY --from=builder /install /usr/local
+
+# Shared module
+COPY shared/ shared/
+
+# Service source
+COPY services/chat/app/ app/
+
+RUN chown -R appuser:appgroup /app
+
+USER appuser
+
+EXPOSE 8000
+
+# Chat service is CPU-bound (LLM calls) — use multiple workers
+CMD ["gunicorn", "app.main:app", \
+     "-k", "uvicorn.workers.UvicornWorker", \
+     "--bind", "0.0.0.0:8000", \
+     "--workers", "2", \
+     "--timeout", "120"]

services/chat/README.md

@@ -0,0 +1,21 @@
+# Chat Service
+
+Owns: deep_agent (home + floating chat), memory middleware, domain agents
+(task, note, project, timeline), LLM orchestration.
+
+## Tables owned
+- `memory_core`
+- `memory_associative`
+- `memory_episodic`
+- `memory_proactive`
+
+## Tables read (cross-service)
+- `users` (for encryption_key — memory decryption)
+
+## Endpoints
+- `POST /chat` (REST fallback)
+
+## Redis channels
+- Subscribe: `chat:request:{user_id}`
+- Publish: `ws:out:{user_id}` (stream frames + tool calls)
+- BRPOP: `tool:result:{call_id}` (30s timeout)

services/chat/app/agents/__init__.py

@@ -0,0 +1 @@
+"""Chat Service domain agents."""

services/chat/app/agents/note_agent.py

@@ -0,0 +1,142 @@
+"""Note agent — Markdown note management (list, get, create, update, delete).
+
+Adapted for Chat Service: import from app.ws_context and app.llm.
+"""
+
+from __future__ import annotations
+
+import re
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.llm import embed
+from app.ws_context import execute_on_client
+
+_UUID_RE = re.compile(
+    r"^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$"
+)
+
+
+def _is_uuid(value: str) -> bool:
+    return bool(_UUID_RE.match(value))
+
+NOTE_SYSTEM_PROMPT = (
+    "You are a note-taking assistant. You help users create, retrieve, update,\n"
+    "and delete Markdown notes in their workspace.\n\n"
+    "Rules:\n"
+    "  - content is always Markdown; preserve formatting when updating\n"
+    "  - project_id is optional; link a note to a project when mentioned\n"
+    "  - When updating, call get_note first if you need to read existing content\n"
+    "    before appending or replacing sections\n"
+    "  - list_notes without project_id returns all notes; scope with project_id\n"
+    "    when the user is working within a specific project\n"
+    "  - project_id must be a UUID; if you only know a project name, do not pass it as project_id\n"
+    "  - Do not fabricate note content — reflect what the user provides or what\n"
+    "    is already in the note (retrieved via get_note)."
+)
+
+
+@tool
+async def list_notes(project_id: str = "") -> str:
+    """List notes, optionally scoped to a project by project_id."""
+    normalized_project_id = project_id if (project_id and _is_uuid(project_id)) else ""
+    result = await execute_on_client(
+        action="select",
+        table="notes",
+        filters={"projectId": normalized_project_id or None},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No notes found."
+    lines = [f"- {r['title']} (id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} note(s):\n" + "\n".join(lines)
+
+
+@tool
+async def get_note(note_id: str) -> str:
+    """Fetch a single note by its UUID to read its full Markdown content."""
+    result = await execute_on_client(action="get", table="notes", data={"id": note_id})
+    row = result.get("row")
+    if not row:
+        return f"Note {note_id} not found."
+    return f"Note '{row['title']}' (id: {row['id']}):\n\n{row['content']}"
+
+
+@tool
+async def create_note(
+    title: str,
+    content: str,
+    project_id: str = "",
+) -> str:
+    """Create a new note.
+    title: note heading (required)
+    content: Markdown body text (required)
+    project_id: optional UUID linking this note to a project
+    """
+    result = await execute_on_client(
+        action="insert",
+        table="notes",
+        data={
+            "title": title,
+            "content": content,
+            "projectId": project_id or None,
+        },
+    )
+    row = result["row"]
+    # Index the note content in the vector store.
+    vector = await embed(content)
+    await execute_on_client(
+        action="vector_upsert",
+        data={"id": row["id"], "projectId": row.get("projectId"), "content": content},
+        vector=vector,
+    )
+    return f"Note created: '{row['title']}' (id: {row['id']})."
+
+
+@tool
+async def update_note(
+    note_id: str,
+    title: str = "",
+    content: str = "",
+) -> str:
+    """Update an existing note. Only pass fields that should change.
+    note_id: UUID of the note (required)
+    If you need to preserve existing content, call get_note first.
+    """
+    updates: dict[str, Any] = {}
+    if title:
+        updates["title"] = title
+    if content:
+        updates["content"] = content
+    result = await execute_on_client(
+        action="update",
+        table="notes",
+        data={"id": note_id, "updates": updates},
+    )
+    row = result["row"]
+    # Re-index if content changed.
+    if content:
+        vector = await embed(content)
+        await execute_on_client(
+            action="vector_upsert",
+            data={"id": note_id, "projectId": row.get("projectId"), "content": content},
+            vector=vector,
+        )
+    return f"Note updated: '{row['title']}' (id: {row['id']})."
+
+
+@tool
+async def delete_note(note_id: str) -> str:
+    """Delete a note permanently by its UUID."""
+    await execute_on_client(action="delete", table="notes", data={"id": note_id})
+    return f"Note {note_id} deleted."
+
+
+NOTE_TOOLS: list[Any] = [
+    list_notes,
+    get_note,
+    create_note,
+    update_note,
+    delete_note,
+]

services/chat/app/agents/project_agent.py

@@ -0,0 +1,146 @@
+"""Project agent — full lifecycle management (list, get, create, update, archive, delete).
+
+Adapted for Chat Service: import from app.ws_context instead of app.core.ws_context.
+"""
+
+from __future__ import annotations
+
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.ws_context import execute_on_client
+
+PROJECT_SYSTEM_PROMPT = (
+    "You are a project management assistant. You help users create, find,\n"
+    "update, and archive projects in their workspace.\n\n"
+    "Rules:\n"
+    "  - status must be one of: active, archived\n"
+    "  - client_id is optional; link to a client only when explicitly mentioned\n"
+    "  - ai_summary is populated only when the user asks for a project summary;\n"
+    "    derive it from context data — do not fabricate content\n"
+    "  - Use list_projects for scoped queries; list_all_projects only when the\n"
+    "    user wants a complete cross-client view including archived projects\n"
+    "  - get_project requires a project UUID; resolve the ID first by calling\n"
+    "    list_projects if you only have a project name\n"
+    "  - Prefer archiving (update_project status=archived) over deletion;\n"
+    "    only call delete_project when the user explicitly confirms deletion."
+)
+
+
+@tool
+async def list_projects(
+    client_id: str = "",
+    include_archived: int = 0,
+) -> str:
+    """List projects, optionally filtered by client_id.
+    include_archived: 1 to include archived projects, 0 for active only (default).
+    """
+    result = await execute_on_client(
+        action="select",
+        table="projects",
+        filters={
+            "clientId": client_id or None,
+            "includeArchived": bool(include_archived),
+        },
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No projects found."
+    lines = [f"- {r['name']} (status: {r['status']}, id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} project(s):\n" + "\n".join(lines)
+
+
+@tool
+async def list_all_projects() -> str:
+    """List every project regardless of client or status.
+    Use only when the user wants a complete cross-client overview.
+    """
+    result = await execute_on_client(action="select", table="projects")
+    rows = result.get("rows", [])
+    if not rows:
+        return "No projects found."
+    lines = [f"- {r['name']} (status: {r['status']}, id: {r['id']})" for r in rows]
+    return f"All projects ({len(rows)}):\n" + "\n".join(lines)
+
+
+@tool
+async def get_project(project_id: str) -> str:
+    """Fetch a single project by its UUID."""
+    result = await execute_on_client(action="get", table="projects", data={"id": project_id})
+    row = result.get("row")
+    if not row:
+        return f"Project {project_id} not found."
+    return (
+        f"Project: '{row['name']}' (id: {row['id']}, status: {row['status']}, "
+        f"clientId: {row.get('clientId', 'none')})"
+    )
+
+
+@tool
+async def create_project(
+    name: str,
+    client_id: str = "",
+) -> str:
+    """Create a new project.
+    name: human-readable project name (required)
+    client_id: optional UUID of the owning client
+    """
+    result = await execute_on_client(
+        action="insert",
+        table="projects",
+        data={"name": name, "clientId": client_id or None},
+    )
+    row = result["row"]
+    return f"Project created: '{row['name']}' (id: {row['id']})"
+
+
+@tool
+async def update_project(
+    project_id: str,
+    name: str = "",
+    client_id: str = "",
+    status: str = "",
+    ai_summary: str = "",
+) -> str:
+    """Update a project. Only pass fields that should change.
+    project_id: UUID of the project (required)
+    status: active | archived
+    ai_summary: AI-generated summary text (populate only when explicitly requested)
+    """
+    updates: dict[str, Any] = {}
+    if name:
+        updates["name"] = name
+    if client_id:
+        updates["clientId"] = client_id
+    if status:
+        updates["status"] = status
+    if ai_summary:
+        updates["aiSummary"] = ai_summary
+    result = await execute_on_client(
+        action="update",
+        table="projects",
+        data={"id": project_id, "updates": updates},
+    )
+    row = result["row"]
+    return f"Project updated: '{row['name']}' (id: {row['id']}, status: {row['status']})"
+
+
+@tool
+async def delete_project(project_id: str) -> str:
+    """Permanently delete a project and orphan its tasks.
+    IMPORTANT: prefer update_project(status='archived') unless the user
+    has explicitly confirmed they want permanent deletion.
+    """
+    await execute_on_client(action="delete", table="projects", data={"id": project_id})
+    return f"Project {project_id} permanently deleted."
+
+
+PROJECT_TOOLS: list[Any] = [
+    list_projects,
+    list_all_projects,
+    get_project,
+    create_project,
+    update_project,
+    delete_project,
+]

services/chat/app/agents/task_agent.py

@@ -0,0 +1,240 @@
+"""Task agent — full CRUD for tasks and task comments.
+
+Adapted for Chat Service: import from app.ws_context instead of app.core.ws_context.
+"""
+
+from __future__ import annotations
+
+from datetime import datetime, timezone
+import re
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.ws_context import execute_on_client
+
+_UUID_RE = re.compile(
+    r"^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$"
+)
+
+
+def _is_uuid(value: str) -> bool:
+    return bool(_UUID_RE.match(value))
+
+TASK_SYSTEM_PROMPT = (
+    "You are a task management assistant for a project workspace.\n"
+    "You create, update, list, and track tasks and their comments.\n\n"
+    "Rules:\n"
+    "  - status must be one of: todo, in_progress, done\n"
+    "  - priority must be one of: high, medium, low\n"
+    "  - due_date is a Unix timestamp in milliseconds; convert human dates\n"
+    "  - assignees is a JSON-encoded array of strings (e.g. '[\"Alice\",\"Bob\"]')\n"
+    "  - project_id is optional; link to a project when the user mentions one\n"
+    "  - is_ai_suggested: 1 only when proactively proposing a task the user\n"
+    "    did not explicitly request; 0 otherwise\n"
+    "  - is_ai_suggested: 1 only when proactively proposing a task the user did not explicitly request; 0 otherwise\n"
+    "  - Use list_tasks_due_today for 'what's due today' queries\n"
+    "  - For update_task, use -1 for integer fields you do not want to change\n"
+    "  - Always confirm the action in plain, user-friendly language."
+)
+
+
+# ── Task tools ────────────────────────────────────────────────────────
+
+
+@tool
+async def list_tasks(
+    project_id: str = "",
+    status: str = "",
+    search: str = "",
+    order_by: str = "",
+) -> str:
+    """List tasks, optionally filtered by project_id, status (todo|in_progress|done),
+    a search string, or an order_by field name (dueDate|priority|createdAt)."""
+    normalized_project_id = project_id if (project_id and _is_uuid(project_id)) else ""
+    result = await execute_on_client(
+        action="select",
+        table="tasks",
+        filters={
+            "projectId": normalized_project_id or None,
+            "status": status or None,
+            "search": search or None,
+            "orderBy": order_by or None,
+        },
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No tasks found matching the given filters."
+    lines = [
+        f"- {r['title']} (status: {r['status']}, priority: {r['priority']}, id: {r['id']})"
+        for r in rows
+    ]
+    return f"Found {len(rows)} task(s):\n" + "\n".join(lines)
+
+
+@tool
+async def create_task(
+    title: str,
+    description: str = "",
+    status: str = "todo",
+    priority: str = "medium",
+    assignees: str = "[]",
+    due_date: int = 0,
+    project_id: str = "",
+    is_ai_suggested: int = 0,
+) -> str:
+    """Create a new task.
+    title: task title (required)
+    description: optional details
+    status: todo | in_progress | done  (default: todo)
+    priority: high | medium | low  (default: medium)
+    assignees: JSON-encoded array of assignee names, e.g. '["Alice"]'
+    due_date: Unix timestamp in milliseconds; 0 means no due date
+    project_id: optional UUID of the parent project
+    is_ai_suggested: 1 if proactively suggested, 0 if user-requested
+    """
+    result = await execute_on_client(
+        action="insert",
+        table="tasks",
+        data={
+            "title": title,
+            "description": description or None,
+            "status": status,
+            "priority": priority,
+            "assignee": assignees,
+            "dueDate": due_date or None,
+            "projectId": project_id or None,
+            "isAiSuggested": is_ai_suggested,
+        },
+    )
+    row = result["row"]
+    return (
+        f"Task created: '{row['title']}' "
+        f"(id: {row['id']}, status: {row['status']}, priority: {row['priority']})"
+    )
+
+
+@tool
+async def update_task(
+    task_id: str,
+    title: str = "",
+    description: str = "",
+    status: str = "",
+    priority: str = "",
+    assignees: str = "",
+    due_date: int = -1,
+    project_id: str = "",
+) -> str:
+    """Update fields on an existing task. Only pass fields you want to change.
+    task_id: the task's UUID (required)
+    due_date: -1 means unchanged; 0 clears the due date; any positive value sets it
+    """
+    updates: dict[str, Any] = {}
+    if title:
+        updates["title"] = title
+    if description:
+        updates["description"] = description
+    if status:
+        updates["status"] = status
+    if priority:
+        updates["priority"] = priority
+    if assignees:
+        updates["assignee"] = assignees
+    if due_date != -1:
+        updates["dueDate"] = due_date or None
+    if project_id:
+        updates["projectId"] = project_id
+    result = await execute_on_client(
+        action="update",
+        table="tasks",
+        data={"id": task_id, "updates": updates},
+    )
+    row = result["row"]
+    return f"Task updated: '{row['title']}' (id: {row['id']}, status: {row['status']})"
+
+
+@tool
+async def delete_task(task_id: str) -> str:
+    """Delete a task permanently by its UUID."""
+    await execute_on_client(action="delete", table="tasks", data={"id": task_id})
+    return f"Task {task_id} deleted."
+
+
+@tool
+async def list_tasks_due_today() -> str:
+    """List all tasks whose due date falls on today's date."""
+    now = datetime.now(tz=timezone.utc)
+    start_ms = int(datetime(now.year, now.month, now.day, tzinfo=timezone.utc).timestamp() * 1000)
+    end_ms = start_ms + 86_400_000 - 1  # last ms of today
+    result = await execute_on_client(
+        action="select",
+        table="tasks",
+        filters={"dueDateFrom": start_ms, "dueDateTo": end_ms},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No tasks are due today."
+    lines = [
+        f"- {r['title']} (priority: {r['priority']}, status: {r['status']}, id: {r['id']})"
+        for r in rows
+    ]
+    return f"Tasks due today ({len(rows)}):\n" + "\n".join(lines)
+
+
+# ── Task comment tools ────────────────────────────────────────────────
+
+
+@tool
+async def list_task_comments(task_id: str) -> str:
+    """List all comments on a task by its UUID."""
+    result = await execute_on_client(
+        action="select",
+        table="taskComments",
+        filters={"taskId": task_id},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return f"No comments found for task {task_id}."
+    lines = [f"- [{r['author']}]: {r['content']} (id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} comment(s):\n" + "\n".join(lines)
+
+
+@tool
+async def add_task_comment(task_id: str, author: str, content: str) -> str:
+    """Add a comment to a task.
+    task_id: UUID of the task to comment on
+    author: name or ID of the comment author
+    content: comment text
+    """
+    result = await execute_on_client(
+        action="insert",
+        table="taskComments",
+        data={"taskId": task_id, "author": author, "content": content},
+    )
+    row = result.get("row", {})
+    row_author = row.get("author", author)
+    row_task_id = row.get("taskId") or row.get("task_id") or task_id
+    row_comment_id = row.get("id", "unknown")
+    return f"Comment added by {row_author} on task {row_task_id} (comment id: {row_comment_id})."
+
+
+@tool
+async def delete_task_comment(comment_id: str) -> str:
+    """Delete a task comment by its UUID."""
+    await execute_on_client(action="delete", table="taskComments", data={"id": comment_id})
+    return f"Comment {comment_id} deleted."
+
+
+# ── Agent ─────────────────────────────────────────────────────────────
+
+
+TASK_TOOLS: list[Any] = [
+    list_tasks,
+    create_task,
+    update_task,
+    delete_task,
+    list_tasks_due_today,
+    list_task_comments,
+    add_task_comment,
+    delete_task_comment,
+]

services/chat/app/agents/timeline_agent.py

@@ -0,0 +1,117 @@
+"""Timeline agent — project milestone management (list, create, update, delete).
+
+Adapted for Chat Service: import from app.ws_context instead of app.core.ws_context.
+"""
+
+from __future__ import annotations
+
+import re
+from typing import Any
+
+from langchain_core.tools import tool
+
+from app.ws_context import execute_on_client
+
+_UUID_RE = re.compile(
+    r"^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$"
+)
+
+
+def _is_uuid(value: str) -> bool:
+    return bool(_UUID_RE.match(value))
+
+TIMELINE_SYSTEM_PROMPT = (
+    "You are a project timeline assistant. Timelines are milestone dates that\n"
+    "track progress on a project — they are not calendar events.\n\n"
+    "Rules:\n"
+    "  - project_id is REQUIRED for every create; confirm with the user if unknown\n"
+    "  - For listing, project_id must be a UUID; never pass plain names as project_id\n"
+    "  - date is a Unix timestamp in milliseconds; convert human-readable dates\n"
+    "  - is_ai_suggested: 1 when proactively proposing a timeline, 0 otherwise\n"
+    "  - is_ai_suggested: 1 when proactively proposing a timeline, 0 otherwise\n"
+    "  - For update_timeline, use -1 for integer fields you do not want to change\n"
+    "  - Listing without a project_id returns all timelines across projects\n"
+    "  - Always echo the title and formatted date in your confirmation."
+)
+
+
+@tool
+async def list_timelines(project_id: str = "") -> str:
+    """List timelines. Provide project_id to scope to a specific project."""
+    normalized_project_id = project_id if (project_id and _is_uuid(project_id)) else ""
+    result = await execute_on_client(
+        action="select",
+        table="timelines",
+        filters={"projectId": normalized_project_id or None},
+    )
+    rows = result.get("rows", [])
+    if not rows:
+        return "No timelines found."
+    lines = [f"- {r['title']} (date: {r['date']}, id: {r['id']})" for r in rows]
+    return f"Found {len(rows)} timeline(s):\n" + "\n".join(lines)
+
+
+@tool
+async def create_timeline(
+    project_id: str,
+    title: str,
+    date: int,
+    is_ai_suggested: int = 0,
+) -> str:
+    """Create a project timeline (milestone).
+    project_id: REQUIRED UUID of the parent project
+    title: descriptive name for the milestone
+    date: Unix timestamp in milliseconds
+    is_ai_suggested: 1 if proactively suggested, 0 if user-requested
+    """
+    result = await execute_on_client(
+        action="insert",
+        table="timelines",
+        data={
+            "projectId": project_id,
+            "title": title,
+            "date": date,
+            "isAiSuggested": is_ai_suggested,
+        },
+    )
+    row = result["row"]
+    return f"Timeline created: '{row['title']}' (id: {row['id']}, date: {row['date']})"
+
+
+@tool
+async def update_timeline(
+    timeline_id: str,
+    title: str = "",
+    date: int = -1,
+) -> str:
+    """Update a timeline. Only pass fields that should change.
+    timeline_id: UUID of the timeline (required)
+    date: -1 means unchanged; any other value sets the new date (ms timestamp)
+    """
+    updates: dict[str, Any] = {}
+    if title:
+        updates["title"] = title
+    if date != -1:
+        updates["date"] = date
+    result = await execute_on_client(
+        action="update",
+        table="timelines",
+        data={"id": timeline_id, "updates": updates},
+    )
+    row = result["row"]
+    return f"Timeline updated: '{row['title']}' (id: {row['id']})"
+
+
+@tool
+async def delete_timeline(timeline_id: str) -> str:
+    """Delete a timeline permanently by its UUID."""
+    await execute_on_client(action="delete", table="timelines", data={"id": timeline_id})
+    return f"Timeline {timeline_id} deleted."
+
+
+TIMELINE_TOOLS: list[Any] = [
+    list_timelines,
+    create_timeline,
+    update_timeline,
+    delete_timeline,
+]

services/chat/app/deep_agent.py

@@ -0,0 +1,883 @@
+"""Single-agent runners for home and floating chat contexts.
+
+Adapted from app/core/deep_agent.py for the Chat Service.
+Import paths changed to use local app modules and shared/.
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+import re
+from datetime import date
+from collections.abc import AsyncGenerator
+from typing import Any, Literal
+
+from langchain_core.messages import AIMessage, HumanMessage, SystemMessage, ToolMessage
+from langchain_core.tools import tool
+
+from app.agents.note_agent import NOTE_TOOLS
+from app.agents.project_agent import PROJECT_TOOLS
+from app.agents.task_agent import TASK_TOOLS
+from app.agents.timeline_agent import TIMELINE_TOOLS
+from app.llm import get_llm
+from app.memory_middleware import MemoryMiddleware
+from app.ws_context import clear_tool_result_collector, execute_on_client, set_tool_result_collector
+from app import tracing
+from shared.db import async_session
+
+logger = logging.getLogger(__name__)
+
+FloatingDomainType = Literal["task", "timeline", "project", "node"]
+FloatingDomainSection = Literal["task", "timeline", "note"]
+
+_HOME_SINGLE_AGENT_SYSTEM = (
+    "You are the home assistant with direct access to all tools: tasks, projects, notes, timelines, and memory tools. "
+    "Always use tools for factual data retrieval before answering. "
+    "When the user asks to remember, forget, or update what you know about them, use memory tools. "
+    "If context.context.resolved_project_id exists, use it as project_id for scoped list calls. "
+    "Return markdown and use tags when relevant: <project>[ids]</project>, <task>[ids]</task>, "
+    "<note>[ids]</note>, <timeline>[ids]</timeline>, <chart>{json}</chart>. "
+    "When listing tasks or timelines, each id tag must be on its own line with no prefix/suffix text. "
+    "Never put titles, priorities, or dates on the same line as <task> or <timeline> tags. "
+    "For questions about upcoming timelines (e.g. 'prossimi eventi'), include only future items in the current month unless the user asks a different range. "
+    "For upcoming tasks, after tag lines add a short recommendation based on due date and priority."
+)
+
+_FLOATING_SINGLE_AGENT_SYSTEM = (
+    "You are the floating assistant with direct access to all tools: tasks, projects, notes, timelines, and memory tools. "
+    "Stay focused on the floating scope in context.scope and answer concisely. "
+    "Return plain text only. Do not output XML/HTML-like tags such as <task>, <project>, <note>, <timeline>, or any bracketed id tag wrappers. "
+    "Always use tools for factual data retrieval before answering. "
+    "When the user asks to remember, forget, or update what you know about them, use memory tools. "
+    "If context.context.resolved_project_id exists, use it as project_id for scoped list calls. "
+)
+
+_FLOATING_DOMAIN_CLASSIFIER_SYSTEM = (
+    "You are a strict domain classifier for websocket floating requests. "
+    "Return ONLY a JSON object with keys: type, id, section. "
+    "Allowed type values: task, timeline, project, node. "
+    "Allowed section values: task, timeline, note, or null. "
+    "Rules: infer from user message intent first; do not blindly trust scope.type. "
+    "If user asks tasks/timeline/notes for a project, set type=project and section accordingly. "
+    "If project id is unknown but context.resolved_project_id exists, use it as id. "
+    "If id is unknown, use null. "
+    "No markdown, no prose, JSON only."
+)
+
+
+def _as_text(content: Any) -> str:
+    if content is None:
+        return ""
+    if isinstance(content, str):
+        return content
+    if isinstance(content, list):
+        parts: list[str] = []
+        for item in content:
+            if isinstance(item, str):
+                parts.append(item)
+            elif isinstance(item, dict):
+                text = item.get("text")
+                if isinstance(text, str):
+                    parts.append(text)
+        return "".join(parts)
+    return str(content)
+
+
+def _candidate_tokens(message: str) -> list[str]:
+    tokens = re.findall(r"[a-zA-Z0-9_-]+", message.lower())
+    return [token for token in tokens if len(token) >= 3]
+
+
+async def _resolve_project_id_from_message(message: str) -> str | None:
+    """Resolve likely project UUID from user message using client project list."""
+    try:
+        result = await execute_on_client(action="select", table="projects")
+    except Exception as exc:
+        logger.warning("deep_agent: project resolve select failed: %s", exc)
+        return None
+
+    rows = result.get("rows", [])
+    if not isinstance(rows, list) or not rows:
+        return None
+
+    tokens = _candidate_tokens(message)
+    scored: list[tuple[int, dict[str, Any]]] = []
+    for row in rows:
+        if not isinstance(row, dict):
+            continue
+        name = str(row.get("name", "")).lower()
+        score = sum(1 for token in tokens if token in name)
+        if score > 0:
+            scored.append((score, row))
+
+    if not scored:
+        return None
+
+    scored.sort(key=lambda item: item[0], reverse=True)
+    top_score = scored[0][0]
+    top_rows = [row for score, row in scored if score == top_score]
+    if len(top_rows) != 1:
+        return None
+
+    project_id = top_rows[0].get("id")
+    return project_id if isinstance(project_id, str) else None
+
+
+def _needs_project_resolution(message: str) -> bool:
+    lowered = message.lower()
+    return any(keyword in lowered for keyword in ["project", "progetto", "progetti", "whitelist"])
+
+
+async def _prepare_context(message: str, context: dict[str, Any]) -> dict[str, Any]:
+    prepared = dict(context)
+    if _needs_project_resolution(message):
+        resolved_project_id = await _resolve_project_id_from_message(message)
+        if resolved_project_id:
+            prepared["resolved_project_id"] = resolved_project_id
+            logger.info("deep_agent: resolved_project_id=%s", resolved_project_id)
+    return prepared
+
+
+def _all_tools() -> list[Any]:
+    return [*TASK_TOOLS, *PROJECT_TOOLS, *NOTE_TOOLS, *TIMELINE_TOOLS]
+
+
+def _trace_id_from_context(context: dict[str, Any]) -> str | None:
+    debug = context.get("_debug")
+    if isinstance(debug, dict):
+        request_id = debug.get("request_id")
+        if isinstance(request_id, str) and request_id:
+            return request_id
+    return None
+
+
+def _context_for_model(context: dict[str, Any]) -> dict[str, Any]:
+    sanitized = dict(context)
+    sanitized.pop("_debug", None)
+    return sanitized
+
+
+_TAG_LINE_RE = re.compile(r"<(task|timeline)>\[[^\]]+\]</\1>")
+_TIMELINE_DMY_RE = re.compile(r"(?P<d>\d{2})/(?P<m>\d{2})/(?P<y>\d{4})")
+
+
+def _is_upcoming_timeline_query(message: str) -> bool:
+    lowered = message.lower()
+    has_upcoming = "prossim" in lowered or "upcoming" in lowered or "next" in lowered
+    has_timeline_topic = any(
+        token in lowered
+        for token in ("event", "evento", "eventi", "timeline", "milestone", "scaden")
+    )
+    return has_upcoming and has_timeline_topic
+
+
+def _timeline_date_in_current_month_or_future(dmy: str) -> bool:
+    match = _TIMELINE_DMY_RE.search(dmy)
+    if not match:
+        return True
+    try:
+        parsed = date(
+            int(match.group("y")),
+            int(match.group("m")),
+            int(match.group("d")),
+        )
+    except ValueError:
+        return True
+
+    today = date.today()
+    return parsed >= today and parsed.year == today.year and parsed.month == today.month
+
+
+def _normalize_tagged_list_lines(text: str, message: str) -> str:
+    if not text:
+        return text
+
+    upcoming_timeline_only = _is_upcoming_timeline_query(message)
+    output_lines: list[str] = []
+
+    for line in text.splitlines():
+        matches = list(_TAG_LINE_RE.finditer(line))
+        if not matches:
+            output_lines.append(line)
+            continue
+
+        had_non_tag_text = _TAG_LINE_RE.sub("", line).strip(" -\t0123456789.*:)")
+        if not had_non_tag_text and len(matches) == 1:
+            tag_text = matches[0].group(0)
+            if (
+                upcoming_timeline_only
+                and "<timeline>" in tag_text
+                and not _timeline_date_in_current_month_or_future(line)
+            ):
+                continue
+            output_lines.append(tag_text)
+            continue
+
+        for match in matches:
+            tag_text = match.group(0)
+            if (
+                upcoming_timeline_only
+                and "<timeline>" in tag_text
+                and not _timeline_date_in_current_month_or_future(line)
+            ):
+                continue
+            output_lines.append(tag_text)
+
+    return "\n".join(output_lines)
+
+
+_GENERIC_TAG_RE = re.compile(r"</?(task|project|note|timeline|chart)>", re.IGNORECASE)
+_BRACKETED_ID_RE = re.compile(r"\[(?:[0-9a-fA-F-]{8,}|[A-Za-z0-9_-]{8,})\]")
+_FLOATING_EMPTY_FALLBACK = "No results found."
+
+
+def _strip_floating_markup_fragment(text: str) -> str:
+    if not text:
+        return text
+    cleaned = _GENERIC_TAG_RE.sub("", text)
+    return _BRACKETED_ID_RE.sub("", cleaned)
+
+
+def _strip_floating_markup(text: str) -> str:
+    """Ensure floating responses stay plain text with no XML-like tag wrappers."""
+    if not text:
+        return text
+
+    cleaned = _strip_floating_markup_fragment(text)
+    lines = [re.sub(r"[ \t]{2,}", " ", line).strip() for line in cleaned.splitlines()]
+    return "\n".join(line for line in lines if line)
+
+
+def _fallback_from_raw_floating_text(raw_text: str) -> str:
+    fallback = _strip_floating_markup_fragment(raw_text or "")
+    fallback = re.sub(r"[ \t]{2,}", " ", fallback).strip()
+    return fallback or _FLOATING_EMPTY_FALLBACK
+
+
+class _FloatingStreamSanitizer:
+    """Streaming sanitizer that removes floating markup without buffering the full answer."""
+
+    def __init__(self) -> None:
+        self._pending = ""
+
+    @staticmethod
+    def _split_safe_boundary(text: str) -> tuple[str, str]:
+        boundary = len(text)
+
+        last_lt = text.rfind("<")
+        if last_lt != -1 and ">" not in text[last_lt:]:
+            boundary = min(boundary, last_lt)
+
+        last_lb = text.rfind("[")
+        if last_lb != -1 and "]" not in text[last_lb:]:
+            boundary = min(boundary, last_lb)
+
+        if boundary == len(text):
+            return text, ""
+        return text[:boundary], text[boundary:]
+
+    def feed(self, chunk: str) -> str:
+        combined = f"{self._pending}{chunk}"
+        safe_text, self._pending = self._split_safe_boundary(combined)
+        return _strip_floating_markup_fragment(safe_text)
+
+    def finalize(self) -> str:
+        tail = re.sub(r"<[^>\n]*$", "", self._pending)
+        tail = re.sub(r"\[[^\]\n]*$", "", tail)
+        self._pending = ""
+        return _strip_floating_markup_fragment(tail)
+
+
+def _normalize_memory_label(path_or_label: str) -> str:
+    value = path_or_label.strip()
+    if value.startswith("/memories/"):
+        value = value[len("/memories/"):]
+    value = value.strip("/")
+    return value
+
+
+def _memory_tools(user_id: str, trace_id: str | None) -> list[Any]:
+    @tool
+    async def memory_list_blocks() -> str:
+        """List all core memory blocks currently stored for the user."""
+        logger.info("deep_agent: memory_list_blocks trace=%s user=%s", trace_id or "-", user_id)
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            blocks = await memory.list_core_blocks(user_id)
+        if not blocks:
+            return "No memory blocks found."
+        lines = [f"- {b['label']}: {b['value']}" for b in blocks]
+        return "Memory blocks:\n" + "\n".join(lines)
+
+    @tool
+    async def memory_get(path_or_label: str) -> str:
+        """Get one memory block by label or /memories/<label> path."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_get trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            value = await memory.get_core_block(user_id, label)
+        if value is None:
+            return f"Memory block '{label}' not found."
+        return f"Memory block '{label}':\n{value}"
+
+    @tool
+    async def memory_create(path_or_label: str, value: str) -> str:
+        """Create or overwrite a memory block value by label or /memories/<label> path."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_create trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            await memory.update_core(user_id, label, value, trace_id=trace_id)
+        return f"Memory block '{label}' saved."
+
+    @tool
+    async def memory_append(path_or_label: str, content: str) -> str:
+        """Append content to a memory block, creating it if missing."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_append trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            await memory.append_core(user_id, label, content)
+        return f"Memory block '{label}' appended."
+
+    @tool
+    async def memory_replace(path_or_label: str, old_string: str, new_string: str) -> str:
+        """Replace one exact string in a memory block."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_replace trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            changed = await memory.replace_core(user_id, label, old_string, new_string)
+        if not changed:
+            return f"No replacement made in '{label}' (old string not found)."
+        return f"Memory block '{label}' updated."
+
+    @tool
+    async def memory_delete(path_or_label: str) -> str:
+        """Delete a memory block by label or /memories/<label> path."""
+        label = _normalize_memory_label(path_or_label)
+        logger.info("deep_agent: memory_delete trace=%s user=%s label=%s", trace_id or "-", user_id, label)
+        if not label:
+            return "Invalid memory label."
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            deleted = await memory.delete_core(user_id, label)
+        if not deleted:
+            return f"Memory block '{label}' not found."
+        return f"Memory block '{label}' deleted."
+
+    @tool
+    async def archival_memory_insert(content: str) -> str:
+        """Insert a long-term archival memory entry."""
+        logger.info("deep_agent: archival_memory_insert trace=%s user=%s", trace_id or "-", user_id)
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            await memory.insert_archival(user_id, content, source="assistant")
+        return "Archival memory saved."
+
+    @tool
+    async def archival_memory_search(query: str, top_k: int = 5) -> str:
+        """Search long-term archival memory by semantic fallback (keyword currently)."""
+        logger.info("deep_agent: archival_memory_search trace=%s user=%s query=%s", trace_id or "-", user_id, query[:80])
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            results = await memory.search_archival(user_id, query, top_k=top_k)
+        if not results:
+            return "No archival memory results found."
+        lines = [f"- {item}" for item in results]
+        return "Archival memory results:\n" + "\n".join(lines)
+
+    @tool
+    async def conversation_search(query: str, top_k: int = 5) -> str:
+        """Search recall memory from prior episodic conversation summaries."""
+        logger.info("deep_agent: conversation_search trace=%s user=%s query=%s", trace_id or "-", user_id, query[:80])
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            results = await memory.search_recall(user_id, query, top_k=top_k)
+        if not results:
+            return "No recall memory results found."
+        lines = [f"- {item}" for item in results]
+        return "Recall memory results:\n" + "\n".join(lines)
+
+    return [
+        memory_list_blocks,
+        memory_get,
+        memory_create,
+        memory_append,
+        memory_replace,
+        memory_delete,
+        archival_memory_insert,
+        archival_memory_search,
+        conversation_search,
+    ]
+
+
+def _all_tools_for_user(user_id: str, trace_id: str | None) -> list[Any]:
+    return [*_all_tools(), *_memory_tools(user_id, trace_id)]
+
+
+def _detect_domain_section(message: str) -> FloatingDomainSection | None:
+    lowered = message.lower()
+    if any(keyword in lowered for keyword in ["timeline", "milestone", "release", "schedule"]):
+        return "timeline"
+    if any(keyword in lowered for keyword in ["task", "tasks", "todo", "attivit", "azione"]):
+        return "task"
+    if any(keyword in lowered for keyword in ["note", "notes", "memo", "document"]):
+        return "note"
+    return None
+
+
+def _normalize_domain_payload(payload: dict[str, Any], fallback_id: str | None) -> dict[str, str | None]:
+    type_raw = str(payload.get("type") or "").strip().lower()
+    domain_type: FloatingDomainType = "task"
+    if type_raw in {"task", "timeline", "project", "node"}:
+        domain_type = type_raw
+
+    id_value = payload.get("id")
+    domain_id = id_value if isinstance(id_value, str) and id_value.strip() else None
+    if domain_type == "project" and not domain_id:
+        domain_id = fallback_id
+
+    section_raw = payload.get("section")
+    section: FloatingDomainSection | None = None
+    if isinstance(section_raw, str):
+        section_candidate = section_raw.strip().lower()
+        if section_candidate in {"task", "timeline", "note"}:
+            section = section_candidate
+
+    if domain_type != "project":
+        section = None
+
+    return {
+        "type": domain_type,
+        "id": domain_id,
+        "section": section,
+    }
+
+
+def _parse_json_object(text: str) -> dict[str, Any] | None:
+    raw = text.strip()
+    if not raw:
+        return None
+    try:
+        parsed = json.loads(raw)
+        return parsed if isinstance(parsed, dict) else None
+    except json.JSONDecodeError:
+        pass
+
+    match = re.search(r"\{.*\}", raw, re.DOTALL)
+    if not match:
+        return None
+    try:
+        parsed = json.loads(match.group(0))
+    except json.JSONDecodeError:
+        return None
+    return parsed if isinstance(parsed, dict) else None
+
+
+def _infer_floating_domain_rule_based(message: str, context: dict[str, Any]) -> dict[str, str | None]:
+    section = _detect_domain_section(message)
+    scope = context.get("scope") if isinstance(context, dict) else None
+    resolved_project_id = context.get("resolved_project_id") if isinstance(context, dict) else None
+    project_id = resolved_project_id if isinstance(resolved_project_id, str) and resolved_project_id else None
+
+    if isinstance(scope, dict):
+        scope_type = str(scope.get("type") or "").strip().lower()
+        scope_id = scope.get("id")
+        scope_id_value = scope_id if isinstance(scope_id, str) and scope_id else None
+
+        if scope_type in {"task", "tasks"}:
+            return {"type": "task", "id": scope_id_value, "section": None}
+        if scope_type in {"project", "projects"}:
+            project_scope_id = scope_id_value or project_id
+            return {
+                "type": "project",
+                "id": project_scope_id,
+                "section": section,
+            }
+        if scope_type in {"note", "notes"}:
+            return {
+                "type": "node",
+                "id": scope_id_value,
+                "section": None,
+            }
+        if scope_type in {"timeline", "timelines"}:
+            return {"type": "timeline", "id": scope_id_value, "section": None}
+
+    lowered = message.lower()
+    if any(keyword in lowered for keyword in ["project", "progetto", "client"]) or project_id:
+        return {
+            "type": "project",
+            "id": project_id,
+            "section": section,
+        }
+    if section == "timeline":
+        return {"type": "timeline", "id": None, "section": None}
+    if section == "note":
+        return {"type": "node", "id": None, "section": None}
+    return {"type": "task", "id": None, "section": None}
+
+
+async def _infer_floating_domain(
+    message: str, context: dict[str, Any], *, langfuse_handler: Any | None = None,
+) -> dict[str, str | None]:
+    resolved_project_id = context.get("resolved_project_id") if isinstance(context, dict) else None
+    project_id = resolved_project_id if isinstance(resolved_project_id, str) and resolved_project_id else None
+
+    classifier_context = {
+        "scope": context.get("scope") if isinstance(context.get("scope"), dict) else None,
+        "resolved_project_id": project_id,
+    }
+
+    try:
+        classifier_prompt = _get_system_prompt(
+            "floating_domain_classifier", _FLOATING_DOMAIN_CLASSIFIER_SYSTEM,
+        )
+        callbacks = _build_callbacks(langfuse_handler)
+        llm = get_llm(callbacks=callbacks)
+        response = await llm.ainvoke(
+            [
+                SystemMessage(content=classifier_prompt),
+                HumanMessage(
+                    content=(
+                        f"Message:\n{message}\n\n"
+                        f"Context:\n{json.dumps(classifier_context, ensure_ascii=True)}"
+                    )
+                ),
+            ]
+        )
+        parsed = _parse_json_object(_as_text(response.content))
+        if parsed is not None:
+            domain = _normalize_domain_payload(parsed, project_id)
+            logger.info(
+                "deep_agent: floating_domain_classified type=%s id=%s section=%s",
+                domain.get("type"),
+                domain.get("id"),
+                domain.get("section"),
+            )
+            return domain
+        logger.warning("deep_agent: floating_domain classifier returned non-json output")
+    except Exception as exc:
+        logger.warning("deep_agent: floating_domain classifier failed: %s", exc)
+
+    return _infer_floating_domain_rule_based(message, context)
+
+
+def _get_system_prompt(langfuse_name: str, fallback: str) -> str:
+    """Fetch a managed prompt from Langfuse, falling back to the hardcoded string."""
+    managed = tracing.get_prompt(langfuse_name, fallback=None)
+    return managed if managed is not None else fallback
+
+
+def _build_callbacks(langfuse_handler: Any | None) -> list[Any] | None:
+    """Return a callbacks list if a Langfuse handler is available."""
+    if langfuse_handler is None:
+        return None
+    return [langfuse_handler]
+
+
+async def _run_single_agent(
+    *,
+    user_id: str,
+    system_prompt: str,
+    message: str,
+    context: dict[str, Any],
+    max_steps: int = 6,
+    langfuse_handler: Any | None = None,
+) -> str:
+    trace_id = _trace_id_from_context(context)
+    callbacks = _build_callbacks(langfuse_handler)
+    llm = get_llm(callbacks=callbacks)
+    tools = _all_tools_for_user(user_id, trace_id)
+    model_context = _context_for_model(context)
+    logger.info("deep_agent: run_single_agent_start trace=%s user=%s", trace_id or "-", user_id)
+    llm_with_tools = llm.bind_tools(tools)
+    messages: list[Any] = [
+        SystemMessage(content=system_prompt),
+        HumanMessage(
+            content=(
+                f"User message:\n{message}\n\n"
+                f"Context:\n{json.dumps({'context': model_context}, ensure_ascii=True)[:3500]}"
+            )
+        ),
+    ]
+
+    tool_calls_count = 0
+    collected: list[dict[str, Any]] = []
+    set_tool_result_collector(collected)
+    try:
+        for _ in range(max_steps):
+            response: AIMessage = await llm_with_tools.ainvoke(messages)
+            messages.append(response)
+
+            if not response.tool_calls:
+                final_text = _as_text(response.content)
+                logger.info(
+                    "deep_agent: run_single_agent_end trace=%s user=%s tool_calls=%d response_chars=%d",
+                    trace_id or "-",
+                    user_id,
+                    tool_calls_count,
+                    len(final_text),
+                )
+                return final_text
+
+            tool_map = {tool_def.name: tool_def for tool_def in tools}
+            for call in response.tool_calls:
+                tool_calls_count += 1
+                call_id = str(call.get("id", ""))
+                call_name = str(call.get("name", ""))
+                call_args = call.get("args", {})
+                logger.info(
+                    "deep_agent: AI->Tool tool_call_id=%s tool=%s args=%s",
+                    call_id,
+                    call_name,
+                    json.dumps(call_args, ensure_ascii=True)[:800],
+                )
+
+                tool_fn = tool_map.get(call_name)
+                if tool_fn is None:
+                    tool_output = f"Unknown tool: {call_name}"
+                else:
+                    tool_output = await tool_fn.ainvoke(call_args)
+
+                logger.info(
+                    "deep_agent: Tool->AI tool_call_id=%s tool=%s output=%s",
+                    call_id,
+                    call_name,
+                    str(tool_output)[:1200],
+                )
+
+                messages.append(ToolMessage(content=str(tool_output), tool_call_id=call["id"]))
+
+        final = await llm.ainvoke(messages)
+        final_text = _as_text(final.content)
+        logger.info(
+            "deep_agent: run_single_agent_end trace=%s user=%s tool_calls=%d response_chars=%d fallback=1",
+            trace_id or "-",
+            user_id,
+            tool_calls_count,
+            len(final_text),
+        )
+        return final_text
+    finally:
+        clear_tool_result_collector()
+
+
+async def _run_single_agent_stream(
+    *,
+    user_id: str,
+    system_prompt: str,
+    message: str,
+    context: dict[str, Any],
+    max_steps: int = 6,
+    langfuse_handler: Any | None = None,
+) -> AsyncGenerator[tuple[str, Any], None]:
+    trace_id = _trace_id_from_context(context)
+    callbacks = _build_callbacks(langfuse_handler)
+    llm = get_llm(callbacks=callbacks)
+    tools = _all_tools_for_user(user_id, trace_id)
+    model_context = _context_for_model(context)
+    logger.info("deep_agent: run_single_agent_stream_start trace=%s user=%s", trace_id or "-", user_id)
+    llm_with_tools = llm.bind_tools(tools)
+    messages: list[Any] = [
+        SystemMessage(content=system_prompt),
+        HumanMessage(
+            content=(
+                f"User message:\n{message}\n\n"
+                f"Context:\n{json.dumps({'context': model_context}, ensure_ascii=True)[:3500]}"
+            )
+        ),
+    ]
+
+    tool_calls_count = 0
+    streamed_chars = 0
+    collected: list[dict[str, Any]] = []
+    set_tool_result_collector(collected)
+    try:
+        for _ in range(max_steps):
+            response: AIMessage = await llm_with_tools.ainvoke(messages)
+            messages.append(response)
+
+            if not response.tool_calls:
+                emitted_any = False
+                async for chunk in llm.astream(messages):
+                    token = _as_text(getattr(chunk, "content", ""))
+                    if token:
+                        streamed_chars += len(token)
+                        emitted_any = True
+                        yield "token", token
+
+                if not emitted_any:
+                    fallback_text = _as_text(response.content)
+                    if fallback_text:
+                        streamed_chars += len(fallback_text)
+                        yield "token", fallback_text
+                logger.info(
+                    "deep_agent: run_single_agent_stream_end trace=%s user=%s tool_calls=%d response_chars=%d",
+                    trace_id or "-",
+                    user_id,
+                    tool_calls_count,
+                    streamed_chars,
+                )
+                return
+
+            tool_map = {tool_def.name: tool_def for tool_def in tools}
+            for call in response.tool_calls:
+                tool_calls_count += 1
+                call_id = str(call.get("id", ""))
+                call_name = str(call.get("name", ""))
+                call_args = call.get("args", {})
+                logger.info(
+                    "deep_agent: AI->Tool tool_call_id=%s tool=%s args=%s",
+                    call_id,
+                    call_name,
+                    json.dumps(call_args, ensure_ascii=True)[:800],
+                )
+
+                tool_fn = tool_map.get(call_name)
+                if tool_fn is None:
+                    tool_output = f"Unknown tool: {call_name}"
+                else:
+                    tool_output = await tool_fn.ainvoke(call_args)
+
+                logger.info(
+                    "deep_agent: Tool->AI tool_call_id=%s tool=%s output=%s",
+                    call_id,
+                    call_name,
+                    str(tool_output)[:1200],
+                )
+
+                messages.append(ToolMessage(content=str(tool_output), tool_call_id=call["id"]))
+
+        async for chunk in llm.astream(messages):
+            token = _as_text(getattr(chunk, "content", ""))
+            if token:
+                streamed_chars += len(token)
+                yield "token", token
+        logger.info(
+            "deep_agent: run_single_agent_stream_end trace=%s user=%s tool_calls=%d response_chars=%d fallback=1",
+            trace_id or "-",
+            user_id,
+            tool_calls_count,
+            streamed_chars,
+        )
+    finally:
+        clear_tool_result_collector()
+
+
+async def run_home(user_id: str, message: str, context: dict[str, Any], *, langfuse_handler: Any | None = None) -> str:
+    prepared_context = await _prepare_context(message, context)
+    system_prompt = _get_system_prompt("home_system", _HOME_SINGLE_AGENT_SYSTEM)
+    response = await _run_single_agent(
+        user_id=user_id,
+        system_prompt=system_prompt,
+        message=message,
+        context=prepared_context,
+        langfuse_handler=langfuse_handler,
+    )
+    return _normalize_tagged_list_lines(response, message)
+
+
+async def run_floating(user_id: str, message: str, context: dict[str, Any], *, langfuse_handler: Any | None = None) -> tuple[str, dict[str, str | None]]:
+    prepared_context = await _prepare_context(message, context)
+    domain = await _infer_floating_domain(message, prepared_context, langfuse_handler=langfuse_handler)
+    system_prompt = _get_system_prompt("floating_system", _FLOATING_SINGLE_AGENT_SYSTEM)
+    response = await _run_single_agent(
+        user_id=user_id,
+        system_prompt=system_prompt,
+        message=message,
+        context=prepared_context,
+        langfuse_handler=langfuse_handler,
+    )
+    sanitized = _strip_floating_markup(response)
+    if not sanitized and response:
+        sanitized = _fallback_from_raw_floating_text(response)
+    return sanitized, domain
+
+
+async def run_home_stream(
+    user_id: str,
+    message: str,
+    context: dict[str, Any],
+    *,
+    langfuse_handler: Any | None = None,
+) -> AsyncGenerator[tuple[str, Any], None]:
+    prepared_context = await _prepare_context(message, context)
+    system_prompt = _get_system_prompt("home_system", _HOME_SINGLE_AGENT_SYSTEM)
+    text_chunks: list[str] = []
+    async for event in _run_single_agent_stream(
+        user_id=user_id,
+        system_prompt=system_prompt,
+        message=message,
+        context=prepared_context,
+        langfuse_handler=langfuse_handler,
+    ):
+        event_type, data = event
+        if event_type != "token":
+            yield event
+            continue
+        text_chunks.append(str(data or ""))
+
+    normalized = _normalize_tagged_list_lines("".join(text_chunks), message)
+    if normalized:
+        yield "token", normalized
+
+
+async def run_floating_stream(
+    user_id: str,
+    message: str,
+    context: dict[str, Any],
+    *,
+    langfuse_handler: Any | None = None,
+) -> AsyncGenerator[tuple[str, Any], None]:
+    prepared_context = await _prepare_context(message, context)
+    domain = await _infer_floating_domain(message, prepared_context, langfuse_handler=langfuse_handler)
+    yield "floating_domain", domain
+
+    system_prompt = _get_system_prompt("floating_system", _FLOATING_SINGLE_AGENT_SYSTEM)
+    sanitizer = _FloatingStreamSanitizer()
+    emitted_sanitized = False
+    raw_chunks: list[str] = []
+    async for event in _run_single_agent_stream(
+        user_id=user_id,
+        system_prompt=system_prompt,
+        message=message,
+        context=prepared_context,
+        langfuse_handler=langfuse_handler,
+    ):
+        event_type, data = event
+        if event_type != "token":
+            yield event
+            continue
+
+        raw_chunk = str(data or "")
+        raw_chunks.append(raw_chunk)
+        sanitized_chunk = sanitizer.feed(raw_chunk)
+        if sanitized_chunk:
+            emitted_sanitized = True
+            yield "token", sanitized_chunk
+
+    tail = sanitizer.finalize()
+    if tail:
+        emitted_sanitized = True
+        yield "token", tail
+
+    if not emitted_sanitized and raw_chunks:
+        yield "token", _fallback_from_raw_floating_text("".join(raw_chunks))
+
+
+async def update_core_memory(user_id: str, key: str, value: str) -> None:
+    """Compatibility helper kept for callers that expect explicit memory update API."""
+    async with async_session() as db:
+        memory = MemoryMiddleware(db)
+        await memory.update_core(user_id, key, value)

services/chat/app/llm.py

@@ -0,0 +1,72 @@
+"""LLM factory — centralised model instantiation via LiteLLM.
+
+Adapted from app/core/llm.py for the Chat Service.
+Uses shared.config.settings instead of app.config.settings.
+"""
+
+from __future__ import annotations
+
+import os
+import warnings
+
+from openai import AsyncOpenAI
+import litellm
+
+from langchain_openai import ChatOpenAI
+from langchain_litellm import ChatLiteLLM
+
+from shared.config import settings
+
+litellm.drop_params = True
+
+warnings.filterwarnings(
+    "ignore",
+    message=r"PydanticSerializationUnexpectedValue\(Expected `ResponseAPIUsage`",
+    category=UserWarning,
+)
+
+
+def _api_key_for_model(model: str) -> str | None:
+    if model.startswith("anthropic/"):
+        return settings.ANTHROPIC_API_KEY or None
+    if model.startswith("gemini/") or model.startswith("google/"):
+        return settings.GOOGLE_API_KEY or None
+    if model.startswith("cerebras/"):
+        return settings.CEREBRAS_API_KEY or None
+    if model.startswith("github_copilot/"):
+        return None
+    return settings.OPENAI_API_KEY or None
+
+
+def get_llm(
+    *,
+    model: str | None = None,
+    temperature: float = 0,
+    callbacks: list | None = None,
+) -> ChatOpenAI | ChatLiteLLM:
+    model = model or settings.LLM_MODEL
+
+    if settings.GITHUB_COPILOT_TOKEN_DIR:
+        os.environ.setdefault("GITHUB_COPILOT_TOKEN_DIR", settings.GITHUB_COPILOT_TOKEN_DIR)
+
+    if "/" in model:
+        return ChatLiteLLM(model=model, temperature=temperature, callbacks=callbacks)
+
+    return ChatOpenAI(
+        model=model,
+        temperature=temperature,
+        api_key=_api_key_for_model(model),
+        callbacks=callbacks,
+    )
+
+
+async def embed(text: str) -> list[float]:
+    model = settings.LLM_EMBED_MODEL
+
+    if model.startswith("github_copilot/") or "/" in model:
+        response = await litellm.aembedding(model=model, input=[text])
+        return response.data[0]["embedding"]
+
+    client = AsyncOpenAI(api_key=settings.OPENAI_API_KEY)
+    response = await client.embeddings.create(model=model, input=text)
+    return response.data[0].embedding

services/chat/app/main.py

@@ -0,0 +1,87 @@
+"""Chat Service — LLM orchestration, domain agents, memory.
+
+Consumes chat requests from Redis, executes deep_agent (home/floating),
+streams responses back via Redis pub/sub to WS Gateway.
+
+Owns: memory_core, memory_associative, memory_episodic, memory_proactive tables.
+"""
+
+import sys
+from contextlib import asynccontextmanager
+import logging
+from pathlib import Path
+
+# Ensure the repo root is on sys.path so "shared" is importable in local dev.
+_repo_root = str(Path(__file__).resolve().parents[3])
+if _repo_root not in sys.path:
+    sys.path.insert(0, _repo_root)
+
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+
+from shared.config import settings
+
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s %(levelname)s %(name)s: %(message)s",
+)
+logging.getLogger("sqlalchemy.engine").setLevel(logging.WARNING)
+logging.getLogger("sqlalchemy.pool").setLevel(logging.WARNING)
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    # Initialise Langfuse tracing (no-op if keys are missing)
+    from app.tracing import init_langfuse
+
+    init_langfuse()
+
+    # Start Redis consumer in background
+    from app.redis_consumer import start_consumer
+
+    consumer_task = start_consumer()
+    yield
+    consumer_task.cancel()
+
+    from app.tracing import shutdown as shutdown_langfuse
+
+    shutdown_langfuse()
+
+    from shared.db import engine
+
+    await engine.dispose()
+
+    from shared.redis import redis_client
+
+    await redis_client.aclose()
+
+
+def create_app() -> FastAPI:
+    app = FastAPI(
+        title="Adiuva Chat Service",
+        version="0.1.0",
+        docs_url="/docs" if settings.ENV == "dev" else None,
+        redoc_url=None,
+        lifespan=lifespan,
+    )
+
+    app.add_middleware(
+        CORSMiddleware,
+        allow_origins=settings.CORS_ORIGINS,
+        allow_credentials=True,
+        allow_methods=["*"],
+        allow_headers=["*"],
+    )
+
+    from app.routes import router
+
+    app.include_router(router, prefix="/api/v1")
+
+    @app.get("/api/v1/health", tags=["health"])
+    async def health() -> dict:
+        return {"status": "ok", "service": "chat", "version": app.version}
+
+    return app
+
+
+app = create_app()

services/chat/app/memory_middleware.py

@@ -0,0 +1,295 @@
+"""Memory Middleware — adapted for Chat Service.
+
+Uses shared.models instead of app.models. Otherwise identical to the
+monolith's app/core/memory_middleware.py.
+"""
+
+from __future__ import annotations
+
+import logging
+import uuid
+from typing import Any
+
+from cryptography.fernet import Fernet, InvalidToken
+from sqlalchemy import select
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from shared.models import (
+    MemoryAssociative,
+    MemoryCore,
+    MemoryEpisodic,
+    MemoryProactive,
+    User,
+)
+
+logger = logging.getLogger(__name__)
+
+_ASSOCIATIVE_TOP_K = 5
+_EPISODIC_RECENT_N = 10
+_PROACTIVE_CONFIDENCE_THRESHOLD = 0.6
+
+
+class MemoryMiddleware:
+
+    def __init__(self, db: AsyncSession) -> None:
+        self._db = db
+
+    async def enrich_context(
+        self,
+        user_id: str,
+        message: str,
+        trace_id: str | None = None,
+        session_id: str | None = None,
+    ) -> dict[str, Any]:
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return {}
+
+        core = await self._load_core(user_id, fernet)
+        associative = await self._load_associative(user_id, message, fernet)
+        episodic = await self._load_episodic(user_id, fernet, session_id=session_id)
+        proactive = await self._load_proactive(user_id, fernet)
+
+        logger.info(
+            "memory: enrich_context trace=%s user=%s core=%d assoc=%d episodic=%d proactive=%d",
+            trace_id or "-", user_id, len(core), len(associative), len(episodic), len(proactive),
+        )
+
+        return {
+            "core_memory": core,
+            "associative_memory": associative,
+            "episodic_memory": episodic,
+            "proactive_hints": proactive,
+        }
+
+    async def store_episode(
+        self, user_id: str, session_id: str, message: str, response: str,
+        trace_id: str | None = None,
+    ) -> None:
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return
+
+        summary = f"User: {message[:200]}\nAssistant: {response[:200]}"
+        encrypted = _encrypt(fernet, summary)
+
+        row = MemoryEpisodic(
+            id=str(uuid.uuid4()),
+            user_id=user_id,
+            summary_encrypted=encrypted,
+            session_id=session_id,
+        )
+        self._db.add(row)
+        try:
+            await self._db.commit()
+        except Exception as exc:
+            logger.error("memory: store_episode failed user=%s: %s", user_id, exc)
+            await self._db.rollback()
+
+    async def update_core(self, user_id: str, key: str, value: str, trace_id: str | None = None) -> None:
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return
+
+        encrypted = _encrypt(fernet, value)
+        result = await self._db.execute(
+            select(MemoryCore).where(MemoryCore.user_id == user_id, MemoryCore.key == key)
+        )
+        existing = result.scalar_one_or_none()
+        if existing is not None:
+            existing.value_encrypted = encrypted
+        else:
+            self._db.add(MemoryCore(
+                id=str(uuid.uuid4()), user_id=user_id, key=key, value_encrypted=encrypted,
+            ))
+        try:
+            await self._db.commit()
+        except Exception as exc:
+            logger.error("memory: update_core failed user=%s key=%s: %s", user_id, key, exc)
+            await self._db.rollback()
+
+    async def list_core_blocks(self, user_id: str) -> list[dict[str, str]]:
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return []
+        result = await self._db.execute(
+            select(MemoryCore).where(MemoryCore.user_id == user_id).order_by(MemoryCore.key.asc())
+        )
+        out: list[dict[str, str]] = []
+        for row in result.scalars().all():
+            plaintext = _safe_decrypt(fernet, row.value_encrypted)
+            if plaintext is not None:
+                out.append({"label": row.key, "value": plaintext})
+        return out
+
+    async def get_core_block(self, user_id: str, label: str) -> str | None:
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return None
+        result = await self._db.execute(
+            select(MemoryCore).where(MemoryCore.user_id == user_id, MemoryCore.key == label)
+        )
+        row = result.scalar_one_or_none()
+        if row is None:
+            return None
+        return _safe_decrypt(fernet, row.value_encrypted)
+
+    async def delete_core(self, user_id: str, label: str) -> bool:
+        result = await self._db.execute(
+            select(MemoryCore).where(MemoryCore.user_id == user_id, MemoryCore.key == label)
+        )
+        row = result.scalar_one_or_none()
+        if row is None:
+            return False
+        await self._db.delete(row)
+        try:
+            await self._db.commit()
+            return True
+        except Exception as exc:
+            logger.error("memory: delete_core failed user=%s label=%s: %s", user_id, label, exc)
+            await self._db.rollback()
+            return False
+
+    async def append_core(self, user_id: str, label: str, content: str) -> None:
+        current = await self.get_core_block(user_id, label)
+        if current is None:
+            await self.update_core(user_id, label, content)
+            return
+        await self.update_core(user_id, label, f"{current}\n{content}")
+
+    async def replace_core(self, user_id: str, label: str, old: str, new: str) -> bool:
+        current = await self.get_core_block(user_id, label)
+        if current is None or old not in current:
+            return False
+        await self.update_core(user_id, label, current.replace(old, new, 1))
+        return True
+
+    async def insert_archival(self, user_id: str, content: str, source: str = "manual") -> None:
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return
+        encrypted = _encrypt(fernet, content)
+        row = MemoryAssociative(
+            id=str(uuid.uuid4()), user_id=user_id,
+            content_encrypted=encrypted, embedding=None,
+            entity_type=source, entity_id=None,
+        )
+        self._db.add(row)
+        try:
+            await self._db.commit()
+        except Exception as exc:
+            logger.error("memory: insert_archival failed user=%s: %s", user_id, exc)
+            await self._db.rollback()
+
+    async def search_archival(self, user_id: str, query: str, top_k: int = 5) -> list[str]:
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return []
+        result = await self._db.execute(
+            select(MemoryAssociative).where(MemoryAssociative.user_id == user_id)
+            .order_by(MemoryAssociative.updated_at.desc()).limit(100)
+        )
+        needle = query.strip().lower()
+        out: list[str] = []
+        for row in result.scalars().all():
+            plaintext = _safe_decrypt(fernet, row.content_encrypted)
+            if plaintext is None:
+                continue
+            if not needle or needle in plaintext.lower():
+                out.append(plaintext)
+            if len(out) >= max(top_k, 1):
+                break
+        return out
+
+    async def search_recall(self, user_id: str, query: str, top_k: int = 5) -> list[str]:
+        fernet = await self._get_fernet(user_id)
+        if fernet is None:
+            return []
+        result = await self._db.execute(
+            select(MemoryEpisodic).where(MemoryEpisodic.user_id == user_id)
+            .order_by(MemoryEpisodic.created_at.desc()).limit(100)
+        )
+        needle = query.strip().lower()
+        out: list[str] = []
+        for row in result.scalars().all():
+            plaintext = _safe_decrypt(fernet, row.summary_encrypted)
+            if plaintext is None:
+                continue
+            if not needle or needle in plaintext.lower():
+                out.append(plaintext)
+            if len(out) >= max(top_k, 1):
+                break
+        return out
+
+    # ── Private ───────────────────────────────────────────────────────
+
+    async def _get_fernet(self, user_id: str) -> Fernet | None:
+        result = await self._db.execute(select(User).where(User.id == user_id))
+        user = result.scalar_one_or_none()
+        if user is None or not user.encryption_key:
+            logger.warning("memory: no encryption_key for user=%s", user_id)
+            return None
+        return Fernet(user.encryption_key.encode())
+
+    async def _load_core(self, user_id: str, fernet: Fernet) -> dict[str, str]:
+        result = await self._db.execute(
+            select(MemoryCore).where(MemoryCore.user_id == user_id)
+        )
+        out: dict[str, str] = {}
+        for row in result.scalars().all():
+            plaintext = _safe_decrypt(fernet, row.value_encrypted)
+            if plaintext is not None:
+                out[row.key] = plaintext
+        return out
+
+    async def _load_associative(self, user_id: str, message: str, fernet: Fernet) -> list[str]:
+        result = await self._db.execute(
+            select(MemoryAssociative).where(MemoryAssociative.user_id == user_id)
+            .order_by(MemoryAssociative.updated_at.desc()).limit(_ASSOCIATIVE_TOP_K)
+        )
+        out: list[str] = []
+        for row in result.scalars().all():
+            plaintext = _safe_decrypt(fernet, row.content_encrypted)
+            if plaintext is not None:
+                out.append(plaintext)
+        return out
+
+    async def _load_episodic(self, user_id: str, fernet: Fernet, session_id: str | None = None) -> list[str]:
+        query = select(MemoryEpisodic).where(MemoryEpisodic.user_id == user_id)
+        if session_id:
+            query = query.where(MemoryEpisodic.session_id == session_id)
+        result = await self._db.execute(
+            query.order_by(MemoryEpisodic.created_at.desc()).limit(_EPISODIC_RECENT_N)
+        )
+        out: list[str] = []
+        for row in result.scalars().all():
+            plaintext = _safe_decrypt(fernet, row.summary_encrypted)
+            if plaintext is not None:
+                out.append(plaintext)
+        return out
+
+    async def _load_proactive(self, user_id: str, fernet: Fernet) -> list[str]:
+        result = await self._db.execute(
+            select(MemoryProactive).where(
+                MemoryProactive.user_id == user_id,
+                MemoryProactive.confidence >= _PROACTIVE_CONFIDENCE_THRESHOLD,
+            ).order_by(MemoryProactive.confidence.desc())
+        )
+        out: list[str] = []
+        for row in result.scalars().all():
+            plaintext = _safe_decrypt(fernet, row.pattern_encrypted)
+            if plaintext is not None:
+                out.append(plaintext)
+        return out
+
+
+def _encrypt(fernet: Fernet, plaintext: str) -> str:
+    return fernet.encrypt(plaintext.encode()).decode()
+
+
+def _safe_decrypt(fernet: Fernet, ciphertext: str) -> str | None:
+    try:
+        return fernet.decrypt(ciphertext.encode()).decode()
+    except (InvalidToken, Exception) as exc:
+        logger.warning("memory: decrypt failed: %s", exc)
+        return None

services/chat/app/output_formatter.py

@@ -0,0 +1,50 @@
+"""Output formatter for deep-agent stream events — Chat Service copy.
+
+Converts (event_type, data) tuples into WebSocket frame Pydantic models.
+"""
+
+from __future__ import annotations
+
+from collections.abc import AsyncGenerator
+from typing import Any
+
+from shared.schemas import WsFloatingDomain, WsStreamEnd, WsStreamStart, WsStreamText
+
+WsFrame = WsStreamStart | WsStreamText | WsStreamEnd | WsFloatingDomain
+
+
+class StreamFormatter:
+    """Convert `(event_type, data)` stream events into websocket frame models."""
+
+    def __init__(self, request_id: str) -> None:
+        self.request_id = request_id
+
+    async def format(
+        self,
+        event_stream: AsyncGenerator[tuple[str, Any], None],
+    ) -> AsyncGenerator[WsFrame, None]:
+        started = False
+
+        async for event_type, data in event_stream:
+            if event_type == "floating_domain":
+                if isinstance(data, dict):
+                    yield WsFloatingDomain(
+                        request_id=self.request_id,
+                        domain=data,
+                    )
+                continue
+
+            if event_type != "token":
+                continue
+
+            if not started:
+                yield WsStreamStart(request_id=self.request_id)
+                started = True
+
+            text = str(data or "")
+            if text:
+                yield WsStreamText(request_id=self.request_id, chunk=text)
+
+        if not started:
+            yield WsStreamStart(request_id=self.request_id)
+        yield WsStreamEnd(request_id=self.request_id)

services/chat/app/redis_consumer.py

@@ -0,0 +1,209 @@
+"""Redis consumer — listens for chat requests and dispatches to deep_agent.
+
+Subscribes to a Redis pattern channel chat:request:* so it receives
+requests for ALL users. Each request is processed in a separate asyncio task.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import logging
+from uuid import uuid4
+
+from shared.db import async_session
+from shared.redis import redis_client, ws_out_channel
+
+from app.deep_agent import run_floating_stream, run_home_stream
+from app.memory_middleware import MemoryMiddleware
+from app.output_formatter import StreamFormatter
+from app.ws_context import clear_current_user, set_current_user
+from app import tracing
+
+logger = logging.getLogger(__name__)
+
+
+def start_consumer() -> asyncio.Task:
+    """Start the Redis consumer as a background asyncio task."""
+    return asyncio.create_task(_consumer_loop())
+
+
+async def _consumer_loop() -> None:
+    """Subscribe to chat:request:* and dispatch incoming frames."""
+    pubsub = redis_client.pubsub()
+    await pubsub.psubscribe("chat:request:*")
+    logger.info("redis_consumer: subscribed to chat:request:*")
+
+    try:
+        while True:
+            message = await pubsub.get_message(
+                ignore_subscribe_messages=True, timeout=1.0
+            )
+            if message is not None and message["type"] == "pmessage":
+                frame = json.loads(message["data"])
+                asyncio.create_task(_dispatch(frame))
+            else:
+                await asyncio.sleep(0.01)
+    except asyncio.CancelledError:
+        logger.info("redis_consumer: shutting down")
+    finally:
+        await pubsub.punsubscribe()
+        await pubsub.aclose()
+
+
+async def _dispatch(frame: dict) -> None:
+    """Route a chat request frame to the appropriate handler."""
+    frame_type = frame.get("type")
+    user_id = frame.get("user_id")
+
+    if not user_id:
+        logger.warning("redis_consumer: frame missing user_id: %s", frame.get("type"))
+        return
+
+    if frame_type == "home_request":
+        await _handle_home_request(user_id, frame)
+    elif frame_type == "floating_request":
+        await _handle_floating_request(user_id, frame)
+    else:
+        logger.debug("redis_consumer: unknown frame type %r", frame_type)
+
+
+async def _publish_frame(user_id: str, frame_data: str) -> None:
+    """Publish a frame to ws:out:{user_id} for the WS Gateway to forward."""
+    channel = ws_out_channel(user_id)
+    await redis_client.publish(channel, frame_data)
+
+
+async def _handle_home_request(user_id: str, frame: dict) -> None:
+    """Process a home_request — enrich with memory, run deep_agent, stream results."""
+    request_id = frame.get("request_id") or str(uuid4())
+    message: str = frame.get("message", "")
+    session_id: str = frame.get("session_id") or str(uuid4())
+
+    logger.info(
+        "redis_consumer: home_request user=%s req=%s msg=%s",
+        user_id, request_id, message[:200],
+    )
+
+    response_chunks: list[str] = []
+
+    with tracing.trace_span(
+        name="home_request",
+        user_id=user_id,
+        session_id=session_id,
+        trace_id=request_id,
+        input=message,
+        metadata={"message_preview": message[:200]},
+        tags=["home"],
+    ) as span:
+        langfuse_handler = tracing.get_langfuse_callback()
+
+        # Enrich with memory context
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            memory_context = await memory.enrich_context(
+                user_id, message,
+                trace_id=request_id, session_id=session_id,
+            )
+
+        context: dict = {
+            "conversation_history": frame.get("conversation_history", []),
+            "_debug": {"request_id": request_id, "session_id": session_id, "user_id": user_id},
+            **memory_context,
+        }
+
+        set_current_user(user_id)
+        try:
+            event_stream = run_home_stream(user_id, message, context, langfuse_handler=langfuse_handler)
+            formatter = StreamFormatter(request_id=request_id)
+            async for ws_frame in formatter.format(event_stream):
+                await _publish_frame(user_id, ws_frame.model_dump_json())
+                if hasattr(ws_frame, "chunk"):
+                    response_chunks.append(ws_frame.chunk)
+        except Exception as exc:
+            logger.error("redis_consumer: home_request failed user=%s req=%s: %s", user_id, request_id, exc)
+        finally:
+            clear_current_user()
+
+        # Link prompt and attach output preview
+        tracing.link_prompt_to_trace(span, "home_system")
+        response_text = "".join(response_chunks)
+        span.update(output=response_text[:500] if response_text else None)
+
+    tracing.flush()
+
+    # Store episode
+    async with async_session() as db:
+        memory = MemoryMiddleware(db)
+        await memory.store_episode(
+            user_id, session_id, message, "".join(response_chunks),
+            trace_id=request_id,
+        )
+
+
+async def _handle_floating_request(user_id: str, frame: dict) -> None:
+    """Process a floating_request — enrich with memory, run deep_agent, stream results."""
+    request_id = frame.get("request_id") or str(uuid4())
+    message: str = frame.get("message", "")
+    session_id: str = frame.get("session_id") or str(uuid4())
+    scope: dict = frame.get("scope", {})
+
+    logger.info(
+        "redis_consumer: floating_request user=%s req=%s scope=%s msg=%s",
+        user_id, request_id, json.dumps(scope)[:200], message[:200],
+    )
+
+    response_chunks: list[str] = []
+
+    with tracing.trace_span(
+        name="floating_request",
+        user_id=user_id,
+        session_id=session_id,
+        trace_id=request_id,
+        input=message,
+        metadata={"message_preview": message[:200], "scope": scope},
+        tags=["floating"],
+    ) as span:
+        langfuse_handler = tracing.get_langfuse_callback()
+
+        # Enrich with memory context
+        async with async_session() as db:
+            memory = MemoryMiddleware(db)
+            memory_context = await memory.enrich_context(
+                user_id, message,
+                trace_id=request_id, session_id=session_id,
+            )
+
+        context: dict = {
+            "scope": scope,
+            "_debug": {"request_id": request_id, "session_id": session_id, "user_id": user_id},
+            **memory_context,
+        }
+
+        set_current_user(user_id)
+        try:
+            event_stream = run_floating_stream(user_id, message, context, langfuse_handler=langfuse_handler)
+            formatter = StreamFormatter(request_id=request_id)
+            async for ws_frame in formatter.format(event_stream):
+                await _publish_frame(user_id, ws_frame.model_dump_json())
+                if hasattr(ws_frame, "chunk"):
+                    response_chunks.append(ws_frame.chunk)
+        except Exception as exc:
+            logger.error("redis_consumer: floating_request failed user=%s req=%s: %s", user_id, request_id, exc)
+        finally:
+            clear_current_user()
+
+        # Link prompt and attach output preview
+        tracing.link_prompt_to_trace(span, "floating_system")
+        response_text = "".join(response_chunks)
+        span.update(output=response_text[:500] if response_text else None)
+
+    tracing.flush()
+
+    # Store episode
+    async with async_session() as db:
+        memory = MemoryMiddleware(db)
+        await memory.store_episode(
+            user_id, session_id, message, "".join(response_chunks),
+            trace_id=request_id,
+        )

services/chat/app/routes.py

@@ -0,0 +1,37 @@
+"""Chat REST route — POST /chat fallback when WS is unavailable."""
+
+from __future__ import annotations
+
+from fastapi import APIRouter, Request
+from fastapi.responses import JSONResponse
+
+from shared.schemas import ChatRequest
+
+from app.deep_agent import run_home
+from app.ws_context import clear_current_user, set_current_user
+
+router = APIRouter(prefix="/chat", tags=["chat"])
+
+
+@router.post("")
+async def chat(body: ChatRequest, request: Request) -> JSONResponse:
+    """REST fallback for home chat.
+
+    In the microservices setup, Traefik ForwardAuth has already validated
+    the JWT and injected X-User-Id / X-User-Email / X-User-Tier headers.
+    """
+    user_id = request.headers.get("X-User-Id", "")
+    if not user_id:
+        return JSONResponse(status_code=401, content={"detail": "Missing X-User-Id header"})
+
+    set_current_user(user_id)
+    try:
+        response = await run_home(
+            user_id=user_id,
+            message=body.message,
+            context=body.context.model_dump(),
+        )
+    finally:
+        clear_current_user()
+
+    return JSONResponse(content={"response": response})

services/chat/app/tracing.py

@@ -0,0 +1,264 @@
+"""Langfuse tracing & prompt management for the Chat Service (v4 SDK).
+
+Provides:
+- ``init_langfuse()`` — initialise the singleton client at startup
+- ``trace_span()`` — context manager that creates a trace + span
+- ``get_langfuse_callback()`` — LangChain callback handler (auto-inherits trace)
+- ``get_prompt()`` — fetch a managed prompt from Langfuse by name
+- ``flush()`` / ``shutdown()`` — lifecycle management
+
+All functions gracefully degrade to no-ops when Langfuse is not configured,
+so the service works identically with or without observability keys.
+
+Requires ``langfuse >= 3.0.0`` (v4 / "Fast Preview" SDK).
+"""
+
+from __future__ import annotations
+
+import logging
+from contextlib import contextmanager
+from typing import Any
+
+from shared.config import settings
+
+logger = logging.getLogger(__name__)
+
+# ── State ────────────────────────────────────────────────────────────────
+
+_initialised: bool = False
+_disabled: bool = False
+
+
+def _is_configured() -> bool:
+    return bool(settings.LANGFUSE_SECRET_KEY and settings.LANGFUSE_PUBLIC_KEY)
+
+
+def init_langfuse() -> None:
+    """Initialise the Langfuse singleton. Call once at startup."""
+    global _initialised, _disabled
+
+    if _initialised or _disabled:
+        return
+
+    if not _is_configured():
+        _disabled = True
+        logger.info("tracing: Langfuse keys not set — tracing disabled")
+        return
+
+    try:
+        from langfuse import Langfuse
+
+        Langfuse(
+            secret_key=settings.LANGFUSE_SECRET_KEY,
+            public_key=settings.LANGFUSE_PUBLIC_KEY,
+            host=settings.LANGFUSE_HOST,
+        )
+        _initialised = True
+        logger.info("tracing: Langfuse client initialised (host=%s)", settings.LANGFUSE_HOST)
+    except Exception as exc:
+        _disabled = True
+        logger.warning("tracing: failed to initialise Langfuse: %s", exc)
+
+
+def _get_client() -> Any | None:
+    """Return the singleton Langfuse client, or *None* if disabled."""
+    if _disabled:
+        return None
+    if not _initialised:
+        init_langfuse()
+    if _disabled:
+        return None
+    try:
+        from langfuse import get_client
+        return get_client()
+    except Exception:
+        return None
+
+
+# ── Null span (no-op when Langfuse is disabled) ─────────────────────────
+
+
+class _NullSpan:
+    """Drop-in replacement when Langfuse is disabled."""
+
+    def update(self, **_: Any) -> None: ...
+    def set_trace_io(self, **_: Any) -> None: ...
+    def score_trace(self, **_: Any) -> None: ...
+
+
+# ── Trace context manager ───────────────────────────────────────────────
+
+
+@contextmanager
+def trace_span(
+    *,
+    name: str,
+    user_id: str,
+    session_id: str | None = None,
+    trace_id: str | None = None,
+    input: Any = None,
+    metadata: dict[str, Any] | None = None,
+    tags: list[str] | None = None,
+):
+    """Context manager that creates a Langfuse trace/span.
+
+    Yields the span object (or a ``_NullSpan`` if Langfuse is disabled).
+    A ``CallbackHandler`` created inside this block auto-inherits the trace
+    context, so there is no need to pass trace IDs manually.
+    """
+    lf = _get_client()
+    if lf is None:
+        yield _NullSpan()
+        return
+
+    try:
+        from langfuse import Langfuse, propagate_attributes
+
+        trace_ctx: dict[str, str] = {}
+        if trace_id is not None:
+            trace_ctx["trace_id"] = Langfuse.create_trace_id(seed=trace_id)
+
+        with lf.start_as_current_observation(
+            as_type="span",
+            name=name,
+            input=input,
+            metadata=metadata or {},
+            **({"trace_context": trace_ctx} if trace_ctx else {}),
+        ) as span:
+            with propagate_attributes(
+                user_id=user_id,
+                session_id=session_id,
+                tags=tags or [],
+            ):
+                yield span
+    except Exception as exc:
+        logger.warning("tracing: trace_span(%s) failed: %s", name, exc)
+        yield _NullSpan()
+
+
+# ── LangChain callback handler ──────────────────────────────────────────
+
+
+def get_langfuse_callback() -> Any | None:
+    """Return a LangChain ``CallbackHandler`` that auto-inherits the current trace.
+
+    Must be called inside a ``trace_span()`` block for proper linking.
+    Returns *None* when Langfuse is disabled.
+    """
+    if _disabled and not _initialised:
+        return None
+
+    try:
+        from langfuse.langchain import CallbackHandler
+        return CallbackHandler()
+    except Exception as exc:
+        logger.warning("tracing: get_langfuse_callback failed: %s", exc)
+        return None
+
+
+# ── Prompt management ────────────────────────────────────────────────────
+
+
+def get_prompt(
+    name: str,
+    *,
+    version: int | None = None,
+    label: str | None = None,
+    fallback: str | None = None,
+    cache_ttl_seconds: int = 300,
+) -> str | None:
+    """Fetch a managed prompt from Langfuse by name.
+
+    Returns the compiled prompt string, or *fallback* if the prompt is not
+    found or Langfuse is disabled.
+    """
+    lf = _get_client()
+    if lf is None:
+        return fallback
+
+    try:
+        kwargs: dict[str, Any] = {
+            "name": name,
+            "cache_ttl_seconds": cache_ttl_seconds,
+        }
+        if version is not None:
+            kwargs["version"] = version
+        if label is not None:
+            kwargs["label"] = label
+        prompt = lf.get_prompt(**kwargs)
+        return prompt.prompt
+    except Exception as exc:
+        logger.warning("tracing: get_prompt(%s) failed: %s", name, exc)
+        return fallback
+
+
+def link_prompt_to_trace(
+    span: Any,
+    prompt_name: str,
+    *,
+    version: int | None = None,
+    label: str | None = None,
+) -> None:
+    """Attach prompt metadata to a span/trace."""
+    lf = _get_client()
+    if lf is None or isinstance(span, _NullSpan):
+        return
+
+    try:
+        kwargs: dict[str, Any] = {"name": prompt_name}
+        if version is not None:
+            kwargs["version"] = version
+        if label is not None:
+            kwargs["label"] = label
+        prompt = lf.get_prompt(**kwargs)
+        span.update(metadata={"prompt": {"name": prompt_name, "version": prompt.version}})
+    except Exception as exc:
+        logger.warning("tracing: link_prompt_to_trace(%s) failed: %s", prompt_name, exc)
+
+
+# ── Scoring helper ───────────────────────────────────────────────────────
+
+
+def score_trace(
+    trace_id: str,
+    name: str,
+    value: float,
+    *,
+    comment: str | None = None,
+) -> None:
+    """Post a score to a trace (e.g. user feedback, latency, quality)."""
+    lf = _get_client()
+    if lf is None:
+        return
+
+    try:
+        lf.create_score(trace_id=trace_id, name=name, value=value, comment=comment)
+    except Exception as exc:
+        logger.warning("tracing: score_trace failed: %s", exc)
+
+
+# ── Shutdown ─────────────────────────────────────────────────────────────
+
+
+def flush() -> None:
+    """Flush pending Langfuse events."""
+    lf = _get_client()
+    if lf is not None:
+        try:
+            lf.flush()
+        except Exception as exc:
+            logger.warning("tracing: flush failed: %s", exc)
+
+
+def shutdown() -> None:
+    """Flush and close the Langfuse client."""
+    global _initialised, _disabled
+    lf = _get_client()
+    if lf is not None:
+        try:
+            lf.flush()
+            lf.shutdown()
+        except Exception as exc:
+            logger.warning("tracing: shutdown failed: %s", exc)
+    _initialised = False
+    _disabled = False

services/chat/app/ws_context.py

@@ -0,0 +1,115 @@
+"""WebSocket context for Chat Service — Redis-based tool call round-trip.
+
+Replaces the monolith's ws_context.py. Instead of calling Electron directly
+via WebSocket, this publishes tool_call frames to Redis (ws:out:{user_id})
+and awaits the result via BRPOP on tool:result:{call_id}.
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+from contextvars import ContextVar
+from typing import Any
+from uuid import uuid4
+
+from shared.redis import redis_client, tool_result_key, ws_out_channel
+
+logger = logging.getLogger(__name__)
+
+_TOOL_CALL_TIMEOUT = 30  # seconds — BRPOP timeout
+
+# Per-request user_id context var (set before agent runs)
+_current_user_id: ContextVar[str | None] = ContextVar("_current_user_id", default=None)
+
+# Optional collector for debug
+_tool_result_collector: ContextVar[list[dict] | None] = ContextVar(
+    "_tool_result_collector", default=None
+)
+
+
+def set_current_user(user_id: str) -> None:
+    _current_user_id.set(user_id)
+
+
+def clear_current_user() -> None:
+    _current_user_id.set(None)
+
+
+def set_tool_result_collector(lst: list[dict]) -> None:
+    _tool_result_collector.set(lst)
+
+
+def clear_tool_result_collector() -> None:
+    _tool_result_collector.set(None)
+
+
+async def execute_on_client(
+    action: str,
+    table: str | None = None,
+    data: dict[str, Any] | None = None,
+    filters: dict[str, Any] | None = None,
+    vector: list[float] | None = None,
+    limit: int | None = None,
+) -> dict[str, Any]:
+    """Send a tool_call to Electron via Redis and await the result.
+
+    1. Build tool_call payload
+    2. Publish to ws:out:{user_id} (WS Gateway forwards to Electron)
+    3. BRPOP on tool:result:{call_id} (WS Gateway pushes when Electron replies)
+    4. Return result dict
+
+    Raises RuntimeError if no user_id is set or if the call times out.
+    """
+    user_id = _current_user_id.get()
+    if not user_id:
+        raise RuntimeError(
+            "execute_on_client() called without a user_id — "
+            "set_current_user() must be called first."
+        )
+
+    call_id = str(uuid4())
+    payload: dict[str, Any] = {
+        "type": "tool_call",
+        "id": call_id,
+        "action": action,
+    }
+    if table is not None:
+        payload["table"] = table
+    if data is not None:
+        payload["data"] = data
+    if filters is not None:
+        payload["filters"] = {k: v for k, v in filters.items() if v is not None}
+    if vector is not None:
+        payload["vector"] = vector
+    if limit is not None:
+        payload["limit"] = limit
+
+    # Publish tool_call to WS Gateway → Electron
+    channel = ws_out_channel(user_id)
+    await redis_client.publish(channel, json.dumps(payload))
+
+    # Wait for Electron's tool_result
+    result_key = tool_result_key(call_id)
+    response = await redis_client.brpop(result_key, timeout=_TOOL_CALL_TIMEOUT)
+
+    if response is None:
+        raise RuntimeError(
+            f"Tool call {call_id} timed out after {_TOOL_CALL_TIMEOUT}s — "
+            f"device may be offline or unresponsive."
+        )
+
+    # response is (key, value) tuple
+    _, raw = response
+    result = json.loads(raw)
+
+    # Collect for debug if requested
+    collector = _tool_result_collector.get(None)
+    if collector is not None:
+        collector.append({
+            "action": action,
+            "table": table,
+            "data": result,
+        })
+
+    return result

services/chat/requirements.txt

@@ -0,0 +1,17 @@
+fastapi>=0.115.0
+uvicorn[standard]>=0.34.0
+gunicorn>=22.0.0
+pydantic>=2.10.0
+pydantic-settings>=2.7.0
+sqlalchemy>=2.0.0
+asyncpg>=0.30.0
+redis>=5.0.0
+cryptography>=42.0.0
+python-dotenv>=1.0.0
+langchain-core>=0.3.0
+langchain-openai>=0.3.0
+langchain-litellm>=0.3.0
+litellm>=1.50.0
+openai>=1.50.0
+httpx>=0.27.0
+langfuse>=3.0.0

services/ws-gateway/Dockerfile

@@ -0,0 +1,36 @@
+# ── builder ──────────────────────────────────────────────────────────────────
+FROM python:3.12-slim AS builder
+
+WORKDIR /build
+
+COPY services/ws-gateway/requirements.txt ./requirements.txt
+RUN pip install --upgrade pip && \
+    pip install --no-cache-dir --prefix=/install -r requirements.txt
+
+# ── runtime ──────────────────────────────────────────────────────────────────
+FROM python:3.12-slim AS runtime
+
+RUN addgroup --system appgroup && adduser --system --ingroup appgroup appuser
+
+WORKDIR /app
+
+COPY --from=builder /install /usr/local
+
+# Shared module
+COPY shared/ shared/
+
+# Service source
+COPY services/ws-gateway/app/ app/
+
+RUN chown -R appuser:appgroup /app
+
+USER appuser
+
+EXPOSE 8000
+
+# Single worker — each instance handles many WS connections via asyncio
+CMD ["gunicorn", "app.main:app", \
+     "-k", "uvicorn.workers.UvicornWorker", \
+     "--bind", "0.0.0.0:8000", \
+     "--workers", "1", \
+     "--timeout", "0"]