Make the governed RAG evidence path executable end to end

Separate local parsing from model indexing, bind review decisions to immutable manifests, persist vectors behind active profiles, and expose retrieval, chat, evaluation, and document workflows through the React workbench.

Constraint: Live Bailian authentication currently fails for all three configured capabilities

Rejected: Direct upload-to-embedding flow | bypasses local review and manifest binding

Confidence: high

Scope-risk: broad

Directive: Keep private-data deployment blocked until authentication, RBAC, and separate database roles land

Tested: make verify; fresh and replay Docker document smoke; worker recovery smoke; frozen synthetic evaluation; migration 0003-0004 roundtrip

Not-tested: Successful live Bailian calls, OCR, real multi-user authorization

compose.yaml

@@ -10,6 +10,7 @@ x-runtime-config: &runtime-config
   POSTGRES_PASSWORD_FILE: /run/secrets/postgres_app_password
   UPLOAD_ROOT: ${UPLOAD_ROOT:-/data/uploads}
   MAX_UPLOAD_MB: "${MAX_UPLOAD_MB:-100}"
+  DOCUMENT_NAMESPACE_MODE: ${DOCUMENT_NAMESPACE_MODE:-fake}
 
 x-rag-config: &rag-config
   DASHSCOPE_API_KEY_FILE: /run/secrets/bailian_api_key
@@ -87,6 +88,29 @@ services:
       - data
     restart: "no"
 
+  upload-init:
+    build:
+      context: ./backend
+    command: ["python", "-m", "app.tools.init_upload_storage"]
+    environment:
+      UPLOAD_ROOT: /data/uploads
+    user: "0:0"
+    network_mode: none
+    volumes:
+      - uploads_data:/data/uploads
+    read_only: true
+    tmpfs:
+      - /tmp
+    security_opt:
+      - no-new-privileges:true
+    cap_drop:
+      - ALL
+    cap_add:
+      - CHOWN
+      - FOWNER
+      - DAC_OVERRIDE
+    restart: "no"
+
   api:
     build:
       context: ./backend
@@ -96,6 +120,8 @@ services:
         condition: service_healthy
       migrate:
         condition: service_completed_successfully
+      upload-init:
+        condition: service_completed_successfully
       model-gateway:
         condition: service_healthy
     environment:
@@ -103,6 +129,8 @@ services:
     secrets:
       - postgres_app_password
       - model_gateway_api_token
+    volumes:
+      - uploads_data:/data/uploads
     networks:
       - data
       - model
@@ -165,6 +193,70 @@ services:
       - ALL
     restart: unless-stopped
 
+  worker-local:
+    build:
+      context: ./backend
+    command: ["python", "-m", "app.worker"]
+    depends_on:
+      db:
+        condition: service_healthy
+      migrate:
+        condition: service_completed_successfully
+      upload-init:
+        condition: service_completed_successfully
+    environment:
+      <<: *runtime-config
+      WORKER_CAPABILITIES: document_parse
+    secrets:
+      - postgres_app_password
+    volumes:
+      - uploads_data:/data/uploads
+    networks:
+      - data
+    init: true
+    read_only: true
+    tmpfs:
+      - /tmp
+    security_opt:
+      - no-new-privileges:true
+    cap_drop:
+      - ALL
+    stop_grace_period: 150s
+    restart: unless-stopped
+
+  worker-model:
+    build:
+      context: ./backend
+    command: ["python", "-m", "app.worker"]
+    depends_on:
+      db:
+        condition: service_healthy
+      migrate:
+        condition: service_completed_successfully
+      model-gateway:
+        condition: service_healthy
+    environment:
+      <<: [*runtime-config, *model-client-config]
+      WORKER_CAPABILITIES: embedding
+      MODEL_GATEWAY_TOKEN_FILE: /run/secrets/model_gateway_worker_token
+      MODEL_GATEWAY_CALLER: worker
+    secrets:
+      - postgres_app_password
+      - model_gateway_worker_token
+    networks:
+      - data
+      - model
+    init: true
+    read_only: true
+    tmpfs:
+      - /tmp
+    security_opt:
+      - no-new-privileges:true
+    cap_drop:
+      - ALL
+    stop_grace_period: 150s
+    restart: unless-stopped
+
   gateway:
     build:
       context: ./backend
@@ -298,8 +390,63 @@ services:
       - ./data/samples/public:/demo:ro
     restart: "no"
 
+  worker-smoke:
+    build:
+      context: ./backend
+    command: ["python", "-m", "app.tools.worker_smoke"]
+    profiles: ["tools"]
+    depends_on:
+      migrate:
+        condition: service_completed_successfully
+    environment:
+      <<: *runtime-config
+    secrets:
+      - postgres_app_password
+    networks:
+      - data
+    init: true
+    read_only: true
+    tmpfs:
+      - /tmp
+    security_opt:
+      - no-new-privileges:true
+    cap_drop:
+      - ALL
+    restart: "no"
+
+  document-pipeline-smoke:
+    build:
+      context: ./backend
+    command: ["python", "-m", "app.tools.document_pipeline_smoke"]
+    profiles: ["tools"]
+    depends_on:
+      gateway:
+        condition: service_healthy
+      worker-local:
+        condition: service_started
+      worker-model:
+        condition: service_started
+    environment:
+      RAG_BASE_URL: http://gateway:8000
+      RAG_UPLOAD_SAMPLE: /demo/upload_demo.md
+      DOCUMENT_NAMESPACE_MODE: ${DOCUMENT_NAMESPACE_MODE:-fake}
+    volumes:
+      - ./data/samples/public:/demo:ro
+    networks:
+      - ingress
+    init: true
+    read_only: true
+    tmpfs:
+      - /tmp
+    security_opt:
+      - no-new-privileges:true
+    cap_drop:
+      - ALL
+    restart: "no"
+
 volumes:
   postgres_data:
+  uploads_data:
 
 networks:
   edge: