Merge branch 'main' into fix-2804

.env.example

@@ -9,8 +9,9 @@ JINA_API_KEY=your-jina-api-key
 
 # InfoQuest API Key
 INFOQUEST_API_KEY=your-infoquest-api-key
-# CORS Origins (comma-separated) - e.g., http://localhost:3000,http://localhost:3001
-# CORS_ORIGINS=http://localhost:3000
+# Browser CORS allowlist for split-origin or port-forwarded deployments (comma-separated exact origins).
+# Leave unset when using the unified nginx endpoint, e.g. http://localhost:2026.
+# GATEWAY_CORS_ORIGINS=http://localhost:3000,http://127.0.0.1:3000
 
 # Optional:
 # FIRECRAWL_API_KEY=your-firecrawl-api-key

CONTRIBUTING.md

@@ -46,12 +46,12 @@ Docker provides a consistent, isolated environment with all dependencies pre-con
    All services will start with hot-reload enabled:
    - Frontend changes are automatically reloaded
    - Backend changes trigger automatic restart
-   - LangGraph server supports hot-reload
+   - Gateway-hosted LangGraph-compatible runtime supports hot-reload
 
 4. **Access the application**:
    - Web Interface: http://localhost:2026
    - API Gateway: http://localhost:2026/api/*
-   - LangGraph: http://localhost:2026/api/langgraph/*
+   - LangGraph-compatible API: http://localhost:2026/api/langgraph/*
 
 #### Docker Commands
 
@@ -94,7 +94,7 @@ Use these as practical starting points for development and review environments:
 If `make docker-init`, `make docker-start`, or `make docker-stop` fails on Linux with an error like below, your current user likely does not have permission to access the Docker daemon socket:
 
 ```text
-unable to get image 'deer-flow-dev-langgraph': permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock
+unable to get image 'deer-flow-gateway': permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock
 ```
 
 Recommended fix: add your current user to the `docker` group so Docker commands work without `sudo`.
@@ -131,9 +131,8 @@ Host Machine
 Docker Compose (deer-flow-dev)
   ├→ nginx (port 2026) ← Reverse proxy
   ├→ web (port 3000) ← Frontend with hot-reload
-  ├→ api (port 8001) ← Gateway API with hot-reload
-   ├→ langgraph (port 2024) ← LangGraph server with hot-reload
-   └→ provisioner (optional, port 8002) ← Started only in provisioner/K8s sandbox mode
+  ├→ gateway (port 8001) ← Gateway API + LangGraph-compatible runtime with hot-reload
+  └→ provisioner (optional, port 8002) ← Started only in provisioner/K8s sandbox mode
 ```
 
 **Benefits of Docker Development**:
@@ -184,17 +183,13 @@ Required tools:
 
 If you need to start services individually:
 
-1. **Start backend services**:
+1. **Start backend service**:
    ```bash
-   # Terminal 1: Start LangGraph Server (port 2024)
-   cd backend
-   make dev
-
-   # Terminal 2: Start Gateway API (port 8001)
+   # Terminal 1: Start Gateway API and embedded LangGraph-compatible runtime (port 8001)
    cd backend
    make gateway
 
-   # Terminal 3: Start Frontend (port 3000)
+   # Terminal 2: Start Frontend (port 3000)
    cd frontend
    pnpm dev
    ```
@@ -212,10 +207,10 @@ If you need to start services individually:
 
 The nginx configuration provides:
 - Unified entry point on port 2026
-- Routes `/api/langgraph/*` to LangGraph Server (2024)
+- Gateway owns `/api/langgraph/*` and translates those public LangGraph-compatible paths to its native `/api/*` routers behind nginx
 - Routes other `/api/*` endpoints to Gateway API (8001)
 - Routes non-API requests to Frontend (3000)
-- Centralized CORS handling
+- Same-origin API routing; split-origin or port-forwarded browser clients should use the Gateway `GATEWAY_CORS_ORIGINS` allowlist
 - SSE/streaming support for real-time agent responses
 - Optimized timeouts for long-running operations
 
@@ -235,8 +230,8 @@ deer-flow/
 │       └── nginx.local.conf # Nginx config for local dev
 ├── backend/                 # Backend application
 │   ├── src/
-│   │   ├── gateway/        # Gateway API (port 8001)
-│   │   ├── agents/         # LangGraph agents (port 2024)
+│   │   ├── gateway/        # Gateway API and LangGraph-compatible runtime (port 8001)
+│   │   ├── agents/         # LangGraph agent definitions
 │   │   ├── mcp/            # Model Context Protocol integration
 │   │   ├── skills/         # Skills system
 │   │   └── sandbox/        # Sandbox execution
@@ -256,8 +251,7 @@ Browser
   ↓
 Nginx (port 2026) ← Unified entry point
   ├→ Frontend (port 3000) ← / (non-API requests)
-  ├→ Gateway API (port 8001) ← /api/models, /api/mcp, /api/skills, /api/threads/*/artifacts
-  └→ LangGraph Server (port 2024) ← /api/langgraph/* (agent interactions)
+  └→ Gateway API (port 8001) ← /api/* and /api/langgraph/* (LangGraph-compatible agent interactions)
 ```
 
 ## Development Workflow

README.md

@@ -245,6 +245,8 @@ make down   # Stop and remove containers
 
 Access: http://localhost:2026
 
+The unified nginx endpoint is same-origin by default and does not emit browser CORS headers. If you run a split-origin or port-forwarded browser client, set `GATEWAY_CORS_ORIGINS` to comma-separated exact origins such as `http://localhost:3000`; the Gateway then applies the CORS allowlist and matching CSRF origin checks.
+
 See [CONTRIBUTING.md](CONTRIBUTING.md) for detailed Docker development guide.
 
 #### Option 2: Local Development

backend/CLAUDE.md

@@ -207,6 +207,8 @@ Configuration priority:
 
 FastAPI application on port 8001 with health check at `GET /health`. Set `GATEWAY_ENABLE_DOCS=false` to disable `/docs`, `/redoc`, and `/openapi.json` in production (default: enabled).
 
+CORS is same-origin by default when requests enter through nginx on port 2026. Split-origin or port-forwarded browser clients must opt in with `GATEWAY_CORS_ORIGINS` (comma-separated exact origins); Gateway `CORSMiddleware` and `CSRFMiddleware` both read that variable so browser CORS and auth-origin checks stay aligned.
+
 **Routers**:
 
 | Router | Endpoints |
@@ -223,7 +225,7 @@ FastAPI application on port 8001 with health check at `GET /health`. Set `GATEWA
 | **Feedback** (`/api/threads/{id}/runs/{rid}/feedback`) | `PUT /` - upsert feedback; `DELETE /` - delete user feedback; `POST /` - create feedback; `GET /` - list feedback; `GET /stats` - aggregate stats; `DELETE /{fid}` - delete specific |
 | **Runs** (`/api/runs`) | `POST /stream` - stateless run + SSE; `POST /wait` - stateless run + block; `GET /{rid}/messages` - paginated messages by run_id `{data, has_more}` (cursor: `after_seq`/`before_seq`); `GET /{rid}/feedback` - list feedback by run_id |
 
-Proxied through nginx: `/api/langgraph/*` → LangGraph, all other `/api/*` → Gateway.
+Proxied through nginx: `/api/langgraph/*` → Gateway LangGraph-compatible runtime, all other `/api/*` → Gateway REST APIs.
 
 ### Sandbox System (`packages/harness/deerflow/sandbox/`)
 

backend/README.md

@@ -14,28 +14,31 @@ DeerFlow is a LangGraph-based AI super agent with sandbox execution, persistent
                                 │                  │
               /api/langgraph/*  │                  │  /api/* (other)
                                 ▼                  ▼
-               ┌────────────────────┐  ┌────────────────────────┐
-               │ LangGraph Server   │  │   Gateway API (8001)   │
-               │    (Port 2024)     │  │   FastAPI REST         │
-               │                    │  │                        │
-               │ ┌────────────────┐ │  │ Models, MCP, Skills,   │
-               │ │  Lead Agent    │ │  │ Memory, Uploads,       │
-               │ │  ┌──────────┐  │ │  │ Artifacts              │
-               │ │  │Middleware│  │ │  └────────────────────────┘
-               │ │  │  Chain   │  │ │
-               │ │  └──────────┘  │ │
-               │ │  ┌──────────┐  │ │
-               │ │  │  Tools   │  │ │
-               │ │  └──────────┘  │ │
-               │ │  ┌──────────┐  │ │
-               │ │  │Subagents │  │ │
-               │ │  └──────────┘  │ │
-               │ └────────────────┘ │
-               └────────────────────┘
+               ┌──────────────────────────────────────────────┐
+               │             Gateway API (8001)               │
+               │  FastAPI REST + LangGraph-compatible runtime │
+               │                                              │
+               │ Models, MCP, Skills, Memory, Uploads,       │
+               │ Artifacts, Threads, Runs, Streaming          │
+               │                                              │
+               │ ┌────────────────┐                           │
+               │ │  Lead Agent    │                           │
+               │ │  ┌──────────┐  │                           │
+               │ │  │Middleware│  │                           │
+               │ │  │  Chain   │  │                           │
+               │ │  └──────────┘  │                           │
+               │ │  ┌──────────┐  │                           │
+               │ │  │  Tools   │  │                           │
+               │ │  └──────────┘  │                           │
+               │ │  ┌──────────┐  │                           │
+               │ │  │Subagents │  │                           │
+               │ │  └──────────┘  │                           │
+               │ └────────────────┘                           │
+               └──────────────────────────────────────────────┘
 ```
 
 **Request Routing** (via Nginx):
-- `/api/langgraph/*` → LangGraph Server - agent interactions, threads, streaming
+- `/api/langgraph/*` → Gateway API - LangGraph-compatible agent interactions, threads, runs, and streaming translated to native `/api/*` routers
 - `/api/*` (other) → Gateway API - models, MCP, skills, memory, artifacts, uploads, thread-local cleanup
 - `/` (non-API) → Frontend - Next.js web interface
 

backend/app/gateway/app.py

@@ -1,6 +1,5 @@
 import asyncio
 import logging
-import os
 from collections.abc import AsyncGenerator
 from contextlib import asynccontextmanager
 
@@ -9,7 +8,7 @@ from fastapi.middleware.cors import CORSMiddleware
 
 from app.gateway.auth_middleware import AuthMiddleware
 from app.gateway.config import get_gateway_config
-from app.gateway.csrf_middleware import CSRFMiddleware
+from app.gateway.csrf_middleware import CSRFMiddleware, get_configured_cors_origins
 from app.gateway.deps import langgraph_runtime
 from app.gateway.routers import (
     agents,
@@ -219,7 +218,9 @@ def create_app() -> FastAPI:
         Configured FastAPI application instance.
     """
     config = get_gateway_config()
-    docs_kwargs = {"docs_url": "/docs", "redoc_url": "/redoc", "openapi_url": "/openapi.json"} if config.enable_docs else {"docs_url": None, "redoc_url": None, "openapi_url": None}
+    docs_url = "/docs" if config.enable_docs else None
+    redoc_url = "/redoc" if config.enable_docs else None
+    openapi_url = "/openapi.json" if config.enable_docs else None
 
     app = FastAPI(
         title="DeerFlow API Gateway",
@@ -239,12 +240,14 @@ API Gateway for DeerFlow - A LangGraph-based AI agent backend with sandbox execu
 
 ### Architecture
 
-LangGraph requests are handled by nginx reverse proxy.
-This gateway provides custom endpoints for models, MCP configuration, skills, and artifacts.
+LangGraph-compatible requests are routed through nginx to this gateway.
+This gateway provides runtime endpoints for agent runs plus custom endpoints for models, MCP configuration, skills, and artifacts.
         """,
         version="0.1.0",
         lifespan=lifespan,
-        **docs_kwargs,
+        docs_url=docs_url,
+        redoc_url=redoc_url,
+        openapi_url=openapi_url,
         openapi_tags=[
             {
                 "name": "models",
@@ -307,25 +310,18 @@ This gateway provides custom endpoints for models, MCP configuration, skills, an
     # CSRF: Double Submit Cookie pattern for state-changing requests
     app.add_middleware(CSRFMiddleware)
 
-    # CORS: when GATEWAY_CORS_ORIGINS is set (dev without nginx), add CORS middleware.
-    # In production, nginx handles CORS and no middleware is needed.
-    cors_origins_env = os.environ.get("GATEWAY_CORS_ORIGINS", "")
-    if cors_origins_env:
-        cors_origins = [o.strip() for o in cors_origins_env.split(",") if o.strip()]
-        # Validate: wildcard origin with credentials is a security misconfiguration
-        for origin in cors_origins:
-            if origin == "*":
-                logger.error("GATEWAY_CORS_ORIGINS contains wildcard '*' with allow_credentials=True. This is a security misconfiguration — browsers will reject the response. Use explicit scheme://host:port origins instead.")
-                cors_origins = [o for o in cors_origins if o != "*"]
-                break
-        if cors_origins:
-            app.add_middleware(
-                CORSMiddleware,
-                allow_origins=cors_origins,
-                allow_credentials=True,
-                allow_methods=["*"],
-                allow_headers=["*"],
-            )
+    # CORS: the unified nginx endpoint is same-origin by default. Split-origin
+    # browser clients must opt in with this explicit Gateway allowlist so CORS
+    # and CSRF origin checks share the same source of truth.
+    cors_origins = sorted(get_configured_cors_origins())
+    if cors_origins:
+        app.add_middleware(
+            CORSMiddleware,
+            allow_origins=cors_origins,
+            allow_credentials=True,
+            allow_methods=["*"],
+            allow_headers=["*"],
+        )
 
     # Include routers
     # Models API is mounted at /api/models
@@ -374,7 +370,7 @@ This gateway provides custom endpoints for models, MCP configuration, skills, an
     app.include_router(runs.router)
 
     @app.get("/health", tags=["health"])
-    async def health_check() -> dict:
+    async def health_check() -> dict[str, str]:
         """Health check endpoint.
 
         Returns:

backend/app/gateway/config.py

@@ -8,7 +8,6 @@ class GatewayConfig(BaseModel):
 
     host: str = Field(default="0.0.0.0", description="Host to bind the gateway server")
     port: int = Field(default=8001, description="Port to bind the gateway server")
-    cors_origins: list[str] = Field(default_factory=lambda: ["http://localhost:3000"], description="Allowed CORS origins")
     enable_docs: bool = Field(default=True, description="Enable Swagger/ReDoc/OpenAPI endpoints")
 
 
@@ -19,11 +18,9 @@ def get_gateway_config() -> GatewayConfig:
     """Get gateway config, loading from environment if available."""
     global _gateway_config
     if _gateway_config is None:
-        cors_origins_str = os.getenv("CORS_ORIGINS", "http://localhost:3000")
         _gateway_config = GatewayConfig(
             host=os.getenv("GATEWAY_HOST", "0.0.0.0"),
             port=int(os.getenv("GATEWAY_PORT", "8001")),
-            cors_origins=cors_origins_str.split(","),
             enable_docs=os.getenv("GATEWAY_ENABLE_DOCS", "true").lower() == "true",
         )
     return _gateway_config

backend/app/gateway/csrf_middleware.py

@@ -6,7 +6,7 @@ State-changing operations require CSRF protection.
 
 import os
 import secrets
-from collections.abc import Callable
+from collections.abc import Awaitable, Callable
 from urllib.parse import urlsplit
 
 from fastapi import Request, Response
@@ -106,6 +106,11 @@ def _configured_cors_origins() -> set[str]:
     return origins
 
 
+def get_configured_cors_origins() -> set[str]:
+    """Return normalized explicit browser origins from GATEWAY_CORS_ORIGINS."""
+    return _configured_cors_origins()
+
+
 def _first_header_value(value: str | None) -> str | None:
     """Return the first value from a comma-separated proxy header."""
     if not value:
@@ -172,7 +177,7 @@ class CSRFMiddleware(BaseHTTPMiddleware):
     def __init__(self, app: ASGIApp) -> None:
         super().__init__(app)
 
-    async def dispatch(self, request: Request, call_next: Callable) -> Response:
+    async def dispatch(self, request: Request, call_next: Callable[[Request], Awaitable[Response]]) -> Response:
         _is_auth = is_auth_endpoint(request)
 
         if should_check_csrf(request) and _is_auth and not is_allowed_auth_origin(request):

backend/app/gateway/services.py

@@ -19,6 +19,7 @@ from langchain_core.messages import HumanMessage
 
 from app.gateway.deps import get_run_context, get_run_manager, get_stream_bridge
 from app.gateway.utils import sanitize_log_param
+from deerflow.config.app_config import get_app_config
 from deerflow.runtime import (
     END_SENTINEL,
     HEARTBEAT_SENTINEL,
@@ -267,6 +268,23 @@ async def start_run(
 
     disconnect = DisconnectMode.cancel if body.on_disconnect == "cancel" else DisconnectMode.continue_
 
+    body_context = getattr(body, "context", None) or {}
+    model_name = body_context.get("model_name")
+
+    # Coerce non-string model_name values to str before truncation.
+    if model_name is not None and not isinstance(model_name, str):
+        model_name = str(model_name)
+
+    # Validate model against the allowlist when a model_name is provided.
+    if model_name:
+        app_config = get_app_config()
+        resolved = app_config.get_model_config(model_name)
+        if resolved is None:
+            raise HTTPException(
+                status_code=400,
+                detail=f"Model {model_name!r} is not in the configured model allowlist",
+            )
+
     try:
         record = await run_mgr.create_or_reject(
             thread_id,
@@ -275,6 +293,7 @@ async def start_run(
             metadata=body.metadata or {},
             kwargs={"input": body.input, "config": body.config},
             multitask_strategy=body.multitask_strategy,
+            model_name=model_name,
         )
     except ConflictError as exc:
         raise HTTPException(status_code=409, detail=str(exc)) from exc

backend/docs/API.md

@@ -6,16 +6,16 @@ This document provides a complete reference for the DeerFlow backend APIs.
 
 DeerFlow backend exposes two sets of APIs:
 
-1. **LangGraph API** - Agent interactions, threads, and streaming (`/api/langgraph/*`)
+1. **LangGraph-compatible API** - Agent interactions, threads, and streaming (`/api/langgraph/*`)
 2. **Gateway API** - Models, MCP, skills, uploads, and artifacts (`/api/*`)
 
 All APIs are accessed through the Nginx reverse proxy at port 2026.
 
-## LangGraph API
+## LangGraph-compatible API
 
 Base URL: `/api/langgraph`
 
-The LangGraph API is provided by the LangGraph server and follows the LangGraph SDK conventions.
+The public LangGraph-compatible API follows LangGraph SDK conventions. In the unified nginx deployment, Gateway owns `/api/langgraph/*` and translates those paths to its native `/api/*` run, thread, and streaming routers.
 
 ### Threads
 
@@ -104,17 +104,11 @@ Content-Type: application/json
 **Recursion Limit:**
 
 `config.recursion_limit` caps the number of graph steps LangGraph will execute
-in a single run. The `/api/langgraph/*` endpoints go straight to the LangGraph
-server and therefore inherit LangGraph's native default of **25**, which is
-too low for plan-mode or subagent-heavy runs — the agent typically errors out
-with `GraphRecursionError` after the first round of subagent results comes
-back, before the lead agent can synthesize the final answer.
-
-DeerFlow's own Gateway and IM-channel paths mitigate this by defaulting to
-`100` in `build_run_config` (see `backend/app/gateway/services.py`), but
-clients calling the LangGraph API directly must set `recursion_limit`
-explicitly in the request body. `100` matches the Gateway default and is a
-safe starting point; increase it if you run deeply nested subagent graphs.
+in a single run. The unified Gateway path defaults to `100` in
+`build_run_config` (see `backend/app/gateway/services.py`), which is a safer
+starting point for plan-mode or subagent-heavy runs. Clients can still set
+`recursion_limit` explicitly in the request body; increase it if you run deeply
+nested subagent graphs.
 
 **Configurable Options:**
 - `model_name` (string): Override the default model
@@ -649,7 +643,7 @@ curl -X POST http://localhost:2026/api/langgraph/threads/abc123/runs \
   }'
 ```
 
-> The `/api/langgraph/*` endpoints bypass DeerFlow's Gateway and inherit
-> LangGraph's native `recursion_limit` default of 25, which is too low for
-> plan-mode or subagent runs. Set `config.recursion_limit` explicitly — see
-> the [Create Run](#create-run) section for details.
+> The unified Gateway path defaults `config.recursion_limit` to 100 for
+> plan-mode and subagent-heavy runs. Clients may still set
+> `config.recursion_limit` explicitly — see the [Create Run](#create-run)
+> section for details.

backend/docs/ARCHITECTURE.md

@@ -14,8 +14,8 @@ This document provides a comprehensive overview of the DeerFlow backend architec
 │                          Nginx (Port 2026)                               │
 │                    Unified Reverse Proxy Entry Point                      │
 │  ┌────────────────────────────────────────────────────────────────────┐  │
-│  │  /api/langgraph/*  →  LangGraph Server (2024)                      │  │
-│  │  /api/*            →  Gateway API (8001)                           │  │
+│  │  /api/langgraph/*  →  Gateway LangGraph-compatible runtime (8001)  │  │
+│  │  /api/*            →  Gateway REST APIs (8001)                     │  │
 │  │  /*                →  Frontend (3000)                               │  │
 │  └────────────────────────────────────────────────────────────────────┘  │
 └─────────────────────────────────┬────────────────────────────────────────┘
@@ -24,8 +24,8 @@ This document provides a comprehensive overview of the DeerFlow backend architec
           │                       │                       │
           ▼                       ▼                       ▼
 ┌─────────────────────┐ ┌─────────────────────┐ ┌─────────────────────┐
-│   LangGraph Server  │ │    Gateway API      │ │     Frontend        │
-│     (Port 2024)     │ │    (Port 8001)      │ │    (Port 3000)      │
+│ Embedded Runtime    │ │    Gateway API      │ │     Frontend        │
+│  (inside Gateway)   │ │    (Port 8001)      │ │    (Port 3000)      │
 │                     │ │                     │ │                     │
 │  - Agent Runtime    │ │  - Models API       │ │  - Next.js App      │
 │  - Thread Mgmt      │ │  - MCP Config       │ │  - React UI         │
@@ -52,9 +52,9 @@ This document provides a comprehensive overview of the DeerFlow backend architec
 
 ## Component Details
 
-### LangGraph Server
+### Embedded LangGraph Runtime
 
-The LangGraph server is the core agent runtime, built on LangGraph for robust multi-agent workflow orchestration.
+The LangGraph-compatible runtime runs inside the Gateway process and is built on LangGraph for robust multi-agent workflow orchestration.
 
 **Entry Point**: `packages/harness/deerflow/agents/lead_agent/agent.py:make_lead_agent`
 
@@ -78,7 +78,7 @@ The LangGraph server is the core agent runtime, built on LangGraph for robust mu
 
 ### Gateway API
 
-FastAPI application providing REST endpoints for non-agent operations.
+FastAPI application providing REST endpoints plus the public LangGraph-compatible `/api/langgraph/*` runtime routes.
 
 **Entry Point**: `app/gateway/app.py`
 
@@ -353,10 +353,10 @@ SKILL.md Format:
    POST /api/langgraph/threads/{thread_id}/runs
    {"input": {"messages": [{"role": "user", "content": "Hello"}]}}
 
-2. Nginx → LangGraph Server (2024)
-   Proxied to LangGraph server
+2. Nginx → Gateway API (8001)
+   Routes `/api/langgraph/*` to the Gateway's LangGraph-compatible runtime
 
-3. LangGraph Server
+3. Embedded LangGraph runtime
    a. Load/create thread state
    b. Execute middleware chain:
       - ThreadDataMiddleware: Set up paths

backend/packages/harness/deerflow/agents/middlewares/dangling_tool_call_middleware.py

@@ -36,42 +36,73 @@ class DanglingToolCallMiddleware(AgentMiddleware[AgentState]):
 
     @staticmethod
     def _message_tool_calls(msg) -> list[dict]:
-        """Return normalized tool calls from structured fields or raw provider payloads."""
+        """Return normalized tool calls from structured fields or raw provider payloads.
+
+        LangChain stores malformed provider function calls in ``invalid_tool_calls``.
+        They do not execute, but provider adapters may still serialize enough of
+        the call id/name back into the next request that strict OpenAI-compatible
+        validators expect a matching ToolMessage. Treat them as dangling calls so
+        the next model request stays well-formed and the model sees a recoverable
+        tool error instead of another provider 400.
+        """
+        normalized: list[dict] = []
+
         tool_calls = getattr(msg, "tool_calls", None) or []
-        if tool_calls:
-            return list(tool_calls)
+        normalized.extend(list(tool_calls))
 
         raw_tool_calls = (getattr(msg, "additional_kwargs", None) or {}).get("tool_calls") or []
-        normalized: list[dict] = []
-        for raw_tc in raw_tool_calls:
-            if not isinstance(raw_tc, dict):
+        if not tool_calls:
+            for raw_tc in raw_tool_calls:
+                if not isinstance(raw_tc, dict):
+                    continue
+
+                function = raw_tc.get("function")
+                name = raw_tc.get("name")
+                if not name and isinstance(function, dict):
+                    name = function.get("name")
+
+                args = raw_tc.get("args", {})
+                if not args and isinstance(function, dict):
+                    raw_args = function.get("arguments")
+                    if isinstance(raw_args, str):
+                        try:
+                            parsed_args = json.loads(raw_args)
+                        except (TypeError, ValueError, json.JSONDecodeError):
+                            parsed_args = {}
+                        args = parsed_args if isinstance(parsed_args, dict) else {}
+
+                normalized.append(
+                    {
+                        "id": raw_tc.get("id"),
+                        "name": name or "unknown",
+                        "args": args if isinstance(args, dict) else {},
+                    }
+                )
+
+        for invalid_tc in getattr(msg, "invalid_tool_calls", None) or []:
+            if not isinstance(invalid_tc, dict):
                 continue
-
-            function = raw_tc.get("function")
-            name = raw_tc.get("name")
-            if not name and isinstance(function, dict):
-                name = function.get("name")
-
-            args = raw_tc.get("args", {})
-            if not args and isinstance(function, dict):
-                raw_args = function.get("arguments")
-                if isinstance(raw_args, str):
-                    try:
-                        parsed_args = json.loads(raw_args)
-                    except (TypeError, ValueError, json.JSONDecodeError):
-                        parsed_args = {}
-                    args = parsed_args if isinstance(parsed_args, dict) else {}
-
             normalized.append(
                 {
-                    "id": raw_tc.get("id"),
-                    "name": name or "unknown",
-                    "args": args if isinstance(args, dict) else {},
+                    "id": invalid_tc.get("id"),
+                    "name": invalid_tc.get("name") or "unknown",
+                    "args": {},
+                    "invalid": True,
+                    "error": invalid_tc.get("error"),
                 }
             )
 
         return normalized
 
+    @staticmethod
+    def _synthetic_tool_message_content(tool_call: dict) -> str:
+        if tool_call.get("invalid"):
+            error = tool_call.get("error")
+            if isinstance(error, str) and error:
+                return f"[Tool call could not be executed because its arguments were invalid: {error}]"
+            return "[Tool call could not be executed because its arguments were invalid.]"
+        return "[Tool call was interrupted and did not return a result.]"
+
     def _build_patched_messages(self, messages: list) -> list | None:
         """Return a new message list with patches inserted at the correct positions.
 
@@ -114,7 +145,7 @@ class DanglingToolCallMiddleware(AgentMiddleware[AgentState]):
                 if tc_id and tc_id not in existing_tool_msg_ids and tc_id not in patched_ids:
                     patched.append(
                         ToolMessage(
-                            content="[Tool call was interrupted and did not return a result.]",
+                            content=self._synthetic_tool_message_content(tc),
                             tool_call_id=tc_id,
                             name=tc.get("name", "unknown"),
                             status="error",

backend/packages/harness/deerflow/mcp/tools.py

@@ -1,11 +1,6 @@
 """Load MCP tools using langchain-mcp-adapters."""
 
-import asyncio
-import atexit
-import concurrent.futures
 import logging
-from collections.abc import Callable
-from typing import Any
 
 from langchain_core.tools import BaseTool
 
@@ -13,46 +8,10 @@ from deerflow.config.extensions_config import ExtensionsConfig
 from deerflow.mcp.client import build_servers_config
 from deerflow.mcp.oauth import build_oauth_tool_interceptor, get_initial_oauth_headers
 from deerflow.reflection import resolve_variable
+from deerflow.tools.sync import make_sync_tool_wrapper
 
 logger = logging.getLogger(__name__)
 
-# Global thread pool for sync tool invocation in async environments
-_SYNC_TOOL_EXECUTOR = concurrent.futures.ThreadPoolExecutor(max_workers=10, thread_name_prefix="mcp-sync-tool")
-
-# Register shutdown hook for the global executor
-atexit.register(lambda: _SYNC_TOOL_EXECUTOR.shutdown(wait=False))
-
-
-def _make_sync_tool_wrapper(coro: Callable[..., Any], tool_name: str) -> Callable[..., Any]:
-    """Build a synchronous wrapper for an asynchronous tool coroutine.
-
-    Args:
-        coro: The tool's asynchronous coroutine.
-        tool_name: Name of the tool (for logging).
-
-    Returns:
-        A synchronous function that correctly handles nested event loops.
-    """
-
-    def sync_wrapper(*args: Any, **kwargs: Any) -> Any:
-        try:
-            loop = asyncio.get_running_loop()
-        except RuntimeError:
-            loop = None
-
-        try:
-            if loop is not None and loop.is_running():
-                # Use global executor to avoid nested loop issues and improve performance
-                future = _SYNC_TOOL_EXECUTOR.submit(asyncio.run, coro(*args, **kwargs))
-                return future.result()
-            else:
-                return asyncio.run(coro(*args, **kwargs))
-        except Exception as e:
-            logger.error(f"Error invoking MCP tool '{tool_name}' via sync wrapper: {e}", exc_info=True)
-            raise
-
-    return sync_wrapper
-
 
 async def get_mcp_tools() -> list[BaseTool]:
     """Get all tools from enabled MCP servers.
@@ -126,7 +85,7 @@ async def get_mcp_tools() -> list[BaseTool]:
         # Patch tools to support sync invocation, as deerflow client streams synchronously
         for tool in tools:
             if getattr(tool, "func", None) is None and getattr(tool, "coroutine", None) is not None:
-                tool.func = _make_sync_tool_wrapper(tool.coroutine, tool.name)
+                tool.func = make_sync_tool_wrapper(tool.coroutine, tool.name)
 
         return tools
 

backend/packages/harness/deerflow/persistence/run/sql.py

@@ -23,6 +23,18 @@ class RunRepository(RunStore):
     def __init__(self, session_factory: async_sessionmaker[AsyncSession]) -> None:
         self._sf = session_factory
 
+    @staticmethod
+    def _normalize_model_name(model_name: str | None) -> str | None:
+        """Normalize model_name for storage: strip whitespace, truncate to 128 chars."""
+        if model_name is None:
+            return None
+        if not isinstance(model_name, str):
+            model_name = str(model_name)
+        normalized = model_name.strip()
+        if len(normalized) > 128:
+            normalized = normalized[:128]
+        return normalized
+
     @staticmethod
     def _safe_json(obj: Any) -> Any:
         """Ensure obj is JSON-serializable. Falls back to model_dump() or str()."""
@@ -70,6 +82,7 @@ class RunRepository(RunStore):
         thread_id,
         assistant_id=None,
         user_id: str | None | _AutoSentinel = AUTO,
+        model_name: str | None = None,
         status="pending",
         multitask_strategy="reject",
         metadata=None,
@@ -85,6 +98,7 @@ class RunRepository(RunStore):
             thread_id=thread_id,
             assistant_id=assistant_id,
             user_id=resolved_user_id,
+            model_name=self._normalize_model_name(model_name),
             status=status,
             multitask_strategy=multitask_strategy,
             metadata_json=self._safe_json(metadata) or {},

backend/packages/harness/deerflow/runtime/journal.py

@@ -20,12 +20,13 @@ from __future__ import annotations
 import asyncio
 import logging
 import time
+from collections.abc import Mapping
 from datetime import UTC, datetime
 from typing import TYPE_CHECKING, Any, cast
 from uuid import UUID
 
 from langchain_core.callbacks import BaseCallbackHandler
-from langchain_core.messages import AnyMessage, BaseMessage, HumanMessage, ToolMessage
+from langchain_core.messages import AIMessage, AnyMessage, BaseMessage, HumanMessage, ToolMessage
 from langgraph.types import Command
 
 if TYPE_CHECKING:
@@ -71,6 +72,7 @@ class RunJournal(BaseCallbackHandler):
         # Dedup: LangChain may fire on_llm_end multiple times for the same run_id
         self._counted_llm_run_ids: set[str] = set()
         self._counted_external_source_ids: set[str] = set()
+        self._counted_message_llm_run_ids: set[str] = set()
 
         # Convenience fields
         self._last_ai_msg: str | None = None
@@ -86,6 +88,50 @@ class RunJournal(BaseCallbackHandler):
 
     # -- Lifecycle callbacks --
 
+    @staticmethod
+    def _message_text(message: BaseMessage) -> str:
+        """Extract displayable text from a message's mixed content shape."""
+        content = getattr(message, "content", None)
+        if isinstance(content, str):
+            return content
+        if isinstance(content, list):
+            parts: list[str] = []
+            for block in content:
+                if isinstance(block, str):
+                    parts.append(block)
+                elif isinstance(block, Mapping):
+                    text = block.get("text")
+                    if isinstance(text, str):
+                        parts.append(text)
+                    else:
+                        nested = block.get("content")
+                        if isinstance(nested, str):
+                            parts.append(nested)
+            return "".join(parts)
+        if isinstance(content, Mapping):
+            for key in ("text", "content"):
+                value = content.get(key)
+                if isinstance(value, str):
+                    return value
+
+        text = getattr(message, "text", None)
+        if isinstance(text, str):
+            return text
+        return ""
+
+    def _record_message_summary(self, message: BaseMessage, *, caller: str | None = None) -> None:
+        """Update run-level convenience fields for persisted run rows."""
+        self._msg_count += 1
+
+        # ``last_ai_message`` should represent the lead agent's user-facing
+        # answer. Middleware/subagent model calls and empty tool-call-only
+        # AI messages must not overwrite the last useful assistant text.
+        is_ai_message = isinstance(message, AIMessage) or getattr(message, "type", None) == "ai"
+        if is_ai_message and (caller is None or caller == "lead_agent"):
+            text = self._message_text(message).strip()
+            if text:
+                self._last_ai_msg = text[:2000]
+
     def on_chain_start(
         self,
         serialized: dict[str, Any],
@@ -164,6 +210,7 @@ class RunJournal(BaseCallbackHandler):
                             content=m.model_dump(),
                             metadata={"caller": caller},
                         )
+                        self._record_message_summary(m, caller=caller)
                         break
                 if self._first_human_msg:
                     break
@@ -222,6 +269,8 @@ class RunJournal(BaseCallbackHandler):
                     "llm_call_index": call_index,
                 },
             )
+            if rid not in self._counted_message_llm_run_ids:
+                self._record_message_summary(message, caller=caller)
 
             # Token accumulation (dedup by langchain run_id to avoid double-counting
             # when the callback fires more than once for the same response)
@@ -245,6 +294,9 @@ class RunJournal(BaseCallbackHandler):
                     else:
                         self._lead_agent_tokens += total_tk
 
+        if messages:
+            self._counted_message_llm_run_ids.add(str(run_id))
+
     def on_llm_error(self, error: BaseException, *, run_id: UUID, **kwargs: Any) -> None:
         self._llm_start_times.pop(str(run_id), None)
         self._put(event_type="llm.error", category="trace", content=str(error))
@@ -260,12 +312,14 @@ class RunJournal(BaseCallbackHandler):
             if isinstance(output, ToolMessage):
                 msg = cast(ToolMessage, output)
                 self._put(event_type="llm.tool.result", category="message", content=msg.model_dump())
+                self._record_message_summary(msg)
             elif isinstance(output, Command):
                 cmd = cast(Command, output)
                 messages = cmd.update.get("messages", [])
                 for message in messages:
                     if isinstance(message, BaseMessage):
                         self._put(event_type="llm.tool.result", category="message", content=message.model_dump())
+                        self._record_message_summary(message)
                     else:
                         logger.warning(f"on_tool_end {run_id}: command update message is not BaseMessage: {type(message)}")
             else:

backend/packages/harness/deerflow/runtime/runs/manager.py

@@ -36,6 +36,7 @@ class RunRecord:
     abort_event: asyncio.Event = field(default_factory=asyncio.Event, repr=False)
     abort_action: str = "interrupt"
     error: str | None = None
+    model_name: str | None = None
 
 
 class RunManager:
@@ -65,6 +66,7 @@ class RunManager:
                 metadata=record.metadata or {},
                 kwargs=record.kwargs or {},
                 created_at=record.created_at,
+                model_name=record.model_name,
             )
         except Exception:
             logger.warning("Failed to persist run %s to store", record.run_id, exc_info=True)
@@ -137,6 +139,18 @@ class RunManager:
                 logger.warning("Failed to persist status update for run %s", run_id, exc_info=True)
         logger.info("Run %s -> %s", run_id, status.value)
 
+    async def update_model_name(self, run_id: str, model_name: str | None) -> None:
+        """Update the model name for a run."""
+        async with self._lock:
+            record = self._runs.get(run_id)
+            if record is None:
+                logger.warning("update_model_name called for unknown run %s", run_id)
+                return
+            record.model_name = model_name
+            record.updated_at = _now_iso()
+        await self._persist_to_store(record)
+        logger.info("Run %s model_name=%s", run_id, model_name)
+
     async def cancel(self, run_id: str, *, action: str = "interrupt") -> bool:
         """Request cancellation of a run.
 
@@ -171,6 +185,7 @@ class RunManager:
         metadata: dict | None = None,
         kwargs: dict | None = None,
         multitask_strategy: str = "reject",
+        model_name: str | None = None,
     ) -> RunRecord:
         """Atomically check for inflight runs and create a new one.
 
@@ -221,6 +236,7 @@ class RunManager:
                 kwargs=kwargs or {},
                 created_at=now,
                 updated_at=now,
+                model_name=model_name,
             )
             self._runs[run_id] = record
 

backend/packages/harness/deerflow/runtime/runs/store/base.py

@@ -23,6 +23,7 @@ class RunStore(abc.ABC):
         thread_id: str,
         assistant_id: str | None = None,
         user_id: str | None = None,
+        model_name: str | None = None,
         status: str = "pending",
         multitask_strategy: str = "reject",
         metadata: dict[str, Any] | None = None,

backend/packages/harness/deerflow/runtime/runs/store/memory.py

@@ -22,6 +22,7 @@ class MemoryRunStore(RunStore):
         thread_id,
         assistant_id=None,
         user_id=None,
+        model_name=None,
         status="pending",
         multitask_strategy="reject",
         metadata=None,
@@ -35,6 +36,7 @@ class MemoryRunStore(RunStore):
             "thread_id": thread_id,
             "assistant_id": assistant_id,
             "user_id": user_id,
+            "model_name": model_name,
             "status": status,
             "multitask_strategy": multitask_strategy,
             "metadata": metadata or {},

backend/packages/harness/deerflow/runtime/runs/worker.py

@@ -230,6 +230,17 @@ async def run_agent(
         else:
             agent = agent_factory(config=runnable_config)
 
+        # Capture the effective (resolved) model name from the agent's metadata.
+        # _resolve_model_name in agent.py may return the default model if the
+        # requested name is not in the allowlist — this update ensures the
+        # persisted model_name reflects the actual model used.
+        if record.model_name is not None:
+            resolved = getattr(agent, "metadata", {}) or {}
+            if isinstance(resolved, dict):
+                effective = resolved.get("model_name")
+                if effective and effective != record.model_name:
+                    await run_manager.update_model_name(record.run_id, effective)
+
         # 4. Attach checkpointer and store
         if checkpointer is not None:
             agent.checkpointer = checkpointer

backend/packages/harness/deerflow/subagents/config.py

@@ -26,7 +26,7 @@ class SubagentConfig:
 
     name: str
     description: str
-    system_prompt: str
+    system_prompt: str | None = None
     tools: list[str] | None = None
     disallowed_tools: list[str] | None = field(default_factory=lambda: ["task"])
     skills: list[str] | None = None

backend/packages/harness/deerflow/subagents/executor.py

@@ -286,11 +286,13 @@ class SubagentExecutor:
         # Reuse shared middleware composition with lead agent.
         middlewares = build_subagent_runtime_middlewares(app_config=app_config, model_name=self.model_name, lazy_init=True)
 
+        # system_prompt is included in initial state messages (see _build_initial_state)
+        # to avoid multiple SystemMessages which some LLM APIs don't support.
         return create_agent(
             model=model,
             tools=tools if tools is not None else self.tools,
             middleware=middlewares,
-            system_prompt=self.config.system_prompt,
+            system_prompt=None,
             state_schema=ThreadState,
         )
 
@@ -365,14 +367,25 @@ class SubagentExecutor:
         Returns:
             Initial state dictionary and tools filtered by loaded skill metadata.
         """
+
         # Load skills as conversation items (Codex pattern)
         skills = await self._load_skills()
         filtered_tools = self._apply_skill_allowed_tools(skills)
         skill_messages = await self._load_skill_messages(skills)
 
+        # Combine system_prompt and skills into a single SystemMessage.
+        # Some LLM APIs reject multiple SystemMessages with
+        # "System message must be at the beginning."
+        system_parts: list[str] = []
+        if self.config.system_prompt:
+            system_parts.append(self.config.system_prompt)
+        for skill_msg in skill_messages:
+            system_parts.append(skill_msg.content)
+
         messages: list[Any] = []
-        # Skill content injected as developer/system messages before the task
-        messages.extend(skill_messages)
+        if system_parts:
+            messages.append(SystemMessage(content="\n\n".join(system_parts)))
+
         # Then the actual task
         messages.append(HumanMessage(content=task))
 

backend/packages/harness/deerflow/tools/skill_manage_tool.py

@@ -10,11 +10,11 @@ from weakref import WeakValueDictionary
 from langchain.tools import tool
 
 from deerflow.agents.lead_agent.prompt import refresh_skills_system_prompt_cache_async
-from deerflow.mcp.tools import _make_sync_tool_wrapper
 from deerflow.skills.security_scanner import scan_skill_content
 from deerflow.skills.storage import get_or_new_skill_storage
 from deerflow.skills.storage.skill_storage import SkillStorage
 from deerflow.skills.types import SKILL_MD_FILE
+from deerflow.tools.sync import make_sync_tool_wrapper
 from deerflow.tools.types import Runtime
 
 logger = logging.getLogger(__name__)
@@ -235,4 +235,4 @@ async def skill_manage_tool(
     )
 
 
-skill_manage_tool.func = _make_sync_tool_wrapper(_skill_manage_impl, "skill_manage")
+skill_manage_tool.func = make_sync_tool_wrapper(_skill_manage_impl, "skill_manage")

backend/packages/harness/deerflow/tools/sync.py

@@ -0,0 +1,36 @@
+"""Utilities for invoking async tools from synchronous agent paths."""
+
+import asyncio
+import atexit
+import concurrent.futures
+import logging
+from collections.abc import Callable
+from typing import Any
+
+logger = logging.getLogger(__name__)
+
+# Shared thread pool for sync tool invocation in async environments.
+_SYNC_TOOL_EXECUTOR = concurrent.futures.ThreadPoolExecutor(max_workers=10, thread_name_prefix="tool-sync")
+
+atexit.register(lambda: _SYNC_TOOL_EXECUTOR.shutdown(wait=False))
+
+
+def make_sync_tool_wrapper(coro: Callable[..., Any], tool_name: str) -> Callable[..., Any]:
+    """Build a synchronous wrapper for an asynchronous tool coroutine."""
+
+    def sync_wrapper(*args: Any, **kwargs: Any) -> Any:
+        try:
+            loop = asyncio.get_running_loop()
+        except RuntimeError:
+            loop = None
+
+        try:
+            if loop is not None and loop.is_running():
+                future = _SYNC_TOOL_EXECUTOR.submit(asyncio.run, coro(*args, **kwargs))
+                return future.result()
+            return asyncio.run(coro(*args, **kwargs))
+        except Exception as e:
+            logger.error("Error invoking tool %r via sync wrapper: %s", tool_name, e, exc_info=True)
+            raise
+
+    return sync_wrapper

backend/packages/harness/deerflow/tools/tools.py

@@ -8,6 +8,7 @@ from deerflow.reflection import resolve_variable
 from deerflow.sandbox.security import is_host_bash_allowed
 from deerflow.tools.builtins import ask_clarification_tool, present_file_tool, task_tool, view_image_tool
 from deerflow.tools.builtins.tool_search import reset_deferred_registry
+from deerflow.tools.sync import make_sync_tool_wrapper
 
 logger = logging.getLogger(__name__)
 
@@ -33,6 +34,13 @@ def _is_host_bash_tool(tool: object) -> bool:
     return False
 
 
+def _ensure_sync_invocable_tool(tool: BaseTool) -> BaseTool:
+    """Attach a sync wrapper to async-only tools used by sync agent callers."""
+    if getattr(tool, "func", None) is None and getattr(tool, "coroutine", None) is not None:
+        tool.func = make_sync_tool_wrapper(tool.coroutine, tool.name)
+    return tool
+
+
 def get_available_tools(
     groups: list[str] | None = None,
     include_mcp: bool = True,
@@ -77,7 +85,7 @@ def get_available_tools(
                 cfg.use,
             )
 
-    loaded_tools = [t for _, t in loaded_tools_raw]
+    loaded_tools = [_ensure_sync_invocable_tool(t) for _, t in loaded_tools_raw]
 
     # Conditionally add tools based on config
     builtin_tools = BUILTIN_TOOLS.copy()

backend/tests/test_dangling_tool_call_middleware.py

@@ -14,6 +14,10 @@ def _ai_with_tool_calls(tool_calls):
     return AIMessage(content="", tool_calls=tool_calls)
 
 
+def _ai_with_invalid_tool_calls(invalid_tool_calls):
+    return AIMessage(content="", tool_calls=[], invalid_tool_calls=invalid_tool_calls)
+
+
 def _tool_msg(tool_call_id, name="test_tool"):
     return ToolMessage(content="result", tool_call_id=tool_call_id, name=name)
 
@@ -22,6 +26,16 @@ def _tc(name="bash", tc_id="call_1"):
     return {"name": name, "id": tc_id, "args": {}}
 
 
+def _invalid_tc(name="write_file", tc_id="write_file:36", error="Failed to parse tool arguments: malformed JSON"):
+    return {
+        "type": "invalid_tool_call",
+        "name": name,
+        "id": tc_id,
+        "args": '{"description":"write report","path":"/mnt/user-data/outputs/report.md","content":"bad {"json"}"}',
+        "error": error,
+    }
+
+
 class TestBuildPatchedMessagesNoPatch:
     def test_empty_messages(self):
         mw = DanglingToolCallMiddleware()
@@ -144,6 +158,42 @@ class TestBuildPatchedMessagesPatching:
         assert patched[1].name == "bash"
         assert patched[1].status == "error"
 
+    def test_invalid_tool_call_is_patched(self):
+        mw = DanglingToolCallMiddleware()
+        msgs = [_ai_with_invalid_tool_calls([_invalid_tc()])]
+        patched = mw._build_patched_messages(msgs)
+        assert patched is not None
+        assert len(patched) == 2
+        assert isinstance(patched[1], ToolMessage)
+        assert patched[1].tool_call_id == "write_file:36"
+        assert patched[1].name == "write_file"
+        assert patched[1].status == "error"
+        assert "arguments were invalid" in patched[1].content
+        assert "Failed to parse tool arguments" in patched[1].content
+
+    def test_valid_and_invalid_tool_calls_are_both_patched(self):
+        mw = DanglingToolCallMiddleware()
+        msgs = [
+            AIMessage(
+                content="",
+                tool_calls=[_tc("bash", "call_1")],
+                invalid_tool_calls=[_invalid_tc()],
+            )
+        ]
+        patched = mw._build_patched_messages(msgs)
+        assert patched is not None
+        tool_msgs = [m for m in patched if isinstance(m, ToolMessage)]
+        assert len(tool_msgs) == 2
+        assert {tm.tool_call_id for tm in tool_msgs} == {"call_1", "write_file:36"}
+
+    def test_invalid_tool_call_already_responded_is_not_patched(self):
+        mw = DanglingToolCallMiddleware()
+        msgs = [
+            _ai_with_invalid_tool_calls([_invalid_tc()]),
+            _tool_msg("write_file:36", "write_file"),
+        ]
+        assert mw._build_patched_messages(msgs) is None
+
 
 class TestWrapModelCall:
     def test_no_patch_passthrough(self):

backend/tests/test_gateway_docs_toggle.py

@@ -122,3 +122,45 @@ def test_health_still_works_when_docs_disabled():
         resp = client.get("/health")
         assert resp.status_code == 200
         assert resp.json()["status"] == "healthy"
+
+
+# ---------------------------------------------------------------------------
+# Runtime CORS behavior
+# ---------------------------------------------------------------------------
+
+
+def _make_gateway_client(cors_origins: str) -> TestClient:
+    with patch.dict(os.environ, {"GATEWAY_CORS_ORIGINS": cors_origins}):
+        _reset_gateway_config()
+        from app.gateway.app import create_app
+
+        return TestClient(create_app())
+
+
+def test_gateway_cors_allows_configured_origin():
+    """GATEWAY_CORS_ORIGINS should control actual browser CORS responses."""
+    client = _make_gateway_client("https://app.example")
+
+    response = client.get("/health", headers={"Origin": "https://app.example"})
+
+    assert response.status_code == 200
+    assert response.headers["access-control-allow-origin"] == "https://app.example"
+    assert response.headers["access-control-allow-credentials"] == "true"
+
+
+def test_gateway_cors_rejects_unconfigured_origin():
+    client = _make_gateway_client("https://app.example")
+
+    response = client.get("/health", headers={"Origin": "https://evil.example"})
+
+    assert response.status_code == 200
+    assert "access-control-allow-origin" not in response.headers
+
+
+def test_gateway_cors_normalizes_configured_default_port():
+    client = _make_gateway_client("https://app.example:443")
+
+    response = client.get("/health", headers={"Origin": "https://app.example"})
+
+    assert response.status_code == 200
+    assert response.headers["access-control-allow-origin"] == "https://app.example"

backend/tests/test_gateway_runtime_cleanup.py

@@ -53,6 +53,29 @@ def test_nginx_routes_official_langgraph_prefix_to_gateway_api():
         assert "proxy_pass http://gateway" in content or "proxy_pass http://$gateway_upstream" in content
 
 
+def test_nginx_defers_cors_to_gateway_allowlist():
+    for path in ("docker/nginx/nginx.local.conf", "docker/nginx/nginx.conf"):
+        content = _read(path)
+
+        assert "Access-Control-Allow-Origin" not in content
+        assert "Access-Control-Allow-Methods" not in content
+        assert "Access-Control-Allow-Headers" not in content
+        assert "Access-Control-Allow-Credentials" not in content
+        assert "proxy_hide_header 'Access-Control-Allow-" not in content
+        assert "if ($request_method = 'OPTIONS')" not in content
+
+
+def test_gateway_cors_configuration_uses_gateway_allowlist():
+    gateway_config = _read("backend/app/gateway/config.py")
+    gateway_app = _read("backend/app/gateway/app.py")
+    csrf_middleware = _read("backend/app/gateway/csrf_middleware.py")
+
+    assert not re.search(r"(?<!GATEWAY_)[\"']CORS_ORIGINS[\"']", gateway_config)
+    assert "cors_origins" not in gateway_config
+    assert "get_configured_cors_origins" in gateway_app
+    assert "GATEWAY_CORS_ORIGINS" in csrf_middleware
+
+
 def test_frontend_rewrites_langgraph_prefix_to_gateway():
     next_config = _read("frontend/next.config.js")
     api_client = _read("frontend/src/core/api/api-client.ts")

backend/tests/test_mcp_sync_wrapper.py

@@ -5,7 +5,8 @@ import pytest
 from langchain_core.tools import StructuredTool
 from pydantic import BaseModel, Field
 
-from deerflow.mcp.tools import _make_sync_tool_wrapper, get_mcp_tools
+from deerflow.mcp.tools import get_mcp_tools
+from deerflow.tools.sync import make_sync_tool_wrapper
 
 
 class MockArgs(BaseModel):
@@ -51,14 +52,13 @@ def test_mcp_tool_sync_wrapper_generation():
 
 
 def test_mcp_tool_sync_wrapper_in_running_loop():
-    """Test the actual helper function from production code (Fix for Comment 1 & 3)."""
+    """Test the shared sync wrapper from production code."""
 
     async def mock_coro(x: int):
         await asyncio.sleep(0.01)
         return f"async_result: {x}"
 
-    # Test the real helper function exported from deerflow.mcp.tools
-    sync_func = _make_sync_tool_wrapper(mock_coro, "test_tool")
+    sync_func = make_sync_tool_wrapper(mock_coro, "test_tool")
 
     async def run_in_loop():
         # This call should succeed due to ThreadPoolExecutor in the real helper
@@ -70,16 +70,16 @@ def test_mcp_tool_sync_wrapper_in_running_loop():
 
 
 def test_mcp_tool_sync_wrapper_exception_logging():
-    """Test the actual helper's error logging (Fix for Comment 3)."""
+    """Test the shared sync wrapper's error logging."""
 
     async def error_coro():
         raise ValueError("Tool failure")
 
-    sync_func = _make_sync_tool_wrapper(error_coro, "error_tool")
+    sync_func = make_sync_tool_wrapper(error_coro, "error_tool")
 
-    with patch("deerflow.mcp.tools.logger.error") as mock_log_error:
+    with patch("deerflow.tools.sync.logger.error") as mock_log_error:
         with pytest.raises(ValueError, match="Tool failure"):
             sync_func()
         mock_log_error.assert_called_once()
         # Verify the tool name is in the log message
-        assert "error_tool" in mock_log_error.call_args[0][0]
+        assert mock_log_error.call_args[0][1] == "error_tool"

backend/tests/test_run_journal.py

@@ -339,6 +339,99 @@ class TestConvenienceFields:
         data = j.get_completion_data()
         assert data["first_human_message"] == "What is AI?"
 
+    @pytest.mark.anyio
+    async def test_completion_data_counts_human_ai_and_tool_messages(self, journal_setup):
+        from langchain_core.messages import HumanMessage, ToolMessage
+
+        j, _ = journal_setup
+        j.on_chat_model_start({}, [[HumanMessage(content="Question")]], run_id=uuid4(), tags=["lead_agent"])
+        j.on_llm_end(_make_llm_response("Answer"), run_id=uuid4(), parent_run_id=None, tags=["lead_agent"])
+        j.on_tool_end(ToolMessage(content="Tool result", tool_call_id="call_1", name="search"), run_id=uuid4())
+
+        data = j.get_completion_data()
+
+        assert data["message_count"] == 3
+        assert data["first_human_message"] == "Question"
+        assert data["last_ai_message"] == "Answer"
+
+    @pytest.mark.anyio
+    async def test_tool_call_only_ai_does_not_clear_last_ai_message(self, journal_setup):
+        j, _ = journal_setup
+        j.on_llm_end(_make_llm_response("Useful answer"), run_id=uuid4(), parent_run_id=None, tags=["lead_agent"])
+        j.on_llm_end(
+            _make_llm_response("", tool_calls=[{"id": "call_1", "name": "search", "args": {}}]),
+            run_id=uuid4(),
+            parent_run_id=None,
+            tags=["lead_agent"],
+        )
+
+        data = j.get_completion_data()
+
+        assert data["message_count"] == 2
+        assert data["last_ai_message"] == "Useful answer"
+
+    @pytest.mark.anyio
+    async def test_last_ai_message_extracts_mixed_content_without_extra_newlines(self, journal_setup):
+        j, _ = journal_setup
+        j.on_llm_end(
+            _make_llm_response(
+                [
+                    {"type": "text", "text": "First "},
+                    {"type": "text", "content": "second"},
+                    " third",
+                    {"type": "image", "url": "ignored"},
+                ]
+            ),
+            run_id=uuid4(),
+            parent_run_id=None,
+            tags=["lead_agent"],
+        )
+
+        data = j.get_completion_data()
+
+        assert data["message_count"] == 1
+        assert data["last_ai_message"] == "First second third"
+
+    @pytest.mark.anyio
+    async def test_last_ai_message_extracts_mapping_content(self, journal_setup):
+        j, _ = journal_setup
+        j.on_llm_end(_make_llm_response({"content": "Nested answer"}), run_id=uuid4(), parent_run_id=None, tags=["lead_agent"])
+
+        data = j.get_completion_data()
+
+        assert data["message_count"] == 1
+        assert data["last_ai_message"] == "Nested answer"
+
+    @pytest.mark.anyio
+    async def test_duplicate_llm_run_id_does_not_double_count_message_summary(self, journal_setup):
+        j, _ = journal_setup
+        run_id = uuid4()
+
+        j.on_llm_end(_make_llm_response("Answer", usage=None), run_id=run_id, parent_run_id=None, tags=["lead_agent"])
+        j.on_llm_end(
+            _make_llm_response("Answer", usage={"input_tokens": 10, "output_tokens": 5, "total_tokens": 15}),
+            run_id=run_id,
+            parent_run_id=None,
+            tags=["lead_agent"],
+        )
+
+        data = j.get_completion_data()
+
+        assert data["message_count"] == 1
+        assert data["last_ai_message"] == "Answer"
+        assert data["total_tokens"] == 15
+
+    @pytest.mark.anyio
+    async def test_subagent_ai_does_not_overwrite_lead_last_ai_message(self, journal_setup):
+        j, _ = journal_setup
+        j.on_llm_end(_make_llm_response("Lead answer"), run_id=uuid4(), parent_run_id=None, tags=["lead_agent"])
+        j.on_llm_end(_make_llm_response("Subagent detail"), run_id=uuid4(), parent_run_id=None, tags=["subagent:research"])
+
+        data = j.get_completion_data()
+
+        assert data["message_count"] == 2
+        assert data["last_ai_message"] == "Lead answer"
+
     @pytest.mark.anyio
     async def test_get_completion_data(self, journal_setup):
         j, _ = journal_setup

backend/tests/test_run_manager.py

@@ -5,6 +5,7 @@ import re
 import pytest
 
 from deerflow.runtime import RunManager, RunStatus
+from deerflow.runtime.runs.store.memory import MemoryRunStore
 
 ISO_RE = re.compile(r"^\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}")
 
@@ -141,3 +142,53 @@ async def test_create_defaults(manager: RunManager):
     assert record.kwargs == {}
     assert record.multitask_strategy == "reject"
     assert record.assistant_id is None
+
+
+@pytest.mark.anyio
+async def test_model_name_create_or_reject():
+    """create_or_reject should accept and persist model_name."""
+    from deerflow.runtime.runs.schemas import DisconnectMode
+
+    store = MemoryRunStore()
+    mgr = RunManager(store=store)
+
+    record = await mgr.create_or_reject(
+        "thread-1",
+        assistant_id="lead_agent",
+        on_disconnect=DisconnectMode.cancel,
+        metadata={"key": "val"},
+        kwargs={"input": {}},
+        multitask_strategy="reject",
+        model_name="anthropic.claude-sonnet-4-20250514-v1:0",
+    )
+    assert record.model_name == "anthropic.claude-sonnet-4-20250514-v1:0"
+    assert record.status == RunStatus.pending
+
+    # Verify model_name was persisted to store
+    stored = await store.get(record.run_id)
+    assert stored is not None
+    assert stored["model_name"] == "anthropic.claude-sonnet-4-20250514-v1:0"
+
+    # Verify retrieval returns the model_name via in-memory record
+    fetched = mgr.get(record.run_id)
+    assert fetched is not None
+    assert fetched.model_name == "anthropic.claude-sonnet-4-20250514-v1:0"
+
+
+@pytest.mark.anyio
+async def test_model_name_default_is_none():
+    """create_or_reject without model_name should default to None."""
+    from deerflow.runtime.runs.schemas import DisconnectMode
+
+    store = MemoryRunStore()
+    mgr = RunManager(store=store)
+
+    record = await mgr.create_or_reject(
+        "thread-1",
+        on_disconnect=DisconnectMode.cancel,
+        model_name=None,
+    )
+    assert record.model_name is None
+
+    stored = await store.get(record.run_id)
+    assert stored["model_name"] is None

backend/tests/test_run_repository.py

@@ -249,3 +249,32 @@ class TestRunRepository:
         rows = await repo.list_by_thread("t1", user_id=None)
         assert len(rows) == 2
         await _cleanup()
+
+    @pytest.mark.anyio
+    async def test_model_name_persistence(self, tmp_path):
+        """RunRepository should persist, normalize, and truncate model_name correctly via SQL."""
+        from deerflow.persistence.engine import get_session_factory, init_engine
+
+        url = f"sqlite+aiosqlite:///{tmp_path / 'test.db'}"
+        await init_engine("sqlite", url=url, sqlite_dir=str(tmp_path))
+        repo = RunRepository(get_session_factory())
+
+        await repo.put("run-1", thread_id="thread-1", model_name="gpt-4o")
+        row = await repo.get("run-1")
+        assert row is not None
+        assert row["model_name"] == "gpt-4o"
+
+        long_name = "a" * 200
+        await repo.put("run-2", thread_id="thread-1", model_name=long_name)
+        row2 = await repo.get("run-2")
+        assert row2["model_name"] == "a" * 128
+
+        await repo.put("run-3", thread_id="thread-1", model_name=123)
+        row3 = await repo.get("run-3")
+        assert row3["model_name"] == "123"
+
+        await repo.put("run-4", thread_id="thread-1", model_name=None)
+        row4 = await repo.get("run-4")
+        assert row4["model_name"] is None
+
+        await _cleanup()

backend/tests/test_subagent_executor.py

@@ -291,7 +291,7 @@ class TestAgentConstruction:
         assert captured["agent"]["model"] is model
         assert captured["agent"]["middleware"] is middlewares
         assert captured["agent"]["tools"] == []
-        assert captured["agent"]["system_prompt"] == base_config.system_prompt
+        assert captured["agent"]["system_prompt"] is None  # system_prompt is merged into initial state messages
 
     @pytest.mark.anyio
     async def test_load_skill_messages_uses_explicit_app_config_for_skill_storage(
@@ -331,6 +331,124 @@ class TestAgentConstruction:
         assert len(messages) == 1
         assert "Use demo skill" in messages[0].content
 
+    @pytest.mark.anyio
+    async def test_build_initial_state_consolidates_system_prompt_and_skills(
+        self,
+        classes,
+        base_config,
+        monkeypatch: pytest.MonkeyPatch,
+        tmp_path,
+    ):
+        """_build_initial_state merges system_prompt and skills into one SystemMessage."""
+        SubagentExecutor = classes["SubagentExecutor"]
+
+        skill_dir = tmp_path / "my-skill"
+        skill_dir.mkdir()
+        skill_file = skill_dir / "SKILL.md"
+        skill_file.write_text("Skill instructions here", encoding="utf-8")
+
+        monkeypatch.setattr(
+            sys.modules["deerflow.skills.storage"],
+            "get_or_new_skill_storage",
+            lambda *, app_config=None: SimpleNamespace(load_skills=lambda *, enabled_only: [SimpleNamespace(name="my-skill", skill_file=skill_file, allowed_tools=None)]),
+        )
+
+        executor = SubagentExecutor(
+            config=base_config,
+            tools=[],
+            thread_id="test-thread",
+        )
+
+        state, _filtered_tools = await executor._build_initial_state("Do the task")
+
+        messages = state["messages"]
+        # Should have exactly 2 messages: one combined SystemMessage + one HumanMessage
+        assert len(messages) == 2
+
+        from langchain_core.messages import HumanMessage, SystemMessage
+
+        assert isinstance(messages[0], SystemMessage)
+        assert isinstance(messages[1], HumanMessage)
+        # SystemMessage should contain both the system_prompt and skill content
+        assert base_config.system_prompt in messages[0].content
+        assert "Skill instructions here" in messages[0].content
+        # HumanMessage should be the task
+        assert messages[1].content == "Do the task"
+
+    @pytest.mark.anyio
+    async def test_build_initial_state_no_skills_only_system_prompt(
+        self,
+        classes,
+        base_config,
+        monkeypatch: pytest.MonkeyPatch,
+    ):
+        """_build_initial_state works when there are no skills."""
+        SubagentExecutor = classes["SubagentExecutor"]
+
+        monkeypatch.setattr(
+            sys.modules["deerflow.skills.storage"],
+            "get_or_new_skill_storage",
+            lambda *, app_config=None: SimpleNamespace(load_skills=lambda *, enabled_only: []),
+        )
+
+        executor = SubagentExecutor(
+            config=base_config,
+            tools=[],
+            thread_id="test-thread",
+        )
+
+        state, _filtered_tools = await executor._build_initial_state("Do the task")
+
+        messages = state["messages"]
+        from langchain_core.messages import HumanMessage, SystemMessage
+
+        assert len(messages) == 2
+        assert isinstance(messages[0], SystemMessage)
+        assert base_config.system_prompt in messages[0].content
+        assert isinstance(messages[1], HumanMessage)
+
+    @pytest.mark.anyio
+    async def test_build_initial_state_no_system_prompt_with_skills(
+        self,
+        classes,
+        monkeypatch: pytest.MonkeyPatch,
+        tmp_path,
+    ):
+        """_build_initial_state works when there is no system_prompt but there are skills."""
+        SubagentConfig = classes["SubagentConfig"]
+
+        config = SubagentConfig(
+            name="test-agent",
+            description="Test agent",
+            system_prompt=None,
+            max_turns=10,
+            timeout_seconds=60,
+        )
+
+        skill_dir = tmp_path / "my-skill"
+        skill_dir.mkdir()
+        skill_file = skill_dir / "SKILL.md"
+        skill_file.write_text("Skill content", encoding="utf-8")
+
+        monkeypatch.setattr(
+            sys.modules["deerflow.skills.storage"],
+            "get_or_new_skill_storage",
+            lambda *, app_config=None: SimpleNamespace(load_skills=lambda *, enabled_only: [SimpleNamespace(name="my-skill", skill_file=skill_file, allowed_tools=None)]),
+        )
+
+        SubagentExecutor = classes["SubagentExecutor"]
+        executor = SubagentExecutor(config=config, tools=[], thread_id="test-thread")
+
+        state, _filtered_tools = await executor._build_initial_state("Do the task")
+
+        messages = state["messages"]
+        from langchain_core.messages import HumanMessage, SystemMessage
+
+        assert len(messages) == 2
+        assert isinstance(messages[0], SystemMessage)
+        assert "Skill content" in messages[0].content
+        assert isinstance(messages[1], HumanMessage)
+
 
 # -----------------------------------------------------------------------------
 # Async Execution Path Tests
@@ -514,6 +632,70 @@ class TestAsyncExecutionPath:
         assert result.status == SubagentStatus.COMPLETED
         assert "Task" in result.result
 
+    @pytest.mark.anyio
+    async def test_aexecute_passes_at_most_one_system_message_to_agent(
+        self,
+        classes,
+        base_config,
+        monkeypatch: pytest.MonkeyPatch,
+        tmp_path,
+    ):
+        """Regression: messages sent to agent.astream must contain at most one
+        SystemMessage and it must be the first message.
+
+        This catches any regression where system_prompt would be re-injected
+        via create_agent() (e.g. system_prompt not passed as None) and appear
+        as a second SystemMessage, which providers like vLLM and Xinference
+        reject with "System message must be at the beginning."
+        """
+        from langchain_core.messages import AIMessage, SystemMessage
+
+        SubagentExecutor = classes["SubagentExecutor"]
+        SubagentStatus = classes["SubagentStatus"]
+
+        # Set up a skill so both system_prompt AND skill content are present,
+        # maximising the chance of catching a double-SystemMessage regression.
+        skill_dir = tmp_path / "regression-skill"
+        skill_dir.mkdir()
+        (skill_dir / "SKILL.md").write_text("Skill instruction text", encoding="utf-8")
+
+        monkeypatch.setattr(
+            sys.modules["deerflow.skills.storage"],
+            "get_or_new_skill_storage",
+            lambda *, app_config=None: SimpleNamespace(load_skills=lambda *, enabled_only: [SimpleNamespace(name="regression-skill", skill_file=skill_dir / "SKILL.md", allowed_tools=None)]),
+        )
+
+        captured_states: list[dict] = []
+
+        async def capturing_astream(state, **kwargs):
+            captured_states.append(state)
+            yield {"messages": [AIMessage(content="Done", id="msg-1")]}
+
+        mock_agent = MagicMock()
+        mock_agent.astream = capturing_astream
+
+        executor = SubagentExecutor(
+            config=base_config,
+            tools=[],
+            thread_id="test-thread",
+        )
+
+        with patch.object(executor, "_create_agent", return_value=mock_agent):
+            result = await executor._aexecute("Do something")
+
+        assert result.status == SubagentStatus.COMPLETED
+        assert len(captured_states) == 1, "astream should be called exactly once"
+        initial_messages = captured_states[0]["messages"]
+
+        system_messages = [m for m in initial_messages if isinstance(m, SystemMessage)]
+        assert len(system_messages) <= 1, f"Expected at most 1 SystemMessage but got {len(system_messages)}: {system_messages}"
+        if system_messages:
+            assert initial_messages[0] is system_messages[0], "SystemMessage must be the first message in the conversation"
+            # The consolidated SystemMessage must carry both the system_prompt
+            # and all skill content — nothing should be split across two messages.
+            assert base_config.system_prompt in system_messages[0].content
+            assert "Skill instruction text" in system_messages[0].content
+
 
 class TestSkillAllowedTools:
     @pytest.mark.anyio

backend/tests/test_tool_deduplication.py

@@ -10,7 +10,8 @@ from __future__ import annotations
 
 from unittest.mock import MagicMock, patch
 
-from langchain_core.tools import BaseTool, tool
+from langchain_core.tools import BaseTool, StructuredTool, tool
+from pydantic import BaseModel, Field
 
 from deerflow.tools.tools import get_available_tools
 
@@ -19,6 +20,10 @@ from deerflow.tools.tools import get_available_tools
 # ---------------------------------------------------------------------------
 
 
+class AsyncToolArgs(BaseModel):
+    x: int = Field(..., description="test input")
+
+
 @tool
 def _tool_alpha(x: str) -> str:
     """Alpha tool."""
@@ -52,10 +57,45 @@ def _make_minimal_config(tools):
     config.tools = tools
     config.models = []
     config.tool_search.enabled = False
+    config.skill_evolution.enabled = False
     config.sandbox = MagicMock()
+    config.acp_agents = {}
     return config
 
 
+@patch("deerflow.tools.tools.get_app_config")
+@patch("deerflow.tools.tools.is_host_bash_allowed", return_value=True)
+@patch("deerflow.tools.tools.reset_deferred_registry")
+def test_config_loaded_async_only_tool_gets_sync_wrapper(mock_reset, mock_bash, mock_cfg):
+    """Config-loaded async-only tools can still be invoked by sync clients."""
+
+    async def async_tool_impl(x: int) -> str:
+        return f"result: {x}"
+
+    async_tool = StructuredTool(
+        name="async_tool",
+        description="Async-only test tool.",
+        args_schema=AsyncToolArgs,
+        func=None,
+        coroutine=async_tool_impl,
+    )
+    tool_cfg = MagicMock()
+    tool_cfg.name = "async_tool"
+    tool_cfg.group = "test"
+    tool_cfg.use = "tests.fake:async_tool"
+    mock_cfg.return_value = _make_minimal_config([tool_cfg])
+
+    with (
+        patch("deerflow.tools.tools.resolve_variable", return_value=async_tool),
+        patch("deerflow.tools.tools.BUILTIN_TOOLS", []),
+    ):
+        result = get_available_tools(include_mcp=False, app_config=mock_cfg.return_value)
+
+    assert async_tool in result
+    assert async_tool.func is not None
+    assert async_tool.invoke({"x": 42}) == "result: 42"
+
+
 @patch("deerflow.tools.tools.get_app_config")
 @patch("deerflow.tools.tools.is_host_bash_allowed", return_value=True)
 @patch("deerflow.tools.tools.reset_deferred_registry")

backend/uv.lock

@@ -4224,11 +4224,11 @@ wheels = [
 
 [[package]]
 name = "urllib3"
-version = "2.6.3"
+version = "2.7.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/c7/24/5f1b3bdffd70275f6661c76461e25f024d5a38a46f04aaca912426a2b1d3/urllib3-2.6.3.tar.gz", hash = "sha256:1b62b6884944a57dbe321509ab94fd4d3b307075e0c2eae991ac71ee15ad38ed", size = 435556, upload-time = "2026-01-07T16:24:43.925Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/53/0c/06f8b233b8fd13b9e5ee11424ef85419ba0d8ba0b3138bf360be2ff56953/urllib3-2.7.0.tar.gz", hash = "sha256:231e0ec3b63ceb14667c67be60f2f2c40a518cb38b03af60abc813da26505f4c", size = 433602, upload-time = "2026-05-07T16:13:18.596Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/39/08/aaaad47bc4e9dc8c725e68f9d04865dbcb2052843ff09c97b08904852d84/urllib3-2.6.3-py3-none-any.whl", hash = "sha256:bf272323e553dfb2e87d9bfd225ca7b0f467b919d7bbd355436d3fd37cb0acd4", size = 131584, upload-time = "2026-01-07T16:24:42.685Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/3e/5db95bcf282c52709639744ca2a8b149baccf648e39c8cc87553df9eae0c/urllib3-2.7.0-py3-none-any.whl", hash = "sha256:9fb4c81ebbb1ce9531cce37674bbc6f1360472bc18ca9a553ede278ef7276897", size = 131087, upload-time = "2026-05-07T16:13:17.151Z" },
 ]
 
 [[package]]

docker/nginx/nginx.conf

@@ -28,21 +28,11 @@ http {
         set $gateway_upstream gateway:8001;
         set $frontend_upstream frontend:3000;
 
-        # Hide CORS headers from upstream to prevent duplicates
-        proxy_hide_header 'Access-Control-Allow-Origin';
-        proxy_hide_header 'Access-Control-Allow-Methods';
-        proxy_hide_header 'Access-Control-Allow-Headers';
-        proxy_hide_header 'Access-Control-Allow-Credentials';
-
-        # CORS headers for all responses (nginx handles CORS centrally)
-        add_header 'Access-Control-Allow-Origin' '*' always;
-        add_header 'Access-Control-Allow-Methods' 'GET, POST, PUT, DELETE, PATCH, OPTIONS' always;
-        add_header 'Access-Control-Allow-Headers' '*' always;
-
-        # Handle OPTIONS requests (CORS preflight)
-        if ($request_method = 'OPTIONS') {
-            return 204;
-        }
+        # Keep the unified nginx endpoint same-origin by default. When split
+        # frontend/backend or port-forwarded deployments need browser CORS,
+        # configure the Gateway allowlist with GATEWAY_CORS_ORIGINS so CORS and
+        # CSRF origin checks stay aligned instead of approving every origin at
+        # the proxy layer.
 
         # LangGraph-compatible API routes served by Gateway.
         # Rewrites /api/langgraph/* to /api/* before proxying to Gateway.

docker/nginx/nginx.local.conf

@@ -28,21 +28,11 @@ http {
         listen [::]:2026;
         server_name _;
 
-        # Hide CORS headers from upstream to prevent duplicates
-        proxy_hide_header 'Access-Control-Allow-Origin';
-        proxy_hide_header 'Access-Control-Allow-Methods';
-        proxy_hide_header 'Access-Control-Allow-Headers';
-        proxy_hide_header 'Access-Control-Allow-Credentials';
-
-        # CORS headers for all responses (nginx handles CORS centrally)
-        add_header 'Access-Control-Allow-Origin' '*' always;
-        add_header 'Access-Control-Allow-Methods' 'GET, POST, PUT, DELETE, PATCH, OPTIONS' always;
-        add_header 'Access-Control-Allow-Headers' '*' always;
-
-        # Handle OPTIONS requests (CORS preflight)
-        if ($request_method = 'OPTIONS') {
-            return 204;
-        }
+        # Keep the unified nginx endpoint same-origin by default. When split
+        # frontend/backend or port-forwarded deployments need browser CORS,
+        # configure the Gateway allowlist with GATEWAY_CORS_ORIGINS so CORS and
+        # CSRF origin checks stay aligned instead of approving every origin at
+        # the proxy layer.
 
         # LangGraph-compatible API routes served by Gateway.
         # Rewrites /api/langgraph/* to /api/* before proxying to Gateway.

frontend/package.json

@@ -68,7 +68,7 @@
     "lucide-react": "^0.562.0",
     "motion": "^12.26.2",
     "nanoid": "^5.1.6",
-    "next": "^16.1.7",
+    "next": "^16.2.6",
     "next-themes": "^0.4.6",
     "nextra": "^4.6.1",
     "nextra-theme-docs": "^4.6.1",

frontend/pnpm-lock.yaml

@@ -156,17 +156,17 @@ importers:
         specifier: ^5.1.6
         version: 5.1.6
       next:
-        specifier: ^16.1.7
-        version: 16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
+        specifier: ^16.2.6
+        version: 16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
       next-themes:
         specifier: ^0.4.6
         version: 0.4.6(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
       nextra:
         specifier: ^4.6.1
-        version: 4.6.1(next@16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3)
+        version: 4.6.1(next@16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3)
       nextra-theme-docs:
         specifier: ^4.6.1
-        version: 4.6.1(@types/react@19.2.13)(next@16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(nextra@4.6.1(next@16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(use-sync-external-store@1.6.0(react@19.2.4))
+        version: 4.6.1(@types/react@19.2.13)(next@16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(nextra@4.6.1(next@16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(use-sync-external-store@1.6.0(react@19.2.4))
       nuxt-og-image:
         specifier: ^5.1.13
         version: 5.1.13(@unhead/vue@2.1.4(vue@3.5.28(typescript@5.9.3)))(unstorage@1.17.4)(vite@7.3.1(@types/node@20.19.33)(jiti@2.6.1)(lightningcss@1.30.2)(yaml@2.8.3))(vue@3.5.28(typescript@5.9.3))
@@ -437,8 +437,8 @@ packages:
   '@emnapi/core@1.8.1':
     resolution: {integrity: sha512-AvT9QFpxK0Zd8J0jopedNm+w/2fIzvtPKPjqyw9jwvBaReTTqPBk9Hixaz7KbjimP+QNz605/XnjFcDAL2pqBg==}
 
-  '@emnapi/runtime@1.9.0':
-    resolution: {integrity: sha512-QN75eB0IH2ywSpRpNddCRfQIhmJYBCJ1x5Lb3IscKAL8bMnVAKnRg8dCoXbHzVLLH7P38N2Z3mtulB7W0J0FKw==}
+  '@emnapi/runtime@1.10.0':
+    resolution: {integrity: sha512-ewvYlk86xUoGI0zQRNq/mC+16R1QeDlKQy21Ki3oSYXNgLb45GV1P6A0M+/s6nyCuNDqe5VpaY84BzXGwVbwFA==}
 
   '@emnapi/wasi-threads@1.1.0':
     resolution: {integrity: sha512-WI0DdZ8xFSbgMjR1sFsKABJ/C5OnRrjT06JXbZKexJGrDuPTzZdDYfFlsgcCXCyf+suG5QU2e/y1Wo2V/OapLQ==}
@@ -1018,56 +1018,56 @@ packages:
   '@napi-rs/wasm-runtime@0.2.12':
     resolution: {integrity: sha512-ZVWUcfwY4E/yPitQJl481FjFo3K22D6qF0DuFH6Y/nbnE11GY5uguDxZMGXPQ8WQ0128MXQD7TnfHyK4oWoIJQ==}
 
-  '@next/env@16.1.7':
-    resolution: {integrity: sha512-rJJbIdJB/RQr2F1nylZr/PJzamvNNhfr3brdKP6s/GW850jbtR70QlSfFselvIBbcPUOlQwBakexjFzqLzF6pg==}
+  '@next/env@16.2.6':
+    resolution: {integrity: sha512-gd8HoHN4ufj73WmR3JmVolrpJR47ILK6LouP5xElPglaVxir6e1a7VzvTvDWkOoPXT9rkkTzyCxBu4yeZfZwcw==}
 
   '@next/eslint-plugin-next@15.5.12':
     resolution: {integrity: sha512-+ZRSDFTv4aC96aMb5E41rMjysx8ApkryevnvEYZvPZO52KvkqP5rNExLUXJFr9P4s0f3oqNQR6vopCZsPWKDcQ==}
 
-  '@next/swc-darwin-arm64@16.1.7':
-    resolution: {integrity: sha512-b2wWIE8sABdyafc4IM8r5Y/dS6kD80JRtOGrUiKTsACFQfWWgUQ2NwoUX1yjFMXVsAwcQeNpnucF2ZrujsBBPg==}
+  '@next/swc-darwin-arm64@16.2.6':
+    resolution: {integrity: sha512-ZJGkkcNfYgrrMkqOdZ7zoLa1TOy0qpcMfk/z4Mh/FKUz40gVO+HNQWqmLxf67Z5WB64DRp0dhEbyHfel+6sJUg==}
     engines: {node: '>= 10'}
     cpu: [arm64]
     os: [darwin]
 
-  '@next/swc-darwin-x64@16.1.7':
-    resolution: {integrity: sha512-zcnVaaZulS1WL0Ss38R5Q6D2gz7MtBu8GZLPfK+73D/hp4GFMrC2sudLky1QibfV7h6RJBJs/gOFvYP0X7UVlQ==}
+  '@next/swc-darwin-x64@16.2.6':
+    resolution: {integrity: sha512-v/YLBHIY132Ced3puBJ7YJKw1lqsCrgcNo2aRJlCEyQrrCeRJlvGlnmxhPxNQI3KE3N1DN5r9TPNPvka3nq5RQ==}
     engines: {node: '>= 10'}
     cpu: [x64]
     os: [darwin]
 
-  '@next/swc-linux-arm64-gnu@16.1.7':
-    resolution: {integrity: sha512-2ant89Lux/Q3VyC8vNVg7uBaFVP9SwoK2jJOOR0L8TQnX8CAYnh4uctAScy2Hwj2dgjVHqHLORQZJ2wH6VxhSQ==}
+  '@next/swc-linux-arm64-gnu@16.2.6':
+    resolution: {integrity: sha512-RPOvqlYBbcQjkz9VQQDZ2T2bARIjXZV1KFlt+V2Mr6SW/e4I9fcKsaA0hdyf2FHoTlsV2xnBd5Y912rP/1Ce6w==}
     engines: {node: '>= 10'}
     cpu: [arm64]
     os: [linux]
 
-  '@next/swc-linux-arm64-musl@16.1.7':
-    resolution: {integrity: sha512-uufcze7LYv0FQg9GnNeZ3/whYfo+1Q3HnQpm16o6Uyi0OVzLlk2ZWoY7j07KADZFY8qwDbsmFnMQP3p3+Ftprw==}
+  '@next/swc-linux-arm64-musl@16.2.6':
+    resolution: {integrity: sha512-URUTu1+dMkxJsPFgm+OeEvq9wf5sujw0EvgYy80TDGHTSLTnIHeqb0Eu8A3sC95IRgjejQL+kC4mw+4yPxiAXA==}
     engines: {node: '>= 10'}
     cpu: [arm64]
     os: [linux]
 
-  '@next/swc-linux-x64-gnu@16.1.7':
-    resolution: {integrity: sha512-KWVf2gxYvHtvuT+c4MBOGxuse5TD7DsMFYSxVxRBnOzok/xryNeQSjXgxSv9QpIVlaGzEn/pIuI6Koosx8CGWA==}
+  '@next/swc-linux-x64-gnu@16.2.6':
+    resolution: {integrity: sha512-DOj182mPV8G3UkrayLoREM5YEYI+Dk5wv7Ox9xl1fFibAELEsFD0lDPfHIeILlutMMfdyhlzYPELG3peuKaurw==}
     engines: {node: '>= 10'}
     cpu: [x64]
     os: [linux]
 
-  '@next/swc-linux-x64-musl@16.1.7':
-    resolution: {integrity: sha512-HguhaGwsGr1YAGs68uRKc4aGWxLET+NevJskOcCAwXbwj0fYX0RgZW2gsOCzr9S11CSQPIkxmoSbuVaBp4Z3dA==}
+  '@next/swc-linux-x64-musl@16.2.6':
+    resolution: {integrity: sha512-HKQ5SP/V/ub73UvF7n/zeJlxk2kLmtL7Wzrg4WfmkjmNos5onJ2tKu7yZOPdL18A6Svfn3max29ym+ry7NkK4g==}
     engines: {node: '>= 10'}
     cpu: [x64]
     os: [linux]
 
-  '@next/swc-win32-arm64-msvc@16.1.7':
-    resolution: {integrity: sha512-S0n3KrDJokKTeFyM/vGGGR8+pCmXYrjNTk2ZozOL1C/JFdfUIL9O1ATaJOl5r2POe56iRChbsszrjMAdWSv7kQ==}
+  '@next/swc-win32-arm64-msvc@16.2.6':
+    resolution: {integrity: sha512-LZXpTlPyS5v7HhSmnvsLGP3iIYgYOBnc8r8ArlT55sGHV89bR2HlDdBjWQ+PY6SJMmk8TuVGFuxalnP3k/0Dwg==}
     engines: {node: '>= 10'}
     cpu: [arm64]
     os: [win32]
 
-  '@next/swc-win32-x64-msvc@16.1.7':
-    resolution: {integrity: sha512-mwgtg8CNZGYm06LeEd+bNnOUfwOyNem/rOiP14Lsz+AnUY92Zq/LXwtebtUiaeVkhbroRCQ0c8GlR4UT1U+0yg==}
+  '@next/swc-win32-x64-msvc@16.2.6':
+    resolution: {integrity: sha512-F0+4i0h9J6C4eE3EAPWsoCk7UW/dbzOjyzxY0qnDUOYFu6FFmdZ6l97/XdV3/Nz3VYyO7UWjyEJUXkGqcoXfMA==}
     engines: {node: '>= 10'}
     cpu: [x64]
     os: [win32]
@@ -1912,6 +1912,9 @@ packages:
   '@swc/helpers@0.5.15':
     resolution: {integrity: sha512-JQ5TuMi45Owi4/BIMAJBoSQoOJu12oOk/gADqlcUL9JEdHB8vyjUSsxqeNXnmXHjYKMi2WcYtezGEEhqUI/E2g==}
 
+  '@swc/helpers@0.5.21':
+    resolution: {integrity: sha512-jI/VAmtdjB/RnI8GTnokyX7Ug8c+g+ffD6QRLa6XQewtnGyukKkKSk3wLTM3b5cjt1jNh9x0jfVlagdN2gDKQg==}
+
   '@t3-oss/env-core@0.12.0':
     resolution: {integrity: sha512-lOPj8d9nJJTt81mMuN9GMk8x5veOt7q9m11OSnCBJhwp1QrL/qR+M8Y467ULBSm9SunosryWNbmQQbgoiMgcdw==}
     peerDependencies:
@@ -2652,8 +2655,8 @@ packages:
   base64-js@1.5.1:
     resolution: {integrity: sha512-AKpaYlHn8t4SVbOHCy+b5+KKgvR4vrsD8vbvrbiQJps7fKDTkjkDry6ji0rUJjC0kzbNePLwzxq8iypo41qeWA==}
 
-  baseline-browser-mapping@2.10.8:
-    resolution: {integrity: sha512-PCLz/LXGBsNTErbtB6i5u4eLpHeMfi93aUv5duMmj6caNu6IphS4q6UevDnL36sZQv9lrP11dbPKGMaXPwMKfQ==}
+  baseline-browser-mapping@2.10.29:
+    resolution: {integrity: sha512-Asa2krT+XTPZINCS+2QcyS8WTkObE77RwkydwF7h6DmnKqbvlalz93m/dnphUyCa6SWSP51VgtEUf2FN+gelFQ==}
     engines: {node: '>=6.0.0'}
     hasBin: true
 
@@ -2710,8 +2713,8 @@ packages:
   camelize@1.0.1:
     resolution: {integrity: sha512-dU+Tx2fsypxTgtLoE36npi3UqcjSSMNYfkqgmoEhtZrraP5VWq0K7FkWVTYa8eMPtnU/G2txVsfdCJTn9uzpuQ==}
 
-  caniuse-lite@1.0.30001780:
-    resolution: {integrity: sha512-llngX0E7nQci5BPJDqoZSbuZ5Bcs9F5db7EtgfwBerX9XGtkkiO4NwfDDIRzHTTwcYC8vC7bmeUEPGrKlR/TkQ==}
+  caniuse-lite@1.0.30001792:
+    resolution: {integrity: sha512-hVLMUZFgR4JJ6ACt1uEESvQN1/dBVqPAKY0hgrV70eN3391K6juAfTjKZLKvOMsx8PxA7gsY1/tLMMTcfFLLpw==}
 
   canvas-confetti@1.9.4:
     resolution: {integrity: sha512-yxQbJkAVrFXWNbTUjPqjF7G+g6pDotOUHGbkZq2NELZUMDpiJ85rIEazVb8GTaAptNW2miJAXbs1BtioA251Pw==}
@@ -4389,8 +4392,8 @@ packages:
       react: ^16.8 || ^17 || ^18 || ^19 || ^19.0.0-rc
       react-dom: ^16.8 || ^17 || ^18 || ^19 || ^19.0.0-rc
 
-  next@16.1.7:
-    resolution: {integrity: sha512-WM0L7WrSvKwoLegLYr6V+mz+RIofqQgVAfHhMp9a88ms0cFX8iX9ew+snpWlSBwpkURJOUdvCEt3uLl3NNzvWg==}
+  next@16.2.6:
+    resolution: {integrity: sha512-qOVgKJg1+At15NpeUP+eJgCHvTCgXsogweq87Ri/Ix7PkqQHg4sdaXmSFqKlgaIXE4kW0g25LE68W87UANlHtw==}
     engines: {node: '>=20.9.0'}
     hasBin: true
     peerDependencies:
@@ -5013,6 +5016,11 @@ packages:
     engines: {node: '>=10'}
     hasBin: true
 
+  semver@7.8.0:
+    resolution: {integrity: sha512-AcM7dV/5ul4EekoQ29Agm5vri8JNqRyj39o0qpX6vDF2GZrtutZl5RwgD1XnZjiTAfncsJhMI48QQH3sN87YNA==}
+    engines: {node: '>=10'}
+    hasBin: true
+
   server-only@0.0.1:
     resolution: {integrity: sha512-qepMx2JxAa5jjfzxG79yPPq+8BuFToHd1hm7kI+Z4zAq1ftQiP7HcxMhDDItrbtwVeLg/cY2JnKnrcFkmiswNA==}
 
@@ -6066,7 +6074,7 @@ snapshots:
       tslib: 2.8.1
     optional: true
 
-  '@emnapi/runtime@1.9.0':
+  '@emnapi/runtime@1.10.0':
     dependencies:
       tslib: 2.8.1
     optional: true
@@ -6343,7 +6351,7 @@ snapshots:
 
   '@img/sharp-wasm32@0.34.5':
     dependencies:
-      '@emnapi/runtime': 1.9.0
+      '@emnapi/runtime': 1.10.0
     optional: true
 
   '@img/sharp-win32-arm64@0.34.5':
@@ -6598,38 +6606,38 @@ snapshots:
   '@napi-rs/wasm-runtime@0.2.12':
     dependencies:
       '@emnapi/core': 1.8.1
-      '@emnapi/runtime': 1.9.0
+      '@emnapi/runtime': 1.10.0
       '@tybys/wasm-util': 0.10.1
     optional: true
 
-  '@next/env@16.1.7': {}
+  '@next/env@16.2.6': {}
 
   '@next/eslint-plugin-next@15.5.12':
     dependencies:
       fast-glob: 3.3.1
 
-  '@next/swc-darwin-arm64@16.1.7':
+  '@next/swc-darwin-arm64@16.2.6':
     optional: true
 
-  '@next/swc-darwin-x64@16.1.7':
+  '@next/swc-darwin-x64@16.2.6':
     optional: true
 
-  '@next/swc-linux-arm64-gnu@16.1.7':
+  '@next/swc-linux-arm64-gnu@16.2.6':
     optional: true
 
-  '@next/swc-linux-arm64-musl@16.1.7':
+  '@next/swc-linux-arm64-musl@16.2.6':
     optional: true
 
-  '@next/swc-linux-x64-gnu@16.1.7':
+  '@next/swc-linux-x64-gnu@16.2.6':
     optional: true
 
-  '@next/swc-linux-x64-musl@16.1.7':
+  '@next/swc-linux-x64-musl@16.2.6':
     optional: true
 
-  '@next/swc-win32-arm64-msvc@16.1.7':
+  '@next/swc-win32-arm64-msvc@16.2.6':
     optional: true
 
-  '@next/swc-win32-x64-msvc@16.1.7':
+  '@next/swc-win32-x64-msvc@16.2.6':
     optional: true
 
   '@nodelib/fs.scandir@2.1.5':
@@ -7192,7 +7200,7 @@ snapshots:
       '@react-aria/interactions': 3.27.1(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
       '@react-aria/utils': 3.33.1(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
       '@react-types/shared': 3.33.1(react@19.2.4)
-      '@swc/helpers': 0.5.15
+      '@swc/helpers': 0.5.21
       clsx: 2.1.1
       react: 19.2.4
       react-dom: 19.2.4(react@19.2.4)
@@ -7203,13 +7211,13 @@ snapshots:
       '@react-aria/utils': 3.33.1(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
       '@react-stately/flags': 3.1.2
       '@react-types/shared': 3.33.1(react@19.2.4)
-      '@swc/helpers': 0.5.15
+      '@swc/helpers': 0.5.21
       react: 19.2.4
       react-dom: 19.2.4(react@19.2.4)
 
   '@react-aria/ssr@3.9.10(react@19.2.4)':
     dependencies:
-      '@swc/helpers': 0.5.15
+      '@swc/helpers': 0.5.21
       react: 19.2.4
 
   '@react-aria/utils@3.33.1(react-dom@19.2.4(react@19.2.4))(react@19.2.4)':
@@ -7218,18 +7226,18 @@ snapshots:
       '@react-stately/flags': 3.1.2
       '@react-stately/utils': 3.11.0(react@19.2.4)
       '@react-types/shared': 3.33.1(react@19.2.4)
-      '@swc/helpers': 0.5.15
+      '@swc/helpers': 0.5.21
       clsx: 2.1.1
       react: 19.2.4
       react-dom: 19.2.4(react@19.2.4)
 
   '@react-stately/flags@3.1.2':
     dependencies:
-      '@swc/helpers': 0.5.15
+      '@swc/helpers': 0.5.21
 
   '@react-stately/utils@3.11.0(react@19.2.4)':
     dependencies:
-      '@swc/helpers': 0.5.15
+      '@swc/helpers': 0.5.21
       react: 19.2.4
 
   '@react-types/shared@3.33.1(react@19.2.4)':
@@ -7437,6 +7445,10 @@ snapshots:
     dependencies:
       tslib: 2.8.1
 
+  '@swc/helpers@0.5.21':
+    dependencies:
+      tslib: 2.8.1
+
   '@t3-oss/env-core@0.12.0(typescript@5.9.3)(zod@3.25.76)':
     optionalDependencies:
       typescript: 5.9.3
@@ -8249,7 +8261,7 @@ snapshots:
 
   base64-js@1.5.1: {}
 
-  baseline-browser-mapping@2.10.8: {}
+  baseline-browser-mapping@2.10.29: {}
 
   best-effort-json-parser@1.2.1: {}
 
@@ -8313,7 +8325,7 @@ snapshots:
 
   camelize@1.0.1: {}
 
-  caniuse-lite@1.0.30001780: {}
+  caniuse-lite@1.0.30001792: {}
 
   canvas-confetti@1.9.4: {}
 
@@ -9643,7 +9655,7 @@ snapshots:
 
   is-bun-module@2.0.0:
     dependencies:
-      semver: 7.7.4
+      semver: 7.8.0
 
   is-callable@1.2.7: {}
 
@@ -10531,25 +10543,25 @@ snapshots:
       react: 19.2.4
       react-dom: 19.2.4(react@19.2.4)
 
-  next@16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4):
+  next@16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4):
     dependencies:
-      '@next/env': 16.1.7
+      '@next/env': 16.2.6
       '@swc/helpers': 0.5.15
-      baseline-browser-mapping: 2.10.8
-      caniuse-lite: 1.0.30001780
+      baseline-browser-mapping: 2.10.29
+      caniuse-lite: 1.0.30001792
       postcss: 8.4.31
       react: 19.2.4
       react-dom: 19.2.4(react@19.2.4)
       styled-jsx: 5.1.6(react@19.2.4)
     optionalDependencies:
-      '@next/swc-darwin-arm64': 16.1.7
-      '@next/swc-darwin-x64': 16.1.7
-      '@next/swc-linux-arm64-gnu': 16.1.7
-      '@next/swc-linux-arm64-musl': 16.1.7
-      '@next/swc-linux-x64-gnu': 16.1.7
-      '@next/swc-linux-x64-musl': 16.1.7
-      '@next/swc-win32-arm64-msvc': 16.1.7
-      '@next/swc-win32-x64-msvc': 16.1.7
+      '@next/swc-darwin-arm64': 16.2.6
+      '@next/swc-darwin-x64': 16.2.6
+      '@next/swc-linux-arm64-gnu': 16.2.6
+      '@next/swc-linux-arm64-musl': 16.2.6
+      '@next/swc-linux-x64-gnu': 16.2.6
+      '@next/swc-linux-x64-musl': 16.2.6
+      '@next/swc-win32-arm64-msvc': 16.2.6
+      '@next/swc-win32-x64-msvc': 16.2.6
       '@opentelemetry/api': 1.9.0
       '@playwright/test': 1.59.1
       sharp: 0.34.5
@@ -10557,13 +10569,13 @@ snapshots:
       - '@babel/core'
       - babel-plugin-macros
 
-  nextra-theme-docs@4.6.1(@types/react@19.2.13)(next@16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(nextra@4.6.1(next@16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(use-sync-external-store@1.6.0(react@19.2.4)):
+  nextra-theme-docs@4.6.1(@types/react@19.2.13)(next@16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(nextra@4.6.1(next@16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(use-sync-external-store@1.6.0(react@19.2.4)):
     dependencies:
       '@headlessui/react': 2.2.9(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
       clsx: 2.1.1
-      next: 16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
+      next: 16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
       next-themes: 0.4.6(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
-      nextra: 4.6.1(next@16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3)
+      nextra: 4.6.1(next@16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3)
       react: 19.2.4
       react-compiler-runtime: 19.1.0-rc.3(react@19.2.4)
       react-dom: 19.2.4(react@19.2.4)
@@ -10575,7 +10587,7 @@ snapshots:
       - immer
       - use-sync-external-store
 
-  nextra@4.6.1(next@16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3):
+  nextra@4.6.1(next@16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4))(react-dom@19.2.4(react@19.2.4))(react@19.2.4)(typescript@5.9.3):
     dependencies:
       '@formatjs/intl-localematcher': 0.6.2
       '@headlessui/react': 2.2.9(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
@@ -10596,7 +10608,7 @@ snapshots:
       mdast-util-gfm: 3.1.0
       mdast-util-to-hast: 13.2.1
       negotiator: 1.0.0
-      next: 16.1.7(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
+      next: 16.2.6(@opentelemetry/api@1.9.0)(@playwright/test@1.59.1)(react-dom@19.2.4(react@19.2.4))(react@19.2.4)
       react: 19.2.4
       react-compiler-runtime: 19.1.0-rc.3(react@19.2.4)
       react-dom: 19.2.4(react@19.2.4)
@@ -10925,7 +10937,7 @@ snapshots:
 
   postcss@8.4.31:
     dependencies:
-      nanoid: 3.3.11
+      nanoid: 3.3.12
       picocolors: 1.1.1
       source-map-js: 1.2.1
 
@@ -11365,6 +11377,8 @@ snapshots:
 
   semver@7.7.4: {}
 
+  semver@7.8.0: {}
+
   server-only@0.0.1: {}
 
   set-function-length@1.2.2:
@@ -11393,7 +11407,7 @@ snapshots:
     dependencies:
       '@img/colour': 1.1.0
       detect-libc: 2.1.2
-      semver: 7.7.4
+      semver: 7.8.0
     optionalDependencies:
       '@img/sharp-darwin-arm64': 0.34.5
       '@img/sharp-darwin-x64': 0.34.5