feishu_mcp_server

This MCP server acts as a bridge between Claude/Cursor and a Feishu (飞书) robot. It receives requests from MCP clients, routes them to the Feishu robot via its API, and returns the responses back through the MCP pipeline. The server enables business-logic workflows by leveraging a Feishu chatbot as the execution layer, so you can compose prompts and model-assisted commands that ultimately trigger Feishu API calls and bot interactions. In practice, you would configure your MCP client to target this feishu_mcp_server, provide the needed webhook or endpoint, and let the server handle the interaction with Feishu.

Once running, you can use the available capabilities to send commands or queries through the MCP client and have the Feishu bot perform actions such as sending messages, responding to events, or executing business processes in Feishu. The server acts as the translator and executor: it interprets the incoming request, calls the Feishu API as needed, and returns the result to the MCP client for display or further processing.

Prerequisites:

• Python environment with uv or uvicorn tooling compatible with the uv workflow described by the project
• Access credentials for Feishu API and a target Feishu bot/webhook
• Node.js not required for this server (this is a Python-based UV workflow as shown in the docs)

1. Install Python dependencies (adjust to your environment):

• Create a virtual environment: python -m venv venv
• Activate it: source venv/bin/activate (Linux/macOS) or venv\Scripts\activate (Windows)
• Install requirements: pip install -r requirements.txt

2. Configure the server:

• Replace YOUR_PATH with the local directory containing bot.py
• Replace YOUR_WEB_HOOK with your Feishu webhook URL or the appropriate webhook endpoint
• If needed, set additional environment variables such as WEBHOOK_URL in your environment or via a .env file

3. Run the MCP server:

• uv --directory YOUR_PATH run bot.py --webhook YOUR_WEB_HOOK
• Ensure the process stays alive and accessible to your MCP client

4. Validate connectivity:

• From your MCP client, send a test request to feishu_mcp_server and verify a response from the Feishu bot

5. Optional: integrate Smithery/CLI for deployment if you want automated installation via Smithery (as shown in the README):

• npx -y @smithery/cli install @Chenzhi-Ana/feishu_mcp_server --client claude

Tips:

• Ensure the Feishu bot has the necessary permissions and that the webhook URL is correctly configured.
• If the Feishu API requires authentication tokens, store them securely as environment variables (e.g., FEISHU_TOKEN) and read them in your bot.py.
• The --webhook flag should point to the Feishu webhook endpoint that your bot uses; keep it consistent with your Feishu app configuration.
• When updating code, restart the MCP server to apply changes.
• If you encounter connectivity issues, check network access between the MCP server host and Feishu API endpoints, and verify firewall rules.
• For debugging, log incoming MCP requests and Feishu API responses to help diagnose problems.