What is the Fledge MCP Server?

The Fledge MCP Server is a Model Context Protocol (MCP) server that connects Fledge functionality to Cursor AI, allowing the AI to interact with Fledge instances via natural language commands.

What are the prerequisites for installing the Fledge MCP Server?

The prerequisites include Fledge installed locally or accessible via API, Cursor AI installed, and Python 3.8+.

How do I install the Fledge MCP Server?

1. Clone the repository: git clone https://github.com/Krupalp525/fledge-mcp.git
2. Navigate to the directory: cd fledge-mcp
3. Install dependencies: pip install -r requirements.txt

How do I run the Fledge MCP Server?

Make sure Fledge is running (fledge start), then start the MCP server with python mcp_server.py. For a secure server, use python secure_mcp_server.py.

How do I verify that the server is running?

Access the health endpoint with curl http://localhost:8082/health. You should receive “Fledge MCP Server is running” as the response.

How do I connect the server to Cursor AI?

In Cursor, go to Settings > MCP Servers, add a new server with the URL http://localhost:8082/tools, and upload the included tools.json file or point to its local path.

How do I use the secure server with API key authentication?

Configure the “X-API-Key” header with the value from the api_key.txt file that is generated when the secure server starts.

What are some available tools in the Fledge MCP Server?

Some available tools include get_sensor_data, list_sensors, ingest_test_data, get_service_status, start_stop_service, update_config, generate_ui_component, and more.

How can I test the API?

You can test the server using the included test scripts: python test_mcp.py for the standard server and python test_secure_mcp.py for the secure server.

How do I add more tools to the server?

1. Add the tool definition to tools.json.
2. Implement the tool handler in mcp_server.py and secure_mcp_server.py.

What are some production considerations for deploying the Fledge MCP Server?

Consider using HTTPS, deploying behind a reverse proxy like Nginx, implementing robust authentication (JWT, OAuth), adding rate limiting, and setting up persistent data storage for subscriptions.

How can I deploy the Fledge MCP Server on Smithery.ai?

1. Build the Docker image: docker build -t fledge-mcp .
2. Deploy to Smithery.ai: smithery deploy
3. Configure environment variables in your Smithery.ai dashboard.

What environment variables should I set on Smithery.ai?

Set FLEDGE_API_URL to your Fledge API endpoint and API_KEY to your secure API key (if using secure mode).

What JSON-RPC methods are supported by the server?

The server supports initialize, tools/list, and tools/call methods.

What are the common JSON-RPC error codes?

Common error codes include -32700 (Parse error), -32600 (Invalid Request), -32601 (Method not found), -32602 (Invalid params), and -32000 (Server error).