Connect Cursor to MCP Servers

What Cursor MCP Integration Gives You

Cursor's Agent mode (Cmd/Ctrl + I) can call MCP tools directly during code generation. This means the AI can query your database, read your documentation, call your APIs, or interact with any service — all without leaving the editor.

Step 1 — Open Cursor MCP Settings

Cursor stores MCP configuration in a dedicated settings file (separate from VS Code's settings.json). There are two ways to open it:

Step 2 — Add an MCP Server

The MCP configuration file uses the same format as Claude Desktop. Here's the basic structure:

{
  "mcpServers": {
    "server-name": {
      "command": "node",
      "args": ["/path/to/server/index.js"],
      "env": {
        "API_KEY": "your-api-key-here"
      }
    }
  }
}

Each server entry has:

• key — a unique name for the server (shown in Cursor UI)
• command — the executable to run (node, python, npx, uvx, etc.)
• args — array of arguments passed to the command
• env — optional environment variables (for API keys, etc.)

Real examples

Filesystem server (via npx)

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-filesystem",
        "/Users/you/projects"
      ]
    }
  }
}

Python MCP server (via uv)

{
  "mcpServers": {
    "my-python-server": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/my-mcp-server",
        "run",
        "server.py"
      ]
    }
  }
}

HTTP/SSE server (remote)

{
  "mcpServers": {
    "remote-server": {
      "url": "https://your-mcp-server.example.com/sse"
    }
  }
}

Step 3 — Enable and Test

After saving your mcp.json, reload Cursor. MCP servers start automatically when Cursor launches.

Troubleshooting