Cursor

Connect Cursor IDE to the CoreClaw MCP Server so your AI coding assistant can discover CoreClaw workers, run them, monitor runs, and retrieve results directly within your editor.

Prerequisites

Cursor installed
A CoreClaw account with an API key from Console -> Settings -> API & Integrations

Configuration

Cursor supports MCP via Streamable HTTP transport. Configure it by creating an mcp.json file in your project or global settings.

Step 1: Create the MCP Configuration File

Project-level:

.your-project/.cursor/mcp.json

Global:

~/.cursor/mcp.json

Step 2: Add the CoreClaw MCP Configuration

{
  "mcpServers": {
    "coreclaw": {
      "url": "https://mcp.coreclaw.com/mcp",
      "headers": {
        "api-key": "YOUR_CORECLAW_API_KEY"
      }
    }
  }
}

Replace YOUR_CORECLAW_API_KEY with your actual CoreClaw API key.

Cursor IDE MCP Configuration

Step 3: Restart Cursor

After saving the configuration, restart Cursor for the changes to take effect.

Verify the Connection

Open Cursor and start a new chat.
Switch to Agent mode.
Ask: “Find Amazon workers on CoreClaw.”
Cursor should call list_store_workers and return matching workers.

Example Conversation

You: Find a Google Maps worker and extract restaurant data near Central Park, New York.

Cursor: I’ll find a Google Maps worker, inspect its input schema, and run it for you. [Calls list_store_workers -> get_worker_input_schema -> run_worker -> get_worker_run -> list_worker_run_results]

Troubleshooting

Tools not appearing

Restart Cursor after MCP config changes.
Ensure you are in Agent mode.
Check that mcp.json is in the correct location.

Authentication errors

Ensure the api-key header value matches your CoreClaw API key exactly.
Verify your key is active in the Console.