Docs
Log in

Use with Cursor

Crawlbase as a Cursor MCP server. Pull live docs, scrape competitive code, and fetch reference content without leaving the editor.

Setup

Cursor reads MCP servers from a JSON file. Open Cursor Settings → Tools and Integrations → Add Custom MCP, or edit the file directly:

ScopePath
Global (all projects)~/.cursor/mcp.json
Per project.cursor/mcp.json in the project root

Configuration

{
  "mcpServers": {
    "crawlbase": {
      "type": "stdio",
      "command": "npx",
      "args": ["@crawlbase/mcp@latest"],
      "env": {
        "CRAWLBASE_TOKEN": "YOUR_TOKEN",
        "CRAWLBASE_JS_TOKEN": "YOUR_JS_TOKEN"
      }
    }
  }
}

Save and reload Cursor (Cmd/Ctrl+Shift+P → "Reload Window"). The Crawlbase tools should now show as a green dot in the MCP settings panel.

Usage in chat

Open the Cursor chat panel (Cmd/Ctrl+L). The AI can now reach for Crawlbase tools when relevant.

Fetch the latest changelog from https://nextjs.org/blog and tell me
what's new in the most recent release. Then update our package.json
to match if our version is older.

Patterns that work well

  • Live docs lookup: "Pull the latest library docs and explain how to do X" — beats whatever's in the model's training data.
  • Competitive analysis: "Look at how competitor implements feature on their public site" — pair with code edits.
  • Build-time scraping: add Crawlbase calls to a tool prompt for one-off data fetches you don't want in the codebase.
Tag a project with .cursor/mcp.json

Per-project configs let different repos use different Crawlbase tokens — useful if you have separate accounts for prod and dev workloads.