DocumentationDeveloper ToolsScrapeless MCPCursor

Cursor

Connect Scrapeless MCP to the Cursor IDE. You’ll need your Scrapeless API token.

Setup

  1. Open Cursor and go to Settings → MCP → Add new global MCP server (this opens ~/.cursor/mcp.json). You can also create a project-scoped .cursor/mcp.json in your workspace.
  2. Add the Scrapeless server:
{
  "mcpServers": {
    "Scrapeless MCP Server": {
      "command": "npx",
      "args": ["-y", "scrapeless-mcp-server"],
      "env": {
        "SCRAPELESS_KEY": "YOUR_SCRAPELESS_KEY"
      }
    }
  }
}
  1. Save the file. Cursor lists the server under Settings → MCP; toggle it on if needed.
  2. In Chat/Composer (Agent mode), ask Cursor to use the web — for example:
    • “Search Google for a solution to this error and summarize the top results.”
    • “Scrape the HTML from this page.”

Cursor will call Scrapeless in the background.

To use the hosted endpoint instead of a local process, use the Streamable HTTP config from Authentication and Transports.

Claude Desktop / Claude CodeCline