Fetch MCP Server for Cursor: Setup & Config

What you can do with Fetch in Cursor

Choose format-specific web retrieval with readable articles, transcripts, and bounded output.

• Read long pages in character-based chunks.
• Extract a YouTube transcript in a selected language.

Before you start: Fetch in Cursor

• Node.js with npx
• Optional yt-dlp for transcript extraction

This is the README MCP launch, not the mcp-fetch CLI invocation. The supplied structured tool catalog and README list different tools; verify the running server tool list.

Fetch Cursor configuration

1. Choose Cursor configuration scope

   For Fetch MCP Server, open Project: .cursor/mcp.json User: ~/.cursor/mcp.json. Merge into mcpServers at project or user scope.

2. Add Fetch MCP Server to Cursor

   Use the fetch-mcp-zcaceres entry below in mcpServers. Keep Fetch MCP Server's documented stdio connection fields and your other Cursor entries.

3. Check Fetch MCP Server in Cursor

   Fetch MCP Server: Enable in Cursor’s MCP settings; inspect tools.

{
  "mcpServers": {
    "fetch-mcp-zcaceres": {
      "command": "npx",
      "args": [
        "mcp-fetch-server"
      ],
      "env": {
        "DEFAULT_LIMIT": "<Optional default response character limit>",
        "MAX_RESPONSE_BYTES": "<Optional response-body byte limit>"
      }
    }
  }
}

Fetch tools available to Cursor

Fetch's stored tool list documents 11 tools; compare it with Cursor's connected tool list, because the stored list is not a live check.

fetch_html

Find fetch_html under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

Retrieve a page as raw HTML.

fetch_markdown

Find fetch_markdown under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

fetch_txt

Find fetch_txt under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

fetch_json

Find fetch_json under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

fetch_readable

Find fetch_readable under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

fetch_youtube_transcript

Find fetch_youtube_transcript under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

fetch

Find fetch under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

Issue an HTTP request with configurable method, headers, body, and options.

get-rules

Find get-rules under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

Retrieve guidance on using the server tools.

graphql

Find graphql under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

Run GraphQL queries or mutations with variables and headers.

puppeteer

Find puppeteer under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

Operate browser pages, execute JavaScript, and capture screenshots.

socket

Find socket under Fetch in Cursor's MCP settings; inspect its arguments when the agent requests a call.

Open, inspect, use, or close WebSocket connections.

Fetch: full stored tool descriptions

Troubleshooting Fetch in Cursor

• Cursor must find npx mcp-fetch-server (Node.js/npm) for Fetch MCP Server. Enable the project/user entry in MCP settings and inspect its tools.
• Review DEFAULT_LIMIT, MAX_RESPONSE_BYTES for Fetch MCP Server before sharing project .cursor/mcp.json; project scope travels with the repository.
• Fetch MCP Server uses stdio here: Cursor must keep the configured process running; protocol output belongs on stdout and diagnostics on stderr.
• Known limitation: Default response character and byte limits constrain large fetches.
• Known limitation: Structured input tools differ from the README tool list.
• Known limitation: Private and localhost targets are blocked by the documented SSRF protection.
• General Cursor troubleshooting lives in the /clients/cursor guide.

Frequently asked questions