Skip to content

Fetch MCP Server

Fetch any web page and convert it to markdown Claude can read in chunks.

Reference serverby Model Context ProtocolRuns on your computerChecked October 18, 2026

Fetch is a reference server from the Model Context Protocol project that downloads a web page and converts its HTML into markdown for Claude to read. Long pages come back in chunks, requests made by the model respect robots.txt, and private network addresses are blocked unless you allow them.

Who it is for

  • Anyone who wants Claude to read specific web pages
  • Developers learning how MCP servers work
  • Researchers summarizing articles and documentation

Prompts to try

Once it is connected, ask in plain words. Claude decides when to use the server and asks before it changes anything important.

Prompt 1
Fetch https://modelcontextprotocol.io/introduction and explain it in plain English.
Prompt 2
Read this blog post and list every statistic it cites along with its source.
Prompt 3
Fetch this library's changelog and tell me what changed in the latest release.

How to connect it

In Claude Code

  1. claude mcp add --transport stdio fetch -- uvx mcp-server-fetch

In the Claude desktop app

Add this to claude_desktop_config.json, then restart the app:

json
{
  "mcpServers": {
    "fetch": {
      "command": "uvx",
      "args": [
        "mcp-server-fetch"
      ]
    }
  }
}

Good to know

Also runs with Docker (docker run -i --rm mcp/fetch); having Node.js installed gives it a more robust HTML simplifier.

Unlock the setup steps

Free. Leave your email once to open the setup steps for all 150 servers, and get the best new connectors each month.

No spam. One click to unsubscribe. We use your email to send you the course and AI tips.

How we checked it

Checked the src/fetch folder of modelcontextprotocol/servers (MIT, last commit 2026-10-10) and the mcp-server-fetch PyPI release 2026.10.10, published 2026-10-11.

Source: github.com/modelcontextprotocol/servers/tree/main/src/fetch