Fetch MCP Server

@phpmac

Basic information

About Fetch MCP Server

Fetch MCP Server

Fetch MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "fetch_mcp": {
      "command": "bun",
      "args": [
        "start"
      ]
    }
  }
}

Fetch MCP Server tools

Fetch website content and return as HTML

Fetch JSON file from URL

Fetch website content and return as plain text (no HTML)

Fetch website content and return as Markdown

Overview

What is Fetch MCP Server?

The Fetch MCP Server is a Model Context Protocol server that fetches web content in various formats—HTML, JSON, plain text, and Markdown. It is designed for MCP clients that need to retrieve and transform web content on demand.

How to use Fetch MCP Server?

Start the server with either bun i && bun start or docker compose up --build -d. Configure your MCP client with an SSE transport pointing to http://localhost:3000/sse, optionally including custom headers and setting useNodeEventSource: true.

Key features of Fetch MCP Server

Fetches web content as HTML, JSON, plain text, or Markdown.
Accepts optional custom headers per request.
Offers four dedicated tools: fetch_html, fetch_json, fetch_txt, fetch_markdown.
Can be started via bun (JavaScript runtime) or Docker.
Uses SSE (Server-Sent Events) transport for MCP integration.

Use cases of Fetch MCP Server

Retrieve web page HTML for scraping or analysis.
Fetch JSON data from APIs and return parsed content.
Convert webpage content to clean Markdown for LLM consumption.
Strip HTML tags and scripts to obtain plain text.

Frequently asked questions

What dependencies are required to run Fetch MCP Server?

You need either `bun` installed or Docker to build and run the container.

How do I authenticate requests with the server?

You can pass custom headers (including Bearer tokens) in the MCP client configuration’s `headers` object.

What transport protocol does it use?

The server communicates via Server-Sent Events (SSE) at the endpoint `http://localhost:3000/sse`.

Does the server store any data or resources?

No, the server does not provide any persistent resources. It fetches and transforms web content on demand.

What output formats are supported?

The server can return HTML, JSON, plain text (with HTML tags removed), and Markdown.

Comments

More Browser Automation MCP servers

Loadster

loadster

Loadster is a load testing, stress testing, and synthetic monitoring platform. Run Playwright, Browser Bots, and Protocol Bots to test and monitor your websites, web apps, and APIs. Loadster MCP gives your agents the abi

Scrapling Fetch MCP

cyberchitta

Helps AI assistants access text content from bot-protected websites. MCP server that fetches HTML/markdown from sites with anti-automation measures using Scrapling.

101