Fetch

Provides web content fetching and Markdown extraction for MCP clients, with chunked reads and optional robots.txt handling.

typescript

7

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "tgambet-mcp-fetch-node": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-fetch-node"
      ]
    }
  }
}

Fetch MCP Server provides web content fetching capabilities for Model Context Protocol clients. It enables LLMs to retrieve web pages, convert HTML to markdown, and read content in chunks using a configurable start index to extract the information you need.

How to use

Start the server using one of the available runtimes, then use your MCP client to invoke the fetch tool with a URL. The server exposes an SSE endpoint at /sse on port 8080 by default, so you can establish a streaming connection and request content from web pages. Use the fetch tool to retrieve a URL and receive the page content as markdown, with content delivered in chunks controlled by the start_index and max_length parameters.

How to install

Prerequisites: ensure you have Node.js and npm or Docker installed on your machine.

npx -y mcp-fetch-node

Alternatively, you can run the server via Docker.

docker run -it tgambet/mcp-fetch-node

Configuration and usage notes

Robots.txt handling and user-agent customization are available. You can disable robots.txt handling with a run-time flag, and you can override the user-agent string by supplying --user-agent=YourUserAgent when starting the server.

By default, requests initiated by a model tool will identify themselves with a specific User-Agent, while user-initiated prompts will use a distinct User-Agent. You can customize this behavior via the same run-time flag described above.

Notes and features

Key features include: fetching and extracting content from a URL, converting to Markdown, respecting robots.txt (optional), and supporting pagination through start_index to read content in chunks.

Available tools

fetch

Fetches a URL and extracts its contents as Markdown. Supports parameters like url (required), max_length (optional), start_index (optional), and raw (optional) to return raw content without Markdown conversion.