Fetch MCP Server 1kd

🚀 MCP Fetch Server

A Model Context Protocol (MCP) server that provides web content scraping capabilities. This server allows Large Language Models (LLMs) to scrape and process content from web pages, converting HTML to Markdown for easy reading.

The fetch tool truncates responses, but by using the start_index parameter, you can specify where to start extracting content. This enables the model to read web pages in segments until it finds the required information.

✨ Features

Available Tools

• fetch - Scrapes a URL on the internet and extracts its content in Markdown format.
  • url (string, required): The URL to scrape.
  • max_length (integer, optional): The maximum number of characters to return (default: 5000).
  • start_index (integer, optional): The character index from which to start extracting content (default: 0).
  • raw (boolean, optional): Get the raw content without Markdown conversion (default: false).

Tips

• fetch
  • Scrapes a URL and extracts its content in Markdown format.
  • Parameters:
    • url (string, required): The URL to scrape.

📦 Installation

Optional: Install node.js, which will enable the fetch server to use a more powerful HTML simplifier.

Using uv (Recommended)

When using uv, no specific installation is required. We'll use uvx to directly run mcp-server-fetch.

Using PIP

Alternatively, you can install mcp-server-fetch via pip:

pip install mcp-server-fetch