Fetch MCP Server
by tgambet
A Node.js port of the official Fetch MCP Server, providing web content fetching capabilities for LLMs. It enables retrieval and processing of web page content, converting HTML to markdown.
Last updated: N/A
What is Fetch MCP Server?
A Model Context Protocol (MCP) server that fetches web content and extracts it as markdown, making it easier for LLMs to consume. It provides an SSE endpoint for integration and can be used as a drop-in replacement for the original Python implementation.
How to use Fetch MCP Server?
The server exposes an SSE endpoint at /sse on port 8080 by default. It can be run using npx -y mcp-fetch-node or via Docker using docker run -it tgambet/mcp-fetch-node. Customization options include ignoring robots.txt with --ignore-robots-txt and setting a custom user-agent with --user-agent=YourUserAgent.
Key features of Fetch MCP Server
Fetch and extract content from URLs
Respects robots.txt (configurable)
User-Agent customization
Markdown conversion
Pagination
Use cases of Fetch MCP Server
Enabling LLMs to access and process web content
Retrieving information from web pages for AI models
Providing context to language models from online sources
Automated web content extraction for analysis
Building AI-powered web crawlers
FAQ from Fetch MCP Server
What is Model Context Protocol (MCP)?
What is Model Context Protocol (MCP)?
MCP is a protocol for providing context to language models.
How does this server differ from the original Fetch MCP Server?
How does this server differ from the original Fetch MCP Server?
This implementation is in TypeScript for Node.js, uses an SSE interface instead of stdio, and has a custom content extraction implementation.
How do I customize the user-agent?
How do I customize the user-agent?
Use the --user-agent=YourUserAgent argument when running the server.
How do I disable robots.txt?
How do I disable robots.txt?
Use the --ignore-robots-txt argument when running the server.
What is the default port?
What is the default port?
The default port is 8080.