Fetch MCP Server

What is Fetch MCP Server?

Fetch MCP Server is a Model Context Protocol server that fetches web content and returns it in multiple formats: HTML, plain text, JSON, and Markdown. It is designed for AI agents or applications that need to retrieve and transform web content on demand.

How to use Fetch MCP Server?

Clone the repository, install dependencies with npm install, and build with npm run build. Start the server directly with npm start (runs on stdio). For desktop app integration, add the provided JSON configuration to the app’s MCP server settings, pointing to the compiled output file.

Key features of Fetch MCP Server

• Fetches web content using the modern fetch API
• Supports custom HTTP headers for requests
• Returns content as HTML, JSON, plain text, or Markdown
• Uses JSDOM for HTML parsing and text extraction
• Uses TurndownService for HTML to Markdown conversion
• No persistent resources – fetches and transforms on demand

Use cases of Fetch MCP Server

• Retrieve a webpage’s raw HTML for analysis or scraping
• Fetch JSON data from an API and make it available to an AI assistant
• Extract plain text from a webpage for summarization or processing
• Convert web content to Markdown for use in documentation or notes
• Provide dynamic web content to an MCP client without storing it

FAQ from Fetch MCP Server

What output formats does Fetch MCP Server support?

It supports four formats: HTML, JSON, plain text (HTML tags removed), and Markdown.

Can I pass custom HTTP headers when fetching?

Yes, each tool accepts an optional headers object to include custom headers in the request.

How do I integrate Fetch MCP Server with a desktop app?

Add the server configuration to the app’s MCP settings using the command node and the path to dist/index.js.

Does Fetch MCP Server store any data?

No. The server does not provide any persistent resources; it fetches and transforms content on demand.

What dependencies does Fetch MCP Server require?

The server requires Node.js and npm. It uses JSDOM for HTML parsing and TurndownService for Markdown conversion.