mcp-url2markdown

What is mcp-url2markdown?

mcp-url2markdown is an MCP server that converts webpage content into clean, formatted Markdown using the crawl4ai library. It is designed for developers who need to extract and reformat web page text for use in AI coding assistants or other tools.

How to use mcp-url2markdown?

Clone the repository and run uv sync to install dependencies. Start the server with fastmcp dev server.py to run in development mode on http://127.0.0.1:6274, or configure it as an MCP tool in Cursor, cline, or roo code using the provided JSON config.

Key features of mcp-url2markdown

• Converts web pages to clean, structured Markdown.
• Powered by the crawl4ai library.
• Works with MCP-compatible AI coding tools.
• Easy local setup with uv and FastMCP.
• Provides a development server for testing.

Use cases of mcp-url2markdown

• Convert API documentation pages into Markdown for reference.
• Extract content from any URL for use in LLM prompts.
• Integrate with AI coding assistants to fetch and format web data.

FAQ from mcp-url2markdown

What does mcp-url2markdown do?

It takes a URL, fetches the webpage, and returns the main content as clean, formatted Markdown.

How do I run the server locally?

Run fastmcp dev server.py after installing dependencies. The server will be available at http://127.0.0.1:6274.

What are the runtime requirements?

The server requires Python, uv, and the crawl4ai library, which are installed via uv sync. It uses FastMCP for the MCP interface.

Can I use mcp-url2markdown with Cursor or roo code?

Yes. Configure it by adding the provided JSON block to your MCP settings, specifying the correct path to the cloned repository.

Does the server require authentication or network access?

No authentication is mentioned. The server fetches public URLs, so a working internet connection is needed to access web pages.