mcp-server-fetch-python

What is mcp-server-fetch-python?

mcp-server-fetch-python is an MCP server that fetches and transforms web content into various formats. It provides tools for extracting raw text, rendered HTML, Markdown, and AI-powered content from media files, supporting JavaScript-rendered pages and visual media analysis.

How to use mcp-server-fetch-python?

Install using uvx mcp-server-fetch-python or clone the repository and run locally with uv sync and uv build. Add the server configuration to your Claude Desktop configuration file (macOS: ~/Library/Application Support/Claude/claude_desktop_config.json; Windows: %APPDATA%/Claude/claude_desktop_config.json) using the uvx command. For the media analysis tool, set the OPENAI_API_KEY environment variable; optionally set MODEL_NAME (default: "gpt-4o") and PYTHONIOENCODING to "utf-8".

Key features of mcp-server-fetch-python

• get-raw-text: Extracts raw text directly from URLs without browser rendering.
• get-rendered-html: Fetches fully rendered HTML using a headless browser.
• get-markdown: Converts web page content to well-formatted Markdown.
• get-markdown-from-media: AI-powered content extraction from images and videos.
• Media tool requires OPENAI_API_KEY environment variable.

Use cases of mcp-server-fetch-python

• Scrape structured data (JSON, XML, CSV) from URLs quickly.
• Extract content from modern web applications and SPAs that require JavaScript rendering.
• Convert web pages to clean Markdown for documentation or note-taking.
• Perform text extraction from images and videos via OpenAI computer vision.
• Integrate content fetching into AI assistants via the MCP protocol.

FAQ from mcp-server-fetch-python

What tools does mcp-server-fetch-python provide?

It provides four tools: get-raw-text, get-rendered-html, get-markdown, and get-markdown-from-media.

Is an OpenAI API key required?

No, only the get-markdown-from-media tool requires an OPENAI_API_KEY. The other three