MCP Server Fetch Python

🚀 mcp-server-fetch-python

An MCP server for scraping and converting web content into various formats. This server offers comprehensive tools to extract content from web pages, including support for JavaScript-rendered content and media files.

✨ Features

Tools

This server provides four professional tools:

• get-raw-text: Extract raw text content directly from a URL without browser rendering.

  • Parameters:
    • url: The URL of the target web page (required).
  • Best suited for handling structured data formats or situations requiring quick access.

• get-rendered-html: Obtain fully rendered HTML content using a headless browser.

  • Parameters:
    • url: The URL of the target web page (required).
  • Essential for modern web applications and SPAs.

• get-markdown: Convert web page content into nicely formatted Markdown.

  • Parameters:
    • url: The URL of the target web page (required).
  • Provides clean and readable text output while preserving structural elements.

• get-markdown-from-media: Perform AI-driven content extraction on media files.

  • Parameters:
    • url: The URL of the target media file (image, video) (required).
  • Utilizes computer vision and OCR technologies for content analysis.
  • Requires setting a valid OPENAI_API_KEY environment variable.
  • If the API key is not set or there are issues processing media files, an error message will be returned.

💻 Usage Examples

Claude Desktop

To use it on Claude Desktop, add the server configuration:

On MacOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json
On Windows: %APPDATA%/Claude/claude_desktop_config.json

"mcpServers": {
  "mcp-server-fetch-python": {
    "command": "uvx",
    "args": [
      "mcp-server-fetch-python"
    ]
  }
}

📚 Documentation

Environment Variables

The following environment variables can be configured:

• OPENAI_API_KEY: Required for the get-markdown-from-media tool. A valid OpenAI API key must be set.
• PYTHONIOENCODING: Set it to "utf-8" to solve character encoding issues in the output.
• MODEL_NAME: Specify the name of the model to use. The default is "gpt-4o".

"mcpServers": {
  "mcp-server-fetch-python": {
    "command": "uvx",
    "args": [
      "mcp-server-fetch-python"
    ],
    "env": {
        "OPENAI_API_KEY": "sk-****",
        "PYTHONIOENCODING": "utf-8",
        "MODEL_NAME": "gpt-4o",        
    }
  }
}

Local Installation

Alternatively, you can install and run the server locally:

git clone https://github.com/tatn/mcp-server-fetch-python.git
cd mcp-server-fetch-python
uv sync
uv build

Then add the following configuration to the Claude Desktop configuration file:

"mcpServers": {
  "mcp-server-fetch-python": {
    "command": "uv",
    "args": [
      "--directory",
      "path\\to\\mcp-server-fetch-python",  # Replace with the actual cloned repository path
      "run",
      "mcp-server-fetch-python"
    ]
  }
}

🔧 Technical Details

Debugging

You can use npx to start the MCP Inspector:

npx @modelcontextprotocol/inspector uvx mcp-server-fetch-python

npx @modelcontextprotocol/inspector uv --directory path\\to\\mcp-server-fetch-python run mcp-server-fetch-python