Jmh108 MCP Server Readability Python

🚀 Mozilla Readability Parser MCP Server

A Python-based implementation of the Model Context Protocol (MCP) server that extracts and converts web page content into clean Markdown format suitable for processing by large language models.

📑 Table of Contents

• ✨ Features
• ❓ Why Not Just Scrape?
• 📦 Installation
• 🚀 Quick Start
• 📚 Tool Reference
• 📋 Dependencies
• 📄 License

✨ Features

• Remove ads, navigation bars, footers, and other non-essential content.
• Convert clean HTML into well-formatted Markdown.
• Handle errors gracefully.
• Optimize for large language model (LLM) processing.
• Lightweight and fast.

❓ Why Not Just Scrape?

Unlike simple scraping requests, this server:

• Uses a readability algorithm to extract only relevant content.
• Eliminates noise such as ads, pop-ups, and navigation menus.
• Reduces token usage by removing unnecessary HTML/CSS.
• Provides a consistent Markdown format to enhance LLM processing.
• Handles complex web pages with dynamic content.

📦 Installation

1. Clone the repository:

git clone https://github.com/jmh108/MCP-server-readability-python.git
cd MCP-server-readability-python

2. Create and activate a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows, use: venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

🚀 Quick Start

1. Start the server:

fastmcp run server.py

2. Example request:

curl -X POST http://localhost:8000/tools/extract_content \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/article"}'

📚 Tool Reference

extract_content

Scrape and convert web page content into clean Markdown.

Parameters:

{
  "url": {
    "Type": "string",
    "Description": "The URL of the website to parse",
    "Required": true
  }
}

Return value:

{
  "content": "Markdown content..."
}

⚙️ MCP Server Configuration

To configure this server in the MCP settings file, add the following:

{
  "mcpServers": {
    "readability": {
      "command": "fastmcp",
      "args": ["run", "server.py"],
      "env": {}
    }
  }
}

Then you can start the server via the MCP protocol and access it using the parse tool.

📋 Dependencies

• readability-lxml - Content extraction
• html2text - HTML to Markdown conversion
• beautifulsoup4 - DOM parsing
• requests - HTTP requests

📄 License

MIT License - See LICENSE for details.