Geeknews

🚀 GeekNews MCP Server

This project is a Model Context Protocol (MCP) server that fetches articles from GeekNews. Implemented in Python, it uses BeautifulSoup for web scraping. To reduce server load, it offers a feature to automatically parse data and store it in the cache daily.

✨ Features

Tools

• get_articles Tool: Fetches articles from GeekNews.

  • Allows specifying the article type (top, new, ask, show) and the number of articles to return.
  • Each response includes information such as title, URL, points, author, time, number of comments, and ranking.
  • Uses cached data to reduce server load.

• get_weekly_news Tool: Retrieves weekly news from GeekNews.

  • Allows specifying a specific weekly news ID or getting the latest weekly news.
  • Provides information about the weekly news, including title, number, ID, content, URL, and project list.
  • Each project includes information such as title, URL, and ranking.
  • Uses cached data to reduce server load.

Caching Feature

• Automatically parses data and stores it in the cache once a day.
• Uses cached data within the validity period without sending requests to the GeekNews server.
• The cache data is valid for 24 hours and will be automatically updated after that.
• A timer periodically checks the cache's validity and updates it if necessary.

📦 Installation

Install via Smithery

🚀 geeknews-mcp-server

Add Server Information to MCP Settings File

{
    "mcp_settings": {
        "servers": {
            "geeknews_mcp_server": {
                "url": "http://localhost:5000",
                "api_key": "your_api_key"
            }
        }
    }
}

Local Installation Guide

1. Clone the Repository

git clone https://github.com/yourusername/geeknews-mcp-server.git
cd geeknews-mcp-server

2. Configure the Environment Install dependencies:

pip install -r requirements.txt

3. Run the Server

python server.py

🔧 Technical Details

Code Structure

• server.py: The main server program that handles requests and responses.
• crawler.py: A crawler module responsible for scraping data from GeekNews.
• cache_manager.py: A module that manages cached data, including reading, writing, and updating operations.
• tools: A directory containing various auxiliary tool functions.

Important Note

⚠️ Important Note

This project relies on the HTML structure of GeekNews. If the structure changes, the crawler logic may need to be adjusted.

💡 Usage Tip

The cached data is stored in the .cache/geeknews-mcp folder in the user's home directory.