MCP News Scraper

A Model Control Protocol (MCP) server that provides news scraping capabilities for Khaleej Times. This service allows you to fetch headlines and article content programmatically.

Features

Get the latest headline from Khaleej Times
Get all main headlines from Khaleej Times
Get full article content including title, paragraphs, author, and date

Prerequisites

Python 3.8 or higher
uv (Python package installer)
typer (for CLI functionality)

Installation

Clone the repository:

git clone <your-repo-url>
cd mcp-news

Install dependencies using uv:

uv run mcp install main.py

If you encounter any typer-related errors, install it separately:

pip install --upgrade typer

Running the Server

Initialize the MCP server:

uv init mcp-server-demo
cd mcp-server-demo

Start the server:

python main.py

Available Tools

The server provides the following tools:

get_khaleej_times(): Returns the latest headline from Khaleej Times
get_khaleej_times_all(): Returns all main headlines from Khaleej Times
get_khaleej_times_article(url): Returns the full content of an article from its URL

Usage with Claude Desktop

Make sure the MCP server is running
In Claude Desktop, you can interact with the news scraper using the available tools
Example queries:
- "Get me the latest headline from Khaleej Times"
- "Show me all headlines from Khaleej Times"
- "Get the content of this article: <article_name>"

Project Structure

mcp-news/
├── main.py              # Main server file with MCP tools
├── news_sources/        # Directory containing news source implementations
│   └── khaleej_times.py # Khaleej Times specific scraping logic
└── README.md           # This file

Contributing

Feel free to submit issues and enhancement requests!

License

MIT License