MCP SSE Server

🚀 GenAI News Assistant MCP Server

This repository contains a template for building a Model Context Protocol (MCP) server, which provides tools for retrieving and analyzing news information using NewsAPI.org and Large Language Models (LLMs).

🚀 Quick Start

This README provides a guide on how to build and run the GenAI News Assistant MCP Server, including project structure, prerequisites, environment variables, build and run steps, testing, MCP tools, development, and licensing information.

✨ Features

• Multi - language implementation: Available in both Python and TypeScript.
• News retrieval and analysis: Use NewsAPI.org and LLMs to search for news and extract information.
• MCP - compliant: Implements specific MCP tools for news processing.

📦 Installation

Prerequisites

• Docker installed on your system.
• A NewsAPI.org API key.
• An LLM API key (provided separately).

Environment Variables

The following environment variables are required:

• NEWS_API_KEY: Your NewsAPI.org API key.
• LLM_API_KEY: Your LLM API key.

Build and Run

Python Version

# Build the Docker image
cd python
docker build -t news - assistant - python .

# Run the container
docker run -p 3000:3000 \
  -e NEWS_API_KEY=your_news_api_key \
  -e LLM_API_KEY=your_llm_api_key \
  news - assistant - python

TypeScript Version

# Build the Docker image
cd typescript
docker build -t news - assistant - typescript .

# Run the container
docker run -p 3000:3000 \
  -e NEWS_API_KEY=your_news_api_key \
  -e LLM_API_KEY=your_llm_api_key \
  news - assistant - typescript

Testing

You can use the MCP Inspector in a development environment to verify that the server is working as expected: https://github.com/modelcontextprotocol/inspector

💻 Usage Examples

The server should implement the following MCP tools:

1. search_news: Search for recent news articles that match a specific query.
2. extract_information_from_article: Extract structured information from a news article.
3. extract_key_info_and_sentiment: Analyze key entities and sentiment in a news article.

For detailed specifications of each tool, please refer to the task documentation.

📚 Documentation

Project Structure

.
├── python/                # Python implementation
│   ├── Dockerfile         # Python Docker configuration
│   └── src/               # Python source code
├── typescript/            # TypeScript implementation
│   ├── Dockerfile         # TypeScript Docker configuration
│   └── src/               # TypeScript source code
└── README.md              # This file

Development

Choose either the Python or TypeScript implementation based on your preference. Both implementations provide the same functionality through the MCP interface.

📄 License

This repository is for internal testing tasks and private use only.