McpWithDeepseek

# DeepSeek Client for Model Context Protocol (MCP) This project is a customized Python client based on the official [Model Context Protocol (MCP)](https://modelcontextprotocol.io/quickstart/client#python) quickstart guide. It replaces the default Claude LLM model with the [DeepSeek LLM](https://deepseek.com/) by updating the API configuration in `client.py`. ## 🚀 Quick Start ### 1. Clone the Repository ```bash git clone https://github.com/your-username/deepseek-mcp-client.git cd deepseek-mcp-client ``` ### 2. Set Up Environment It's recommended to use a virtual environment: ```bash uv venv source .venv/bin/activate # On Windows: .venv\Scripts\activate ``` Then install dependencies: ```bash uv add mcp anthropic python-dotenv ``` ### 3. Configure API Keys Make sure you have an API key for the DeepSeek LLM. ``` DEEPSEEK_API_KEY=your_deepseek_api_key ``` ### 4. Run the Client ```bash uv run client.py ../MCP-SERVER-DEMO/weather.py ``` #### The weather.py is same with https://github.com/modelcontextprotocol/quickstart-resources/blob/main/weather-server-python/weather.py and https://modelcontextprotocol.io/quickstart/server ## 🔧 What's Modified? Only `client.py` is changed from the original MCP implementation: - Replaced Claude LLM API with DeepSeek LLM API. - Preserved the original Model Context Protocol logic and structure. ## 📚 References - Official MCP Documentation: [https://modelcontextprotocol.io/](https://modelcontextprotocol.io/) - DeepSeek: [https://deepseek.com/](https://deepseek.com/) --- Feel free to customize the `client.py` further to suit your use case. ## 📝 License MIT License