MCP Server for NBA Stats Predictor Application

An MCP-powered tool for the NBA stats predictor app that generates player performance forecasts using real-time data analysis and advanced statistical modeling.

Installation

Prerequisites

• Python 3.8+
• pip
• Claude Desktop

Step-by-Step Setup

1. Clone this repository onto your local device

2. Navigate to the project directory:

   cd nba-stats-predictor-application
   

   Copy

3. Create a virtual environment:

   python3 -m venv venv
   

   Copy

4. Activate the virtual environment:

   source venv/bin/activate
   

   Copy

5. Install dependencies:

   pip install -r requirements.txt
   

   Copy

6. Download the necessary data:

   python3 data_pipeline/download_data.py
   

   Copy

7. Train the prediction model:

   python3 models/train_model.py
   

   Copy

8. Start the FastAPI server:

   uvicorn api.fastapi_server:app --reload
   

   Copy

9. Open a new terminal

10. Return to the project directory

11. Install UV package manager:

    curl -LsSf https://astral.sh/uv/install.sh | sh
    

    Copy

12. Restart the terminal in this directory

13. Run the MCP server:

    uv run mcp_main.py
    

    Copy

14. Open another new terminal

15. Configure Claude Desktop:

    code ~/Library/Application\ Support/Claude/claude_desktop_config.json
    

    Copy

    Note: If the file doesn't exist, create it.

16. Add the following configuration to claude_desktop_config.json:

    {
        "mcpServers": {
            "NBA-stats-predictor": {
                "command": "/PATH/TO/PROJECT/DIRECTORY/.venv/bin/uv",
                "args": [
                    "--directory",
                    "/PATH/TO/PROJECT/DIRECTORY/",
                    "run",
                    "mcp_main.py"
                ]
            }
        }
    }
    

    Copy

    Remember to replace /PATH/TO/PROJECT/DIRECTORY/ with the actual path to your project.

17. You should now be able to use this MCP tool on Claude Desktop.

Usage

Once configured, you can use the NBA stats predictor tool in Claude Desktop to get predictions for player performance in upcoming games.

Troubleshooting

• Make sure all paths in the configuration are correct
• Ensure the virtual environment is activated before running commands
• Check that all dependencies are properly installed
• Verify that the FastAPI server is running before using the MCP tool