statsig-mcp

# Statsig MCP Server Model Context Protocol server for [Statsig](https://statsig.com) feature flags and experimentation platform. This MCP server enables AI assistants to interact with Statsig's feature management and experimentation platform, allowing them to check feature flags, retrieve dynamic configurations, get experiment assignments, and log custom events. ## Features - 🚩 **Feature Gate Checking**: Determine if features are enabled for users - ⚙️ **Dynamic Configuration**: Retrieve configuration values based on user context - 🧪 **Experiment Assignment**: Get experiment variations for A/B testing - 📊 **Layer Parameter Access**: Retrieve values from Statsig layers - 📈 **Event Logging**: Track custom events for analytics ## Installation ### Using uv (Recommended) ```bash # Clone the repository git clone https://github.com/GeLi2001/statsig-mcp.git cd statsig-mcp # Install dependencies with uv uv sync # Install with dev dependencies uv sync --extra dev ``` ### Using pip ```bash # Install from source pip install -e . # Or install with dev dependencies pip install -e ".[dev]" ``` ## Quick Start ### 1. Run the Server #### With uv (Recommended): ```bash # Using command-line arguments (MCP best practice) uv run -m statsig_mcp --api-key "your-secret-key" # With additional options uv run -m statsig_mcp \ --api-key "your-secret-key" \ --environment "production" \ --api-timeout 5000 \ --debug # Or with environment variables STATSIG_SERVER_SECRET_KEY="your-key" uv run -m statsig_mcp ``` #### With Python: ```bash # Using flags (recommended) python -m statsig_mcp --api-key "your-secret-key" # Or environment variables export STATSIG_SERVER_SECRET_KEY="your-secret-key" python -m statsig_mcp ``` ### 2. Configuration Options The server supports both **command-line arguments** (recommended for MCP) and **environment variables**: #### Command-Line Arguments (MCP Best Practice) ```bash uv run -m statsig_mcp --help ``` | Argument | Ty…