Frequently Asked Questions about the Statsource MCP Server

Q: What is an MCP Server? A: MCP (Model Context Protocol) Server acts as a bridge, allowing AI models to access and interact with external data sources and tools. It standardizes how applications provide context to LLMs.

Q: What is the Statsource MCP Server? A: The Statsource MCP Server is a specific implementation of an MCP Server that provides statistical analysis capabilities to LLMs. It allows LLMs to analyze data, calculate statistics, and generate predictions.

Q: What data sources does the Statsource MCP Server support? A: The Statsource MCP Server supports CSV files (uploaded to statsource.me), PostgreSQL databases (via connection strings), and external APIs.

Q: What statistical calculations can the Statsource MCP Server perform? A: The server can perform a wide range of statistical calculations, including mean, median, standard deviation, sum, count, min, max, describe, correlation, missing values, unique values, and boxplots.

Q: Can the Statsource MCP Server generate machine learning predictions? A: Yes, the server can generate ML predictions based on historical data.

Q: How do I install the Statsource MCP Server? A: You can install the server using uv (recommended), Docker, or PIP. Detailed instructions are provided in the documentation.

Q: How do I configure the Statsource MCP Server for Claude.app? A: You need to configure the mcpServers settings in your Claude configuration file. Example configurations for different installation methods are provided in the documentation, along with information on configuring environment variables.

Q: What environment variables do I need to configure? A: You may need to configure environment variables such as API_KEY, DB_CONNECTION_STRING, and DB_SOURCE_TYPE, depending on your configuration.

Q: How do I debug the Statsource MCP Server? A: You can use the MCP inspector to debug the server. Instructions are provided in the documentation.

Q: Can I contribute to the Statsource MCP Server? A: Yes, contributions are welcome! The project is open-source and licensed under the MIT License.