Statsource MCP Server

Example Use Cases

• 1

  Compute summary statistics, correlations, and visual outputs (like boxplots) from uploaded CSVs or database tables.

• 2

  Generate short-term ML forecasts by specifying a date column and number of future periods on time-series data.

• 3

  Filter, group, and slice datasets (by date ranges, categorical filters, or groupby keys) before running analyses.

Description

Statsource MCP Server is a Model Context Protocol (MCP) server that enables LLMs to connect to CSV files, databases (e.g., PostgreSQL), or APIs to perform statistical analysis and make ML predictions. It exposes tools like get_statistics for computing summary metrics, correlations, boxplots, and forecasts, and suggest_feature for proposing product improvements. The server is configurable via environment variables (API_KEY, DB_CONNECTION_STRING, DB_SOURCE_TYPE) and supports Docker and pip installation for easy integration into agent workflows.