neiltron/apple-health-mcp

Local-first Apple Health MCP server. Lets AI assistants answer questions about your sleep, workouts, activity and heart data from a local export.

What it solves

This project allows AI assistants (MCP clients) to securely query and analyze personal Apple Health data. It bridges the gap between static CSV exports of health metrics and LLMs, enabling users to get health summaries and run analytical queries on their own data without uploading the files to a cloud service.

How it works

The server acts as a Model Context Protocol (MCP) server that runs locally. It uses DuckDB to read CSV exports generated by the "Simple Health Export CSV" app. When an AI client requests data, the server loads the relevant CSV files into an in-memory database, allowing the LLM to perform SQL-based analysis and generate reports based on the user's local health records.

Who it’s for

Individuals who want to use AI to analyze their Apple Health trends and summaries while keeping their raw data stored locally on their machine.

Highlights

  • Local Processing: Reads files in place and does not make network requests to upload raw data.
  • SQL-Powered Analysis: Uses DuckDB for efficient analytical queries via the health_query tool.
  • Schema Discovery: Includes a health_schema tool so the AI can understand the available tables, columns, and units.
  • Automated Reporting: Provides a health_report tool for generating weekly, monthly, or custom health summaries.

Related

  • Project
  • Project
  • Project
  • Project
  • Project