This project provides an MCP server for querying Apple Health data using natural language and SQL, targeting users who want to analyze their health metrics efficiently. It leverages DuckDB for fast data processing and offers features like automated reports and smart caching, simplifying health data exploration.

How It Works

The server acts as an intermediary, translating natural language queries into SQL statements executed against a DuckDB database populated from Apple Health CSV exports. This approach allows for direct, efficient analysis of health data without requiring complex data wrangling or specialized knowledge, making it accessible to a broader user base.

Quick Start & Requirements

• Install/Run: Use directly via npx @neiltron/apple-health-mcp.
• Prerequisites: Requires Apple Health data exported as CSV files using the "Simple Health Export CSV" iOS app. The HEALTH_DATA_DIR environment variable must be set to the path of the unzipped CSV export.
• Configuration: Add to Claude Desktop's claude_desktop_config.json with the HEALTH_DATA_DIR environment variable specified.
• Docs: Simple Health Export CSV app

Highlighted Details

• Supports natural language querying via MCP clients.
• Direct SQL query execution against health data.
• Automated generation of weekly/monthly health summaries.
• Efficient data loading with configurable time windows and query result caching.

Maintenance & Community

The project is maintained by neiltron. Contributions are welcome, with guidelines for code patterns, TypeScript types, error handling, and performance.

Licensing & Compatibility

Licensed under the MIT license, permitting commercial use and integration with closed-source applications.

Limitations & Caveats

The project relies on a specific CSV export format from a third-party iOS app. Performance may be impacted by large datasets or inefficient queries, with suggestions to reduce memory usage or filter by date ranges.