Leo-Health-Core: Import Health Exports into Local SQLite

Overview

Leo-Health-Core is a Python tool for combining Apple Health XML exports and Whoop CSV exports in a local SQLite database. It addresses the difficulty of working with large, differently structured health exports by parsing and normalizing common metrics for SQL queries and a local dashboard.

It is designed for privacy-conscious users comfortable with a command line. Data remains on the user's machine according to the project description; the dashboard is intended for localhost and single-user use.

Key Features

• Parses Apple Health exports from export.zip, including heart rate, HRV, sleep, workouts, and blood oxygen.
• Imports Whoop CSV data for recovery, strain, and sleep.
• Stores normalized records in SQLite for queries across sources.
• Includes terminal and browser-based dashboard commands.
• Watches a folder for new exports and can ingest them automatically.
• Provides Docker support for running the dashboard across operating systems.
• Uses Python's standard library, with the project describing the tool as having zero dependencies.

Use Cases

• Apple Health users who want to query exported measurements with SQLite instead of browsing a large XML export.
• Whoop users who want recovery, strain, and sleep records in a local database.
• Users of both services who want to compare selected metrics across data sources.
• Developers who want a local foundation for analyzing wearable data or building additional integrations.
• Privacy-conscious users who prefer local files and tools over uploading personal health data to a hosted service.

Project Facts

• Language: Python
• License: MIT
• Stars: 94
• Forks: 9
• Archived: No

Getting Started

git clone https://github.com/sandseb123/Leo-Health-Core.git
cd Leo-Health-Core
pip3 install -e .

See the README for platform-specific setup, Docker instructions, supported exports, and usage details.

Alternatives

• fasten-onprem: Fasten OnPrem organizes medical records from FHIR uploads and manual entry, while Leo-Health-Core imports wearable exports into a normalized SQLite database.

Considerations

• The README lists Python 3.9 or newer for direct installation; Docker is offered as an alternative.
• The dashboard has no authentication and is intended for single-user desktop use. Anyone with access to the machine can read the database.
• Ingestion depends on the formats supported by the current parsers. Oura, Fitbit, and Garmin support are listed as roadmap items, not current features.
• The README describes the AI health coach as coming soon; it is not part of the currently described core workflow.
• The repository reports 10 open issues, so check its issue tracker for known gaps before relying on it for a particular export.