health_ops

KO | EN

gitlite — search

#python #react #ai #typescript #llm #go #android #nodejs #java #awesome #awesome-list #security #ios #openai

(★ 1)

Personal health ops: automated Garmin Smartwatches data collection, visualisation, and AI-powered insights

health_ops 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# HealthOps

Personal health data analysis powered by Garmin Connect data + OpenAI

## Overview

HealthOps is a personal health data management system that:
- Syncs your Garmin Connect data automatically (vitals, sleep, body battery, stress)
- Stores data securely in DuckDB
- Provides interactive visualizations through a Streamlit dashboard
- Offers AI-powered coaching insights using OpenAI

## Architecture

- **Vault**: HashiCorp Vault for secure credential storage
- **Garmin Sync**: Background worker that syncs data hourly
- **Dashboard**: Streamlit web interface with metrics and AI coach

## Quick Start

```bash
./run.sh       # First run: initializes Vault and prompts for credentials
./run.sh       # Subsequent runs: starts all services
./run.sh -r    # Full reset (wipes all data)
```

Access the dashboard at: http://localhost:8888

## Features

- **Automated Sync**: Hourly sync of current day data + 30-day gap-fill on startup
- **Weekly Metrics**: Steps, sleep, body battery, stress averages
- **Interactive Charts**: Sleep score, stress levels, body battery with adjustable time ranges
- **AI Coach**: On-demand readiness analysis powered by OpenAI GPT-5
- **Secure Storage**: All credentials stored in Vault with minimal-privilege app tokens

## Requirements

- Docker & Docker Compose
- Garmin Connect account
- OpenAI API key

## Data Privacy

All data stays local:
- Garmin data synced directly to your local DuckDB
- Vault runs locally in Docker
- Only API calls go to Garmin Connect and OpenAI

health_ops

health_ops 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

health_ops

health_ops 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README