Health_Assistant_Tracker

A comprehensive personal health dashboard built with ASP.NET Core 9.0, AI-driven insights, and a lot of vibecoding. This application aggregates data from your Garmin devices (via Garmin Connect), visualizes your performance, and uses OpenAI to provide daily coaching recommendations.

Note: Currently, this project supports Garmin devices only.

🎯 Project Vision

The core goal of this project is to evolve the Recommendation Agents into a truly intelligent coaching system. We aim to move beyond simple summaries and create an agent that:

1. Gives better recommendations: Deeper context awareness, less generic advice, and "smarter" coaching.
2. Incorporates Routines: Seamlessly suggests and adds workouts to your routine calendar based on your recovery and goals.

Documentation

• User Guide: Walkthrough of the dashboard features.
• Architecture: Technical deep-dive into the stack, data flow, and components.
• Contributing & Roadmap: How to help out and future ideas.

---

How to Run

You can run this project using Docker (recommended) or locally with .NET.

Prerequisites

• Garmin Connect Account: Email and password.
• OpenAI API Key (Optional): For AI coaching features.

Option A: Docker Compose (Recommended)

This is the easiest way to get started. It builds the container (including .NET and Python dependencies) and starts the server.

1. Clone the repository: \\�ash git clone https://github.com/camiloberutti/Health_Assistant_Tracker.git cd Health_Assistant_Tracker \\

2. Configure Environment: Copy the example file and edit it with your credentials. \\powershell cp .env.example .env

   Edit .env with your GARMIN_USERNAME, GARMIN_PASSWORD, etc.

   \\

3. Run: \\powershell docker-compose up -d \
   The app will be available at http://localhost:5180.

4. Update/Rebuild: If you make code changes, run: \\powershell ./apply-changes.ps1 \\

Option B: Local Development (.NET CLI)

1. Install Dependencies:

   • .NET 9.0 SDK
   • Python 3.10+ (Make sure 'python' is in your PATH)
   • Install the Garmin library: \\�ash pip install garminconnect \\

2. Configure: Ensure your .env\ file is created (see step 2 in Option A). The app reads this file on startup.

3. Run: \\powershell cd GarminTempApi dotnet watch run \\

---

Configuration (.env)

Variable	Description	Default
\GARMIN_USERNAME\	Your Garmin Connect login email (Required)	-
\GARMIN_PASSWORD\	Your Garmin Connect password (Required)	-
\GARMIN_REFRESH_MINUTES\	Activity sync interval (minutes)	\60\
\GARMIN_START_DATE\	Start of history import (YYYY-MM-DD)	-
\GARMIN_END_DATE\	End date (Leave empty to always sync up to today)	-
\CHATGPT_API_KEY\	OpenAI Key for insights (Optional)	-

---

Contributing

We want to make this the best self-hosted Garmin dashboard available. Check out the Contributing Guide for:

• Step-by-step development instructions.
• List of planned features (Mobile support, Dark mode, etc.).
• Architecture details.