Skip to main content

​
Prerequisites

Before you begin, ensure you have the following installed:

Docker

Version 20.10 or higher

Docker Compose

Version 1.29 or higher
You can verify your Docker installation by running docker --version and docker-compose --version

​
Installation Steps

1

Clone the repository

Clone the Mirobody Health repository to your local machine:
git clone https://github.com/thetahealth/mirobody-health.git
cd mirobody-health
2

Set up configuration files

Create your configuration files from the provided examples:
cp .env_example .env
cp config.example.yaml config.yaml
For a quick test, you can use the default configuration. For production, see the Configuration guide.
3

Add API keys

Edit config.yaml to add your API keys. Both OPENROUTER_API_KEY and GOOGLE_API_KEY are required:
config.yaml
# AI Provider API Keys - REQUIRED
OPENROUTER_API_KEY: 'your_openrouter_api_key'  # REQUIRED for AI chat/agents
GOOGLE_API_KEY: 'your_google_api_key'  # REQUIRED for file processing (Gemini)

# Optional: Additional AI providers
OPENAI_API_KEY: 'your_openai_api_key'
OPENROUTER_API_KEY is required for AI chat and agent features. Get your API key at OpenRouter.GOOGLE_API_KEY (Gemini) is required for processing files like PDFs, images, and medical records. Get your API key at Google AI Studio.
OpenRouter provides access to multiple LLM providers through a single API, while Google Gemini handles multimodal file processing.
4

Deploy with Docker

Run the deployment script to start all services:
./deploy.sh
This will start three services:
  • Backend API on port 18080
  • PostgreSQL database on port 5432
  • Redis cache on port 6379
Wait for all services to start. You should see logs indicating successful startup.
5

Verify the installation

Open your browser and navigate to:
http://localhost:18080
You should see the Mirobody Health welcome page.
6

Log in

Log in with the default demo account or the accounts set up in config.yaml.
[email protected]
777777
Once logged in, you can:
  • Upload files: Lab reports, medical records, health documents (see File Processing)
  • Chat with AI: Ask questions about your health data
  • Connect providers: Link Garmin, Whoop, and other devices (see Using Providers)

​
Test the API

Once the services are running, you can test the API endpoints:
Verify the server is running:
curl http://localhost:18080/health

​
Available Endpoints

After deployment, the following endpoints are available:

Web Interface

Main web interface and documentation

MCP Protocol

JSON-RPC 2.0 interface for AI agents

Chat API

AI chat interface endpoint

History

Session history endpoint

​
Docker Services

The deployment includes three containerized services:
ServicePortDescription
backend18080FastAPI application server
postgres5432PostgreSQL database
redis6379Redis cache and session store
You can view logs for any service using:
docker-compose logs -f backend
docker-compose logs -f postgres
docker-compose logs -f redis

​
Common Commands

docker-compose down

docker-compose restart

# All services
docker-compose logs -f

# Specific service
docker-compose logs -f backend

docker-compose down
docker-compose build --no-cache
docker-compose up -d

docker-compose exec postgres psql -U holistic_user -d holistic_db

docker-compose exec redis redis-cli

​
Troubleshooting

If port 18080 is already in use, edit config.yaml:
HTTP_PORT: 18081  # Change to an available port
Then restart the services.
Ensure PostgreSQL container is running:
docker-compose ps postgres
Check database logs:
docker-compose logs postgres
Verify Redis is running and accessible:
docker-compose exec redis redis-cli ping
Should return PONG if working correctly.
Make the script executable:
chmod +x deploy.sh

​
Next Steps

Now that you have Mirobody Health running:

Configure Providers

Set up OAuth credentials for Garmin, Whoop, and other providers

Explore the API

Learn how to use the Pulse API to link users and retrieve health data

Understand Architecture

Learn how Mirobody Health processes and standardizes health data

Add New Providers

Follow the guide to integrate additional health device providers
For production deployment, see the Production Deployment guide for security hardening and performance optimization.