Oura MCP Server (Python)
Enables AI assistants to access Oura Ring health data including sleep, readiness, activity, heart rate, and workouts through MCP tools with OAuth2 authentication.
README
Oura MCP Server (Python)
A Python implementation of a Model Context Protocol (MCP) server for accessing Oura Ring health data with OAuth2 authentication.
Features
- OAuth2 Authentication - Secure authentication with automatic token refresh
- 9 MCP Tools - Access sleep, readiness, activity, heart rate, workouts, and more
- Token Encryption - AES-256 encryption for OAuth tokens at rest
- Health Insights - Aggregated health trend analysis
- Pure Python - Built with FastMCP and Starlette
Prerequisites
- Python 3.11 or higher
- Oura Ring with an active Oura account
- Oura API access (OAuth application)
Installation
Option 1: Install from source
# Clone or download this repository
cd oura-mcp-python
# Create virtual environment (recommended)
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install the package
pip install -e .
Option 2: Install with uv (faster)
cd oura-mcp-python
uv pip install -e .
Configuration
Step 1: Register an Oura OAuth Application
- Go to Oura OAuth Applications
- Log in with your Oura account
- Click "New Application"
- Fill in:
- Application Name:
Personal MCP Server - Redirect URI:
http://localhost:8000/oauth/callback - Scopes: Select all available scopes
- Application Name:
- Save and copy your Client ID and Client Secret
Step 2: Create Environment File
cp .env.example .env # copy the template
# then edit .env with your credentials
Note:
.envis gitignored;.env.exampleis the committed template.
Step 3: Generate Encryption Key
# Run this in Python to generate TOKEN_ENCRYPTION_KEY
import secrets
print("TOKEN_ENCRYPTION_KEY:", secrets.token_hex(32))
Step 4: Configure .env
Edit your .env file:
# Oura OAuth Credentials
OURA_CLIENT_ID=your_client_id_here
OURA_CLIENT_SECRET=your_client_secret_here
OURA_REDIRECT_URI=http://localhost:8000/oauth/callback
# Token Encryption
TOKEN_ENCRYPTION_KEY=<generated_encryption_key>
# Server Configuration (optional; defaults shown)
HOST=0.0.0.0
PORT=8000
Running the Server
Start the Server
# Using the installed command
oura-mcp
# Or directly with Python
python -m oura_mcp.server
# Or with uvicorn for development
uvicorn oura_mcp.server:create_app --factory --reload --port 8000
Connect Your Oura Account
- Open your browser to:
http://localhost:8000/oauth/authorize - Log in to your Oura account
- Approve the requested permissions
- You'll see a success message when connected
Verify Connection
Check the health endpoint:
curl http://localhost:8000/health
Expected response:
{
"status": "healthy",
"oauth_connected": true,
"server": "Oura MCP Server (Python)",
"version": "1.0.0"
}
Connecting to Claude Desktop
Add this to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"oura": {
"command": "npx",
"args": ["-y", "mcp-remote", "http://localhost:8000/mcp"],
"env": {
"API_KEY": "YOUR_AUTH_TOKEN_HERE"
}
}
}
}
Alternative: Direct stdio connection (if running locally)
{
"mcpServers": {
"oura": {
"command": "python",
"args": ["-m", "oura_mcp.server"],
"cwd": "/path/to/oura-mcp-python",
"env": {
"OURA_CLIENT_ID": "your_client_id",
"OURA_CLIENT_SECRET": "your_client_secret",
"OURA_REDIRECT_URI": "http://localhost:8000/oauth/callback"
}
}
}
}
Restart Claude Desktop after saving.
Available MCP Tools
1. get_personal_info
Get user's personal information and ring details.
2. get_sleep_summary
Get sleep data for a date range.
Parameters:
start_date(required): YYYY-MM-DDend_date(optional): YYYY-MM-DDinclude_detailed(optional): Include detailed sleep period data
3. get_readiness_score
Get daily readiness scores.
Parameters:
start_date(required): YYYY-MM-DDend_date(optional): YYYY-MM-DD
4. get_activity_summary
Get activity data for a date range.
Parameters:
start_date(required): YYYY-MM-DDend_date(optional): YYYY-MM-DD
5. get_heart_rate
Get heart rate data in 5-minute intervals.
Parameters:
start_datetime(required): ISO 8601 format (e.g., 2024-01-01T00:00:00)end_datetime(optional): ISO 8601 format
6. get_workouts
Get workout sessions.
Parameters:
start_date(required): YYYY-MM-DDend_date(optional): YYYY-MM-DD
7. get_tags
Get user-created tags and notes.
Parameters:
start_date(required): YYYY-MM-DDend_date(optional): YYYY-MM-DD
8. get_sessions
Get guided and unguided session data.
Parameters:
start_date(required): YYYY-MM-DDend_date(optional): YYYY-MM-DD
9. get_health_insights
Get aggregated health insights based on recent data.
Parameters:
days(optional): Number of days to analyze (default: 7)
10. get_connection_status
Check if the Oura account is connected and tokens are valid.
API Endpoints
Health Check
GET /health
OAuth Endpoints
GET /oauth/authorize - Start OAuth flow
GET /oauth/callback - OAuth callback (automatic)
GET /oauth/status - Get connection status
POST /oauth/disconnect - Disconnect and clear tokens
MCP Endpoint
POST /mcp - MCP JSON-RPC endpoint (Streamable HTTP)
Deployment Options
Option 1: Local Development
Run on your machine with the server accessible at localhost.
Option 2: Railway / Cloud Deployment
- Push to a Git repository
- Deploy to Railway, Render, or similar
- Update
OURA_REDIRECT_URIto your deployed URL - Update your Oura OAuth application's redirect URI
Option 3: Docker
FROM python:3.11-slim
WORKDIR /app
COPY . .
RUN pip install -e .
EXPOSE 8000
CMD ["oura-mcp"]
docker build -t oura-mcp-python .
docker run -p 8000:8000 --env-file .env oura-mcp-python
Troubleshooting
"Not authenticated" errors
- Visit
/oauth/authorizeto connect your Oura account - Check that your OAuth credentials are correct in
.env
Token refresh fails
- Ensure
TOKEN_ENCRYPTION_KEYhasn't changed - Try disconnecting (
POST /oauth/disconnect) and reconnecting
"Invalid redirect URI" from Oura
- Ensure your
OURA_REDIRECT_URIexactly matches what's configured in your Oura OAuth application - Include the full path:
http://localhost:8000/oauth/callback
Claude Desktop doesn't see the tools
- Verify the server is running:
curl http://localhost:8000/health - Check Claude Desktop logs: Help → Open Logs Folder
- Ensure
mcp-remoteis available:npx -y mcp-remote --help
Development
Running Tests
pip install -e ".[dev]"
pytest
Code Formatting
ruff format .
ruff check --fix .
License
MIT
Credits
- Based on the meimakes/oura-mcp-server TypeScript implementation
- Built with MCP Python SDK
- Uses the Oura API V2
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.