Oura Mcp Unofficial

Name: Oura Mcp Unofficial
Author: davidmosiah

@davidmosiah

Repository Homepage

About Oura Mcp Unofficial

Local-first MCP server for Oura Ring readiness, sleep, HRV — Claude/Cursor/ChatGPT ready

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "oura": {
      "command": "npx",
      "args": [
        "-y",
        "oura-mcp-unofficial"
      ]
    }
  }
}

Tools

realistic synthetic readiness/sleep/activity payloads (no account needed; see [Quickstart](#quickstart-see-the-data-before-you-connect))

personalized 3-step setup walkthrough that adapts to your current state

verify local setup before calling Oura

inventory supported data domains, scopes, privacy modes and recommended first calls without calling Oura APIs.

readiness, sleep, activity and SpO2 brief for today

scorecard, comparison vs prior week, next-week plan

Overview

What is Oura Mcp Unofficial?

Oura Mcp Unofficial is a local-first MCP server that connects AI agents to your Oura Ring data — readiness scores, sleep stages, HRV, activity, SpO2, and personal info — without needing to copy-paste from the Oura app. It handles the OAuth dance locally, normalizes responses across endpoints, and exposes Oura through the Model Context Protocol. Tokens never leave your machine, and raw payloads are opt-in via privacy-mode defaults.

How to use Oura Mcp Unofficial?

First create an Oura app at cloud.ouraring.com with redirect URI http://127.0.0.1:3000/callback. Then run npx -y oura-mcp-unofficial setup (paste client id + secret), npx -y oura-mcp-unofficial auth (opens browser for OAuth), and npx -y oura-mcp-unofficial doctor to verify. Add the server to your MCP client config as a npx command, or for Claude Desktop run setup --client claude to auto-write the snippet. You can also call oura_demo before connecting to see synthetic payloads.

Key features of Oura Mcp Unofficial

Local-first: OAuth tokens never leave your machine
Supports readiness, sleep, activity, HRV, SpO2, workouts, sessions, tags, personal info
Privacy‑mode defaults to structured; raw JSON is opt‑in
Demo mode (oura_demo) works without an Oura account
Includes prompts like oura_daily_checkin and oura_weekly_review
Caching opt‑in via SQLite; configurable via env or ~/.oura-mcp/config.json
Compatible with Hermes‑based remote setups

Use cases of Oura Mcp Unofficial

Get a daily readiness and sleep brief for your AI agent to plan workouts or rest
Review weekly trends across sleep, HRV, and activity with a next‑week plan
Investigate heart‑rate records without exposing raw data
Let your agent evaluate recovery before recommending training intensity
Set up a private wellness agent using Claude Desktop, Cursor, or Hermes

FAQ from Oura Mcp Unofficial

What data can Oura Mcp Unofficial access?

It can access daily readiness scores, daily sleep scores with sleep periods (stages, timing), daily activity (steps, calories, MET), heart‑rate time series, HRV via readiness contributors, SpO2 daily averages, workouts, sessions, tags, and personal info (DOB, sex, height, weight) — subject to scopes granted in the Oura Cloud API. Continuous sensor telemetry is not available.

What are the requirements to run Oura Mcp Unofficial?

You need Node.js 20+ and an Oura app created at cloud.ouraring.com with the redirect URI set to http://127.0.0.1:3000/callback.

How do I verify my setup?

Run oura_connection_status in your agent after completing setup and auth. You can also use oura_doctor from the command line to confirm everything is ready.

Is my Oura token secure?

Yes. OAuth tokens are stored in ~/.oura-mcp/tokens.json with Unix permissions 0600. The server never prints or returns access/refresh tokens to the MCP client.

Is this an official Oura product?

No. This is an unofficial project not affiliated with, endorsed by, or supported by Ōura Health Oy. Use it only with your own Oura account and in line with the Oura Cloud API terms.

Comments

More AI & Agents MCP servers

MCP-LLM Bridge

patruff

Bridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools

976

MCP Server - Remote MacOs Use

baryhuang

The only general AI agent that does NOT requires extra API key, giving you full control on your local and remote MacOs from Claude Desktop App

487

Solon Ai

opensolon

Java AI application development framework (supports LLM-tool,skill; RAG; MCP; Agent-ReAct,Team-Agent). Compatible with java8 ~ java25. It can also be embedded in SpringBoot, jFinal, Vert.x, Quarkus, and other frameworks.

416