Repolode
thinkwee

thinkwee/HiMe

A self-hosted personal health AI agent that ingests Apple Watch and iPhone health data and sends proactive insights through an iOS app, web dashboard, or IM chat.

Open-source alternative to:WhoopApple Health

Stars
71
Forks
10
Open issues
11
Last commit
14 days ago

What is it?

HiMe is a self-hosted personal health AI agent platform. It ingests wearable data from Apple Watch and iPhone, including heart rate, HRV, SpO2, sleep stages, workouts and mobility metrics, and runs the data through an LLM agent that produces scheduled checks, event-triggered analysis and chart-rich reports.

Users interact through a native iOS app with in-app chat and push notifications, a watchOS companion, a web dashboard with an agent monitor, or optional IM gateways for Telegram, Feishu and WeChat. The agent can design a personalised check-in plan from a short goal survey and generate custom pages for repeated workflows.

The project runs locally via Docker or a native setup script, and it is positioned as research-grade software for personal use rather than a medical device.

Key features

  • 01Ingests Apple Watch and iPhone health data such as heart rate, HRV, SpO2 and sleep stages.
  • 02Native iOS in-app chat with streaming replies and proactive push notifications.
  • 03Scheduled and event-triggered AI analysis delivered as chart-rich reports.
  • 04Optional Telegram, Feishu and WeChat gateways for chatting with the agent.
  • 05Web dashboard with light and dark themes and an agent activity timeline.
  • 06Skills system for reusable analysis playbooks.

Use cases

  • Keeping a self-hosted record of wearable health trends on your own server or homelab.
  • Getting daily or weekly summaries of sleep, activity and heart metrics from an agent.
  • Asking natural-language questions about your own health data from an iPhone or chat app.
  • Building custom health dashboards or Readiness-style scores by describing them to the agent.

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/thinkwee/HiMe.git HiMe
  2. 2

    Enter the project directory

    cd HiMe
  3. 3

    Run the setup wizard to start the server

    ./setup.sh
  4. 4

    Install the iOS app from the App Store and set the Server URL to your host

Things to know

  • Requires an Apple Watch or iPhone health data source and the iOS app for the main experience.
  • Licensed under PolyForm Noncommercial 1.0.0, so commercial use is not permitted.
  • The README describes it as research-grade software and not a medical device.
  • Self-hosting requires running a server with Docker or native setup.

Topics

FAQ

What is HiMe?
A self-hosted personal health AI agent that ingests Apple Watch and iPhone health data and sends proactive insights through an iOS app, web dashboard, or IM chat. HiMe is a self-hosted personal health AI agent platform. It ingests wearable data from Apple Watch and iPhone, including heart rate, HRV, SpO2, sleep stages, workouts and mobility metrics, and runs the data through an LLM agent that produces scheduled checks, event-triggered analysis and chart-rich reports.
Is HiMe free to use? What license does it have?
HiMe is open source under the Other license. Check the license terms before using it in commercial products.
Is HiMe still maintained?
Actively maintained. The last commit to HiMe was on Oct 8, 2026.
What can I use HiMe for?
Keeping a self-hosted record of wearable health trends on your own server or homelab. Getting daily or weekly summaries of sleep, activity and heart metrics from an agent. Asking natural-language questions about your own health data from an iPhone or chat app. Building custom health dashboards or Readiness-style scores by describing them to the agent.
How do I install HiMe?
Clone the repository: git clone https://github.com/thinkwee/HiMe.git HiMe Enter the project directory: cd HiMe Run the setup wizard to start the server: ./setup.sh Install the iOS app from the App Store and set the Server URL to your host
Where can I get HiMe?
The source code and installation instructions are on GitHub: https://github.com/thinkwee/HiMe

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects