LifeOS
A personal AI setup layer that gives an AI coding harness your goals, context, and memory.
- Stars
- 19.7k
- Forks
- 2.5k
- Licence
- MIT
- Language
- TypeScript
- Last update
- 9 days ago
- Type
- AI model / agent
What it is
LifeOS is a set of files, skills, and routing rules that sits on top of an AI coding harness such as Claude Code. It stores your goals, people, and preferences so the AI can keep working toward them without you re-explaining context each time. The README says it is aimed at general life and work tasks, not only programming.
- Ask the AI to research a topic and have a matching workflow trigger automatically via intelligent routing
- Keep goals, contacts, and preferences available across sessions
- Build reusable custom skills for tasks you repeat often
- Use the included skill catalog (97 public skills per the README) for research, writing, security, and similar tasks
Great for
- People already using an AI coding harness who want it to remember their goals and context
- Developers comfortable with TypeScript, Bun, and command-line setup
- Users who want persistent memory and reusable skills across sessions
- People interested in a personal AI assistant that improves its own workflows
Not ideal for
- People who do not use an AI coding harness; the README says a capable harness is required, and Claude Code is the most-tested path
- Users who want a polished consumer app with no setup, since install is driven by an AI and a terminal
- Privacy-focused users who need fully local models today, since local model support is listed only on the roadmap
How to get it
- 1
See the README
You'll need: A capable AI coding harness such as Claude Code · bun
One-line installer for Claude Code on macOS/Linux
curl -fsSL https://ourlifeos.ai/install.sh | bash - 2
Run it
- Install a supported AI coding harness such as Claude Code and bun
- Paste the README prompt into your AI harness: Read https://ourlifeos.ai/install and install LifeOS for me.
- Or run the one-line installer for Claude Code on macOS/Linux
- Approve the steps the AI asks permission for during setup
- Explore the docs at docs.ourlifeos.ai or the walkthrough video linked in the README
- 3
Get the most out of it
- Back up your ~/.claude folder before upgrades, using the command in the README
- Keep personal customizations inside the USER/ folder, which the README says upgrades do not touch
- Use the Skills catalog on ourlifeos.ai/skills to find existing capabilities before writing new ones
- Integrate Fabric patterns into your skills, as the README suggests many users do
- Use git to version your LifeOS setup so you can roll back changes
Downloads
Strengths
- Open source under the MIT license with a large public following (about 19.6k stars at snapshot)
- Designed to be harness-agnostic, with core ideas described as portable to other agents
- Clear recovery guidance in the README, including backups and a note that the USER/ folder is not overwritten by upgrades
Trade-offs
- Install and daily use depend on a capable AI harness, so costs and access to that harness apply
- The README's claims about outcomes such as efficiency and 'Euphoric Surprise' are not backed by measured evidence in the text
- Local model support and remote access are roadmap items, not current features
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| LifeOS (this) | 19.7k | AI model / agent | intermediate | 9 days ago | People already using an AI coding harness who want it to remember their goals and context |
| Fabric | 44.5k | CLI | intermediate | 4 days ago | Command-line users who want to run AI prompts on files or piped text |
| i-have-adhd | 59.7k | Other | beginner | 17 days ago | Developers who find long AI replies hard to scan or act on |
| claudian | 15.9k | App | intermediate | 4 days ago | Obsidian users who already use Claude Code or Codex and want that agent inside their notes |
| superpowers | 311k | Framework | beginner | 7 days ago | Developers who already use a supported coding agent such as Claude Code, Codex, Cursor, or Gemini CLI |
| ECC | 295k | Dev tool | intermediate | 4 days ago | Developers who already use Claude Code and want a shared workflow for planning, testing and review |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.