danielmiessler/LifeOS

What it solves

LifeOS is a general-purpose AI harness designed to transform a raw AI coding agent into a personalized life operating system. It solves the problem of AI lacking persistent memory, personal context, and specialized skills, allowing the AI to understand the user's goals, preferences, and identity to help them move from their "Current State" to an "Ideal State."

How it works

LifeOS acts as a layer on top of an AI harness (such as Claude Code, Cursor, or Codex). It uses universal primitives—including hooks, skills, and context files—to provide the AI with persistent memory (Cortex), intelligent routing (Synapse), and a library of specialized capabilities. The system is harness-agnostic and built using TypeScript and Bash, allowing it to be installed and configured by the AI agent itself via a prompt.

Who it’s for

It is designed for individuals who use AI coding harnesses and want a more integrated, personalized AI assistant that remembers past sessions, manages personal goals, and automates complex workflows across life and work.

Highlights

• Persistent Memory: The Cortex system maintains a typed Knowledge Archive of people, companies, ideas, and research across sessions.
• Custom Skill System: A library of specialized capabilities for research, security, writing, and art.
• Harness Agnostic: Works with various high-end AI harnesses, though optimized for Claude Code.
• Self-Improving: The system can modify itself based on what it learns about the user.
• Intent Engineering: Uses a framework to move users from their current state to an ideal state.