OpenClaw overview
OpenClaw helps you run a personal AI assistant on your own infrastructure. This section is the fastest path from zero to a working assistant with memory, Skills, and scheduled automation.This repository currently contains two tracks: the current
OpenClaw tutorial and a preserved set of
legacy Trellis workflow docs. If your goal is to deploy a personal AI assistant, start here.What you will build
- A self-hosted AI assistant that runs on your own machine or server
- A persistent assistant with memory, personality, and scheduled checks
- A setup you can extend with Skills, tools, and custom workflows
Before you start
- A Linux server or always-on machine is recommended
- An API key for the model provider you plan to use
- Telegram is the easiest chat channel if you want to follow the tutorial as written
Follow the 7-day path
Day 1: Meet OpenClaw
Understand how a personal AI assistant differs from a normal chat app.
Day 2: Build in 10 minutes
Deploy the base system and send your first message.
Day 3: Give it a soul
Define identity, tone, and long-term behavior.
Day 4: Connect tools and data
Add channels, tools, and external context.
Day 5: Install Skills
Extend your assistant with reusable capabilities.
Day 6: Configure heartbeat and memory
Make the assistant proactive and persistent.
Day 7: Advanced workflows
Customize behavior, automation, and long-running routines.
Resources
Jump to the main links for repository, community, Skills, and follow-up reading.