The Guide
Fifteen chapters. From “what is this?” to running a proactive, multi-agent AI setup that works while you sleep.
What it does, why people care, and whether it's for you.
What you need (and what you don't).
Mac, Windows, Linux. Every command explained.
Running the onboarding wizard and getting your Gateway up.
Talk to your AI in the browser. No setup needed.
WhatsApp, Telegram, Discord, Slack — pick your poison.
Give your agent a name, personality, and memory.
Starter projects and ideas so you're not just staring at a cursor.
They will. Here's how to fix them.
Claude vs GPT vs MiniMax -- which to use and when.
SOUL.md, USER.md, MEMORY.md -- how your agent remembers you.
Install plugins, browse ClawHub, teach your agent new tricks.
Run specialized agents for different tasks.
Heartbeats, cron jobs, webhooks -- make your agent work while you sleep.
Keep your stuff safe. Build trust gradually.