Install & First Run
Get OpenClaw running on your machine in under 5 minutes.
1. Prerequisites
- Node.js 22+
- An AI API key (Anthropic Claude recommended, or OpenAI)
- Optional: A chat platform account (WhatsApp, Telegram, Discord, etc.)
2. Install OpenClaw
Install globally via npm:
npm install -g openclaw@latest
Or use the one-line installer:
curl -fsSL https://openclaw.ai/install.sh | bash
Or build from source via GitHub: github.com/openclaw/openclaw
3. Run the Onboarding Wizard
openclaw onboard --install-daemon
The wizard walks you through:
- AI model selection — Claude, GPT-4, or Gemini
- Channel setup — connect your preferred messaging platform
- Security settings — API key storage, permissions
- Daemon installation — launchd (macOS) or systemd (Linux)
4. Your First Conversation
Once onboarding completes, OpenClaw runs as a background daemon. Send a message from your configured channel (WhatsApp, Telegram, etc.) or use the CLI:
openclaw chat "What can you do?"
Or open the dashboard:
openclaw dashboard
5. Verify Setup
openclaw doctor
This checks:
- Node.js version
- Gateway health
- Channel connections
- Model authentication
Troubleshooting
- Check logs:
openclaw logs - Gateway status:
openclaw gateway status - Community: discord.gg/w2sKrd9cSH
Next Steps
- Workspace Config — shape your assistant with AGENTS.md, SOUL.md, TOOLS.md
- Channels — connect to WhatsApp, Telegram, Discord
- Skills & ClawHub — browse 565+ community skills
Understanding Agent Runtimes
OpenClaw's onboarding wizard configures an Agent Runtime — the same concept you'll master when building your own agents. Understanding model selection, tool configuration, and memory is foundational.
Learn Agent Runtime Concepts