OpenClaw Setup Guide: From Zero to Your Own AI Assistant in 15 Minutes

What is OpenClaw?

OpenClaw is an open-source personal AI assistant that runs on your own machine. Unlike cloud-based assistants, your data stays local. It connects to messaging apps you already use — WhatsApp, Telegram, Slack, Discord — and can browse the web, manage files, run scripts, and remember context across conversations.

This is the Quick Start Guide for personal installation on your own computer. If you need to deploy on a server, VPS, or Docker for production use, see the Complete Clawdbot/OpenClaw Setup Guide A-Z.

Prerequisites

Step 1: Install OpenClaw

npm install -g openclaw@latest

Verify the installation:

openclaw --version

Step 2: Run the Onboard Wizard

openclaw onboard --install-daemon

The wizard walks you through 4 steps:

1. Gateway setup — starts the local WebSocket server at ws://127.0.0.1:18789
2. AI model selection — choose Claude, GPT, or a local model
3. Channel pairing — connect your first messaging app
4. Skill installation — pick starter skills from ClawHub

Step 3: Connect a Messaging Channel

During onboard, select your preferred channel. For WhatsApp, scan the QR code displayed in your terminal. For Telegram, enter your bot token from @BotFather.

After pairing, send a test message. OpenClaw responds directly in your chat app.

Step 4: Configure Your Workspace

OpenClaw stores config at ~/.openclaw/openclaw.json. A minimal config:

{
  "agent": {
    "model": "anthropic/claude-sonnet-4-5"
  },
  "channels":