Getting Started with OpenClaw (Clawdbot): A Beginner’s Guide

Ready to self-host your own AI assistant? This beginner-friendly guide walks you through setting up OpenClaw from scratch — no prior experience required.

What We’ll Cover

1. Prerequisites and requirements
2. Installing OpenClaw
3. Running the setup wizard
4. Connecting your AI provider
5. Adding your first messaging channel
6. Your first conversation

Total time: 15-30 minutes

Prerequisites

Before starting, make sure you have:

• A computer — Windows, Mac, or Linux all work
• Node.js 18 or higher — the runtime that powers OpenClaw
• An AI API key — from Anthropic (Claude) or OpenAI (GPT-4)
• Basic terminal comfort — you’ll type a few commands

Installing Node.js

If you don’t have Node.js installed:

Windows/Mac: Download from nodejs.org and run the installer.

Linux:

curl -fsSL https://deb.nodesource.com/setup_20.x | sudo -E bash -
sudo apt-get install -y nodejs

Verify installation: node --version (should show v18 or higher)

Getting an AI API Key

You’ll need an API key from at least one AI provider:

• Anthropic (Claude): console.anthropic.com
• OpenAI (GPT-4): platform.openai.com

Both offer pay-as-you-go pricing. Start with $10-20 credit — it goes a long way for personal use.

Step 1: Install OpenClaw

Open your terminal (Command Prompt on Windows, Terminal on Mac/Linux) and run:

npm install -g openclaw

This installs OpenClaw globally on your system. Wait for it to complete (1-2 minutes).

Verify installation:

openclaw --version

Step 2: Run the Setup Wizard

The onboarding wizard guides you through initial configuration:

openclaw onboard

The wizard will ask you about:

• Your preferred AI provider
• API key configuration
• Basic settings

Follow the prompts — defaults are sensible for most users.

Step 3: Add Your AI Provider

If the wizard didn’t prompt for your API key, add it manually:

openclaw configure --section auth

Enter your Anthropic or OpenAI API key when prompted. The key is stored securely on your machine.

Step 4: Start the Gateway

The gateway is the background service that runs your AI assistant:

openclaw gateway start

Check that it’s running:

openclaw status

You should see the gateway status as “running” with your configured AI provider listed.

Step 5: Open the Dashboard

OpenClaw includes a web dashboard for chatting and configuration:

openclaw dashboard

This opens your browser to the local dashboard. You can chat with your AI right here to verify everything works!

Step 6: Connect Telegram (Optional but Recommended)

To chat from your phone via Telegram:

1. Open Telegram and message @BotFather
2. Send /newbot and follow prompts to create a bot
3. Copy the bot token BotFather gives you
4. Run: openclaw configure --section telegram
5. Paste your bot token
6. Restart the gateway: openclaw gateway restart

Now message your bot on Telegram — your AI assistant responds!

For detailed Telegram setup, see our complete Telegram guide.

Your First Conversation

Whether through the dashboard or Telegram, try asking your AI:

• “Hello! What can you help me with?”
• “What’s the weather like today?” (if web search is enabled)
• “Remember that my favorite color is blue” (testing memory)
• “Set a reminder for 10 minutes from now” (if cron is configured)

Your AI assistant is now running locally, under your control!

Essential Commands Reference

Running 24/7

For always-on availability:

1. Keep the gateway running: Use openclaw gateway start in a persistent terminal or configure it as a system service
2. Dedicated hardware: Consider a mini PC for always-on operation
3. Auto-start: Configure your system to start the gateway on boot

Troubleshooting Common Issues

“Command not found: openclaw”

Node.js global packages aren’t in your PATH. Try restarting your terminal, or run with npx: npx openclaw

“Cannot connect to AI provider”

Verify your API key is correct and has credit. Check: openclaw configure --section auth

“Gateway won’t start”

Check if something else is using the default port: openclaw status shows details. Try a different port in config if needed.

Telegram bot doesn’t respond

Ensure the gateway is running, bot token is correct, and you’ve approved yourself (if using pairing mode).

Next Steps

Now that OpenClaw is running:

• Connect more messaging platforms
• Explore the dashboard settings
• Set up scheduled tasks and reminders
• Customize your AI’s personality in the workspace files
• Join the OpenClaw Discord for community support

Frequently Asked Questions

How much does OpenClaw cost to run?

The software is free. You only pay for AI API usage — typically $5-30/month for personal use depending on how much you chat.

Can I use multiple AI models?

Yes! Configure multiple providers and switch between Claude, GPT-4, and others based on task or preference.

Is my data private?

Your conversation history stays on your machine. Only individual prompts go to your chosen AI provider when you chat.

What if I break something?

OpenClaw stores config in ~/.openclaw/. Worst case, delete that folder and run openclaw onboard again for a fresh start.

Can I run this on a Raspberry Pi?

Yes! A Raspberry Pi 4 or 5 works fine for API-based AI. Just install Node.js for ARM and follow the same steps.

Congratulations!

You now have a self-hosted AI assistant running on your own hardware. Welcome to the world of personal AI ownership!

This article contains affiliate links. As an Amazon Associate, we earn from qualifying purchases at no additional cost to you.