Do I need coding experience to install OpenClaw?

No coding required. The installation is three terminal commands — install Node.js, run npm install, and answer the setup wizard. The wizard walks you through choosing an AI provider, entering your API key, and picking a model. No programming knowledge needed.

Which AI provider should I choose as a beginner?

Start with Claude (Anthropic) or GPT-4o (OpenAI) for best results. Both require a paid API key but cost just a few cents per day for personal use. If you want fully private AI with no API costs, choose Ollama — it runs a model entirely on your own machine.

Does OpenClaw work on Windows?

Yes. OpenClaw runs on macOS, Linux, and Windows. The installation steps are the same on all platforms — install Node.js, run npm install -g openclaw@latest, and follow the wizard. Windows users should use PowerShell or Windows Terminal instead of Command Prompt.

What if the install command fails?

The most common cause is an outdated Node.js version. OpenClaw requires Node 22.16 or higher. Run node --version to check. If your version is too old, visit nodejs.org, download the latest LTS release, install it, and retry the command.

Module 0: Quick Setup – HowOpenClaw

Install OpenClaw and run your first AI conversation in under 15 minutes. Step-by-step for Mac, Linux, and Windows — no coding required.

0 of 10 modules complete0%

5 min read

What you will learn

Understand what OpenClaw is and what it does
Install OpenClaw on your machine
Complete the onboarding wizard
Send your first message and get a real response

Why this matters

In under 15 minutes, you will have your own private AI assistant running on your machine. It is not a cloud service someone else controls. It is yours.

What you'll have when you're done

What you'll have when this is done: A running AI assistant that responds to your messages within seconds. You'll test it with a real command at the end and see it reply.

What is OpenClaw?

OpenClaw is a personal AI assistant that runs on your computer. You install it once, connect it to the apps you already use (WhatsApp, Telegram, Slack), and it handles tasks, answers questions, and runs automations on your behalf.

It is not like ChatGPT or Claude. Those are websites you visit. OpenClaw is software that lives on your machine, keeps your data local, and connects to any AI model you choose (OpenAI, Claude, Ollama for fully local).

Before you start: quick requirements check

OpenClaw is lightweight — it's a Node.js process, not a heavy application. Here's what you need:

What	Minimum	Notes
Node.js	22.16	Node 24 recommended — download here
RAM	2 GB	4 GB+ comfortable; 8 GB+ if using local models (Ollama)
OS	macOS, Linux, Windows (WSL2)	Any modern 64-bit system
Internet	Required for cloud AI	Not needed if using Ollama locally

No GPU required. No special hardware. If you're running macOS, Linux, or Windows with a few GB of RAM, you're good.

→ Running a server, Raspberry Pi, or want 24/7 operation? See the full System Requirements for use-case specific guidance.

Open your Terminal

All commands in this course are run in the Terminal (also called the command line or shell). On macOS, press Cmd + Space, type "Terminal", and hit Enter. On Windows, use PowerShell or Windows Terminal. On Linux, use your system's terminal app.

When you see a code block like this:

some command

That means: click in your Terminal window, type that command exactly, and press Enter.

Install OpenClaw

You need Node 24 (recommended) or Node 22.16+. Check your version first:

node --version

If you need to install or update Node, visit nodejs.org and download the installer for your operating system — it works just like installing any other app.

Then install OpenClaw globally:

npm install -g openclaw@latest

Verify the installation:

openclaw --version

Run the onboarding wizard

The wizard walks you through initial configuration. It takes about two minutes:

openclaw onboard --install-daemon

The wizard has four prompts. Here is exactly what you will see:

Prompt A — Choose your AI provider

? Which AI provider do you want to use?
❯ OpenAI        (GPT-4o, GPT-4o mini)
  Anthropic     (Claude Sonnet, Claude Haiku)
  Ollama        (local — no API key required)
  Other / Skip

Which provider should I pick?

OpenAI (GPT-4o) is the easiest starting point — widely supported, fast, and great at following instructions. Anthropic (Claude) is excellent for long conversations and nuanced tasks. Ollama runs models locally with no API key and no cost, but requires more RAM (8 GB minimum, 16 GB recommended).

Prompt B — Enter your API key

If you chose OpenAI or Anthropic, the wizard will ask for your API key:

? Paste your OpenAI API key (starts with sk-):

Where to get your key:

OpenAI: platform.openai.com/api-keys → Create new secret key
Anthropic: console.anthropic.com/settings/keys → Create key

Your key is stored locally in ~/.openclaw/openclaw.json — it never leaves your machine. If you chose Ollama, skip this — no key needed.

Prompt C — Choose your model

? Which model do you want to use?
❯ gpt-4o          (recommended — best balance of speed and quality)
  gpt-4o-mini     (faster and cheaper — good for simple tasks)

For Anthropic:

❯ claude-sonnet-4-5   (recommended)
  claude-haiku-4-5    (faster, lower cost)

For Ollama:

❯ llama3.2     (general purpose, 8 GB RAM)
  mistral      (fast, good for instructions)
  phi3         (small — works on 4 GB RAM)

You can change this any time

Run openclaw configure to switch providers or models without losing any data.

Prompt D — Set up your workspace

? Where should OpenClaw store your workspace? (~/.openclaw/workspace)
  Press Enter to accept the default

Hit Enter. The wizard then installs the daemon and confirms:

✓ Config saved
✓ Workspace created at ~/.openclaw/workspace
✓ Daemon installed (auto-starts on login)
✓ Gateway started

Open the dashboard

openclaw dashboard

This opens http://127.0.0.1:18789/ in your browser — you should see the OpenClaw interface with your agent status, connected channels, and configuration.

Send your first message

openclaw agent --message "Hello! Tell me what you can do."

You should get a response within a few seconds. If you see a reply, your installation is working correctly.

Verify the installation

Run the built-in health check:

openclaw doctor

All checks green? You are ready for Module 1.

Something not working?

Check the troubleshooting guide for common issues, or run openclaw doctor to diagnose the problem.

Need a command reference? The CLI Cheatsheet has every openclaw command with flags and examples.

Finished this module?

Tracks your progress across all 10 modules

FAQ

Do I need coding experience to install OpenClaw?: No coding required. The installation is three terminal commands — install Node.js, run npm install, and answer the setup wizard. The wizard walks you through choosing an AI provider, entering your API key, and picking a model. No programming knowledge needed.
Which AI provider should I choose as a beginner?: Start with Claude (Anthropic) or GPT-4o (OpenAI) for best results. Both require a paid API key but cost just a few cents per day for personal use. If you want fully private AI with no API costs, choose Ollama — it runs a model entirely on your own machine.
Does OpenClaw work on Windows?: Yes. OpenClaw runs on macOS, Linux, and Windows. The installation steps are the same on all platforms — install Node.js, run npm install -g openclaw@latest, and follow the wizard. Windows users should use PowerShell or Windows Terminal instead of Command Prompt.
What if the install command fails?: The most common cause is an outdated Node.js version. OpenClaw requires Node 22.16 or higher. Run node --version to check. If your version is too old, visit nodejs.org, download the latest LTS release, install it, and retry the command.

Module 0: Quick Setup

On this page