Neura Market Logo

Getting Started with OpenClaw: Install and Run Your First Chat

This page guides you through installing OpenClaw and running your first chat in minutes. It is intended for new users who need to set up the Gateway, configure authentication, and start chatting with an AI assistant.

Read this when

  • First time setup from zero
  • You want the fastest path to a working chat

Set up OpenClaw, complete the onboarding flow, and begin chatting with your AI assistant in roughly five minutes. Once finished, you will have a running Gateway, authentication configured, and an active chat session.

What you need

  • Node.js 22.22.3+, 24.15+, or 25.9+ (version 24 is the recommended default)
  • An API key from a model provider (Anthropic, OpenAI, Google, etc.), the onboarding process will ask for one

Tip

Use node --version to verify your Node version. Windows users: the native Windows Hub app provides the simplest desktop experience. The PowerShell installer and WSL2 Gateway paths are also supported. See Windows. Need to install Node? Refer to Node setup.

Quick setup

Install OpenClaw

macOS / Linux

curl -fsSL https://openclaw.ai/install.sh | bash

Install Script Process

Windows (PowerShell)

iwr -useb https://openclaw.ai/install.ps1 | iex

Note

Alternative installation approaches (Docker, Nix, npm): Install.

Run onboarding

openclaw onboard --install-daemon

The wizard guides you through picking a model provider, providing an API key, and setting up the Gateway. The QuickStart process typically takes just a few minutes, though provider sign-in, channel pairing, daemon installation, network downloads, skills, or optional plugins may extend the full onboarding time. Skip any optional steps and come back to them later using openclaw configure.

For complete reference material, see Onboarding (CLI).

Verify the Gateway is running

openclaw gateway status

The Gateway should be listening on port 18789.

Open the dashboard

openclaw dashboard

This launches the Control UI in your browser. If it loads successfully, everything is functioning correctly.

Send your first message

Enter a message in the Control UI chat and you should receive an AI response.

Prefer to chat from your phone? The quickest channel to configure is Telegram (it only requires a bot token). Check Channels for a full list of options.

Advanced: mount a custom Control UI build

If you maintain a localized or customized dashboard build, set gateway.controlUi.root to a directory containing your built static assets and index.html.

mkdir -p "$HOME/.openclaw/control-ui-custom"
# Copy your built static files into that directory.

Then configure:

{
  "gateway": {
    "controlUi": {
      "enabled": true,
      "root": "$HOME/.openclaw/control-ui-custom"
    }
  }
}

Restart the gateway and reopen the dashboard:

openclaw gateway restart
openclaw dashboard

What to do next

Advanced: environment variables

If OpenClaw runs as a service account or you need custom paths:

  • OPENCLAW_HOME, home directory used for internal path resolution
  • OPENCLAW_STATE_DIR, overrides the state directory
  • OPENCLAW_CONFIG_PATH, overrides the config file path

Full reference: Environment variables.