OmniClaw

Getting Started

Deploy your AI agent
in 2 minutes.

No servers to configure, no code to write. Follow these five steps and your OpenClaw agent will be live on WhatsApp.

1

Sign up with your email

Head to the connect page and enter your email address. You will receive a one-time passcode (OTP) — six digits sent to your inbox. Enter the code to verify your identity. No passwords, no forms, no account setup.

What you will see:

An email input field followed by a 6-digit code input. The code expires after 10 minutes. Check your spam folder if it does not arrive within a few seconds.

2

Scan the QR code

After verification, a QR code appears on screen. Open WhatsApp on your phone, go to Settings > Linked Devices > Link a Device, and scan the code with your camera.

This links your WhatsApp account to your OmniClaw agent. The connection is direct — we use the same Linked Devices protocol that WhatsApp Web uses. Your messages are end-to-end encrypted in transit.

Tip:

Keep the browser tab open while scanning. The QR code refreshes every 30 seconds. If it expires, a new one will appear automatically.

3

Send your first message

Your agent is now live. Open any WhatsApp chat and send a message to yourself or have someone message you. The AI agent will respond automatically.

Try these prompts:

  • "Summarize this article:" followed by a URL
  • "Translate this to Spanish: Good morning"
  • "Write a polite reply to this message"
  • "What are 3 dinner ideas with chicken and rice?"
4

Store API keys in the vault

If you want your agent to use external services — like a custom OpenAI key, a weather API, or a database — store the credentials in the vault from your dashboard.

Vault secrets are encrypted at rest and injected into your agent's VM as environment variables at boot time. They are never written to disk inside the VM. Even the agent code cannot read them back after injection.

How it works:

Go to Dashboard > Vault, add a key-value pair (e.g. OPENAI_API_KEY), and your agent will have access to it as process.env.OPENAI_API_KEY on the next session.

5

Customize your agent

From the dashboard, you can configure your agent's behavior:

  • Model selection — switch between GPT-4o, Claude, or Gemini
  • Skills — enable built-in capabilities like web search, code execution, or file handling
  • System prompt — define your agent's personality and instructions
  • Auto-reply rules — choose which chats the agent responds to

Ready to start?

Free to try. No credit card required. Your agent will be live in under two minutes.