Connect OpenClaw
to WhatsApp.
Scan a QR code and your AI agent is live on WhatsApp. No app installs, no bot accounts, no API keys needed.
Three steps. Under a minute.
- 01
Verify your email
Enter your email and confirm with a one-time code. This creates your account and provisions a microVM for your agent.
- 02
Scan the QR code
A QR code appears on screen. Open WhatsApp on your phone, go to Settings > Linked Devices > Link a Device, and point your camera at the code. This uses WhatsApp's official Linked Devices protocol — the same mechanism as WhatsApp Web.
- 03
Start chatting
Your agent is live immediately. Send a message in any WhatsApp chat and your AI assistant responds within seconds, powered by GPT-4o, Claude, or Gemini.
What you can do.
Personal assistant
"Remind me to call the dentist tomorrow at 9am and draft a message I can send them"
Research and summarization
"Summarize this article and list the three key takeaways: https://..."
Language translation
"Translate to French: We'll arrive at the hotel around 3pm and would like early check-in"
Writing help
"Write a professional reply to this email: [paste email]. Keep it polite but firm."
Your messages, your business.
Encrypted in transit
WhatsApp encrypts messages with the Signal protocol up to the linked device, which is your agent's VM. There your agent processes them in plaintext, inside a VM dedicated to your account.
No message storage
We do not read, log, or train on your conversations. Messages are processed in your VM's memory and discarded after the response is sent.
Hardware isolation
Each agent runs in a dedicated Firecracker microVM with its own Linux kernel. No shared infrastructure with other users.
EU hosting
OmniRun infrastructure runs on servers in the EU (Finland). Message content you send to an LLM is processed by the provider you choose (e.g. OpenAI, Anthropic, Google), which may be outside the EU.