How to Deploy OpenClaw in 2 Minutes

What is OpenClaw?

OpenClaw is an open-source AI assistant framework that connects to messaging platforms like WhatsApp and Telegram. It handles conversations, manages context, and can be extended with custom skills to do things like search the web, manage calendars, or interact with APIs.

OmniClaw is a managed hosting platform for OpenClaw, built by OmniRun. It runs each OpenClaw instance in its own hardware-isolated Firecracker microVM, with an encrypted credential vault and built-in LLM proxy. You get the full power of OpenClaw without managing servers.

Prerequisites

All you need is an email address. No credit card required. OmniClaw comes with $5 free credit to get started.

Step 1: Go to the Connect Page

Open omniclaw.run/connect in your browser. This is the onboarding flow that sets up your OpenClaw instance.

Step 2: Enter Your Email

Type your email address and click Continue. You will receive a one-time verification code. Enter the 6-digit code to confirm your identity. That is your entire account setup — no password to remember.

Step 3: Choose Your Channel

Pick how you want to talk to your AI assistant:

• WhatsApp — Best for personal use. Your assistant gets its own WhatsApp presence linked to your phone.
• Telegram — Best for bots and group chats. Your assistant runs as a standalone Telegram bot.

Step 4: Connect Your Channel

The connection step depends on which channel you chose:

Step 5: Send Your First Message

Open the chat with your assistant and say hello. You should get a response within a few seconds. Your OpenClaw instance is now running in its own isolated sandbox, processing messages through the built-in LLM proxy.

What Happens Behind the Scenes

When you complete the connect flow, OmniClaw provisions a full environment for you:

• A Firecracker microVM boots in under 200ms from a snapshot — the same technology that powers AWS Lambda.
• Your messaging credentials are stored in an encrypted vault and injected at runtime. They never touch the sandbox filesystem.
• All LLM calls route through Kilo, OmniRun's LLM proxy. You can switch models, track usage, and set rate limits without changing your agent config.
• Your instance runs on EU bare-metal servers in Finland and Germany, with full data residency compliance.

Next Steps

Your agent is live. Here is what to explore next:

• Store API keys in the vault — Give your agent access to external services without exposing credentials. Read more about OmniClaw security.
• Customize with skills — Add web search, calendar management, or custom API integrations to your agent.
• Compare hosting options — See how OmniClaw stacks up against self-hosting and container platforms in our hosting comparison.