Run Your Own AI Assistant with OpenClaw and Chat from Your Phone

What you will build

A personal AI assistant that lives on your own server and answers you in the chat app you already use.

Starting point: you only need a Vultr account. This challenge is self-contained, so you can do it without having done any other.

The plan

1. Deploy OpenClaw from the Vultr Marketplace, or with the Docker setup script on an Ubuntu 24.04 instance
2. Add a model: your own provider API key or Vultr Serverless Inference
3. Create a bot in your chat app (Telegram is the quickest) and give OpenClaw its token
4. Set an allowlist so only your own account can message the bot
5. Open the dashboard through an SSH tunnel or HTTPS, never plain HTTP over the internet
6. Chat with the assistant from your phone and screenshot the conversation
7. Ask it to do one small, safe task and describe what it did

git clone https://github.com/openclaw/openclaw.git
cd openclaw && ./docker-setup.sh
docker compose logs openclaw-gateway

How you will be judged

• Strangers cannot use or command your assistant
• Secrets stay out of screenshots and repositories
• You can explain what the assistant is allowed to do and what it is not

An AI agent can act on what it is told. Give it the least access that works, keep the allowlist tight, and treat the gateway token and API keys as passwords. Revoke the bot token and destroy the server when you are done.