Samy6767f/openclaw — reverse-engineered prompt

Reverse engineered prompt

GitHub

Build me a personal AI assistant app called OpenClaw that I can run on my own devices. I want it to feel local, fast, and always available, not like a public chatbot. It should have a setup flow in the terminal that walks me through connecting my AI account, starting the background service, choosing a workspace, and adding chat channels.

The assistant should be able to receive and send messages through popular apps like WhatsApp, Telegram, Slack, Discord, Signal, iMessage, Google Chat, Microsoft Teams, Matrix, LINE, WeChat, and a simple web chat. It should also support speaking and listening on desktop and mobile where possible, plus show a live canvas area that the assistant can update and I can control.

Please include sensible security defaults so random incoming DMs are not trusted, and make it easy to check status, send a test message, run in debug mode, and update later. Look up current docs online if you need to.

Want more depth? Deep Reverse