Moez-lab/JARVIS-AI-Assistant ? reverse-engineered prompt
Reverse engineered prompt
Build me a Jarvis style AI assistant that runs with a Python backend and a 3D web interface.
I want it to listen for a wake word, answer with a natural sounding voice, and show a realistic avatar on screen that moves its mouth in sync with speech. It should use the webcam to recognize me, track my face, and detect objects or people in view, then describe what it sees in plain language. If it does not know a face, it should warn me.
Please make the frontend feel futuristic and polished, with a live 3D character, simple controls to tweak the avatar look, and a realtime connection to the backend so the face tracking and lip sync stay smooth. Also include basic desktop automation and system monitoring so I can ask it to do simple tasks on my computer.
If you need to check current docs for any of the libraries, go ahead and look them up online.
Are you gonna build this?
make sure you review the code using coderabbit