yangsuzhou/wifi-densepose — reverse-engineered prompt

Reverse engineered prompt

GitHub

Build me a working prototype of a privacy first WiFi sensing app like π RuView. I want a web dashboard that can show live human presence, body pose, breathing rate, and heart rate from WiFi signal data, with no cameras and no wearables.

Make it practical for someone to try right away. If real CSI hardware is connected, support an ESP32 S3 style sensor stream. If no hardware is available, include a demo mode with realistic sample signal data so the dashboard still works. The app should explain clearly that normal laptops can only do basic RSSI presence detection, while full pose and vitals need CSI capable hardware.

Include a simple setup flow, status indicators for sensors, live charts for signal quality and vital signs, and a clean room view that draws people as skeletons or outlines. Package it so I can run it locally in Docker and open it in the browser. Please add basic tests and a short README with setup steps. Look up current docs online if you need to.

Want more depth? Deep Reverse