jaswanth-mjy/hand-gesture-media-controller — reverse-engineered prompt

Build me a simple Python app that lets me control videos and music with hand gestures through my webcam. I want it to work with common apps like YouTube, Netflix, Spotify, VLC, QuickTime, and browser players by sending normal keyboard controls.

Use computer vision to detect one hand in real time and show a camera window with clear feedback, like how many fingers it sees, what gesture it thinks I’m making, and the current media state. A closed fist should play, an open palm should pause, two fingers should rewind 10 seconds, three fingers should skip forward 10 seconds, and four fingers should trigger Mac shutdown only if that is safe and clearly confirmed.

Please make it stable so random movement doesn’t spam commands. Require the same gesture for several frames and add a cooldown between actions. Include a way to quit with Q, basic setup instructions, and helpful error messages for camera permission problems or when no hand is detected.