kevinbadi/jev-voice ? reverse-engineered prompt
Reverse engineered prompt
Build me a macOS app that lets me control my Mac with my voice.
I want it to listen locally, wake up when I say a name like Alfred or Jarvis, and then do things like open apps, search the web, type text, scroll, press keys, play music, take screenshots, and lock the screen. It should also support a hands free mode, a hold to talk mode with Caps Lock, and a push to talk terminal mode.
Please make the setup easy on Apple Silicon, with a script that installs everything it needs, downloads the speech model, sets up the launcher, and helps with the Mac permissions for microphone, accessibility, and input monitoring. I also want it to say out loud when it finishes a command.
If a command needs multiple screen steps, it should be able to look at the frontmost window and keep going until it is done. Keep it local as much as possible, and use the external AI only for choosing the right action, not for generating random text.
Are you gonna build this?
make sure you review the code using arcumet