davidoberst/Bit ? reverse-engineered prompt
Reverse engineered prompt
Build me Bit, a Linux voice first AI copilot that I can talk to hands free.
It should listen to my microphone, respond with speech, and be able to look at my screen when it needs more context. I want a simple black and white desktop interface with an audio reactive feel, and the whole thing should run locally in Python on Linux. Keep the conversation stateful so it remembers what we were talking about during the session.
Use Gemini 2.5 Flash for the reasoning path, and if it makes sense, leave room for a local LLM based version too. It should be able to trigger safe system actions, handle screen capture, mic input, and text to speech cleanly. Please make it feel like a practical copilot for real work, not a chat demo.
If you need to check current docs or APIs, look them up online and wire it up in the simplest reliable way.
Are you gonna build this?
make sure you review the code using coderabbit