Invisible AI Assistant

A Python application that analyzes screenshots using Google's Gemini AI and provides answers via typing or text-to-speech. I recommend to modify the prompts to better fit your own needs. If you found this project useful, please drop a ⭐- it means a lot!

Hotkeys

• ALT+Q: General question (types answer)
• ALT+C: Code completion question (types answer)
• ALT+M: Multiple choice question (speaks answer)
• ALT+T: Translation to english (speaks answer)
• ALT+E: Detailed text explanation (speaks answer)
• ALT+R: Repeats last TTS response
• ALT+ESC: Exits application

Setup

1. Install dependencies: pip install -r requirements.txt
2. Create .env file with your Gemini API key (check images for help)
3. Run the application

Notes

• Requires internet connection
• Beep sounds indicate startup and shutdown
• Warning! Long TextToSpeech answers can take up to 20 seconds to load
• To make it actually invisible, change the script extension to .pyw

Inspired by Cluely AI.