Summary
Prompt to file, without opening a studio.
It does two things and keeps the receipts: what comes back is a real file in a dated folder with a sidecar recording the settings that produced it, so a take you liked can be reproduced rather than re-guessed.
What it is
Two Gemini generation calls wired into your assistant — text to image and text to speech — with every result saved to disk alongside a file recording how it was generated.
What you get
- An image generated from a text prompt
- Spoken audio generated from text
- Files sorted into image and audio folders by date
- A companion metadata file for each result, recording the generation settings
Requirements
A Google Gemini API key in GEMINI_API_KEY, and pip or uvx to run it. An optional download-path setting decides where generated files land.
Setup effort
One command plus a key — pip install gemini-gen-mcp, then supply credentials