gigglemorgan/giggle-studio ? reverse-engineered prompt
Reverse engineered prompt
Build me a local first web app for Linux that runs on my own machine and helps me work with open weight AI models in one place. I want a clean browser based studio where I can load models, chat with them, and manage inference without it using more memory than it should.
It should also let me fine tune models with LoRA and QLoRA, show me progress and loss while training, and safely export the result when it is done. I want a built in sandbox for running code with clear approval before anything risky happens, plus a section for defensive binary analysis where I can inspect files without executing them.
Make it privacy focused, self hosted, and reliable on modest hardware, with a simple setup flow, diagnostics, and a polished interface. If you need to check current docs for any libraries or best practices, feel free to look them up online.
Are you gonna build this?
make sure you review the code using coderabbit