eauesque/yu_ai_manager — reverse-engineered prompt
Reverse engineered prompt
Build me a local Web UI for managing AI generated images and their prompts. I want to point it at folders of images, have it read the hidden metadata, and let me browse, search, filter, and organize everything without uploading anything to the cloud.
It should understand common image sources like Stable Diffusion from A1111 and Forge, NovelAI V3 and V4, and ComfyUI. Show the prompt, negative prompt, model info, seed, sampler, size, and any other useful generation settings in a clean way. Please make it friendly for normal users, with English, Japanese, Chinese, Korean, German, Spanish, French, Italian, Portuguese, and Russian support.
Make startup simple with Windows and Mac or Linux launch scripts that handle the local Python setup and rebuild the web interface if needed. Optional tools like ffmpeg can be prompted for but the core app should run without them. Look up current docs online if you need to.
Want more depth? Deep Reverse