Vision Assistant.

This Assistant you tell the bot what you want a/aind then copy the positive prompt. Paste the prompt on the image generation page, done!