VibePrompt

locks wardrobe, world, lighting and selfie type to whatever's showing in the frames above, then rolls a fresh pose, expression, capture angle and grain for each shot — the way a real multi-photo post from one moment actually looks, instead of every shot changing scene at once.

builds one {a|b|c} master prompt for ComfyUI's Random Prompts / dynamic-prompts node, filtered to the current selfie type, lighting mode, and anatomy/quality settings above. Flat wildcard syntax can't preserve the world↔outfit↔lighting linking this tool enforces live, so outfit/lighting/world are pooled independently — expect the occasional mismatched combo (e.g. bikini + snowy street) in bulk ComfyUI batches, unlike the guaranteed-consistent single roll above.