Labubu AI Generator: Turn Photos Into 3D Doll Sets

What Ribbi did

Ribbi reads the photo and breaks down the outfit on its own — beige knit crop top, light-blue wide-leg jeans, neon-blue clogs, sky-blue cap and a blue tote — pins the palette as a fresh yellow-blue-white Y2K street look, and rewrites it all as a 3D designer-toy doll. Following the user's "one shot to confirm first" pacing, it delivers exactly one eye-level poster: crosswalk backdrop, collage sticker details all in place. Direction approved, then move on.

What Ribbi did

With the eye-level shot approved, one "write it into the skill" gets that poster saved as the skill's cover. Then a single OK sets three camera positions — low angle, top-down and side view — running in parallel until the four-angle set is complete: same doll, same outfit, same crosswalk, and the character doesn't drift an inch between angles. Ribbi closes by asking: want the four stills cut into a camera-move video?

What Ribbi did

The first camera-move cut felt stuttery, so the user pulls up the prompt to vet it line by line — Ribbi hands over the full camera plan, cuts the fast snaps per the user's notes, adds a 0.2-second slow-motion hold after every transition, and provides a Chinese version for easy review. The re-run glides through all four angles seamlessly. The user signs off on the spot: the tuned camera plan and the "one shot to confirm, then fill in the rest in parallel" workflow both go back into the skill — with the pitfalls logged as warnings so they never get hit again.