Clean

Named camera moves, checked. A free browser-side linter for AI video platforms that expose named presets: it enforces the One-Move Rule, the four sanctioned simultaneous combinations, the conflict pairs a camera cannot perform at once, the per-shot-type duration bands (including the flash establishing shot that is half a second, not a full one), the check that a stated travel can actually produce its stated end framing, zoom written for a physical move, and the FACS action units with their three-to-four expression ceiling and beat arithmetic. Five lanes over one shot: pick the move from a 28-move catalogue, write the shot prompt, schedule the performance as action units, plan a sequence whose durations sum to the envelope, and triage a take that came back wrong with one of five verdicts and a single-variable retake. Lanes derived from the open-source MIT skill set OSideMedia/higgsfield-ai-prompt-skill - higgsfield-camera, higgsfield-facs, higgsfield-troubleshoot, higgsfield-shotlist-director and higgsfield-motion. Not affiliated with or endorsed by Higgsfield or OSideMedia.

Share

Details

PricingUsage-based + 10% creator margin
Billed model rate$2.20 in / $13.20 out per 1M tokens
Creator margin+10%
Effective rate$2.40 in / $14.40 out per 1M tokens
Security scanClean — skill and frontend scanned
Model gpt-terra
Created2026-08-29
Updated2026-09-03

Every public app is built from a security-scanned skill and must pass a clean scan — skill and frontend — before it can be listed. Have a skill of your own? Turn it into an app — or read the step-by-step walkthrough.