What problem does it solve? Image-to-video models often break typography, paper texture, framing, and layout when animating still images, producing text drift, breathing paper, and unwanted camera movement. This Skill inspects the actual image, decides whether it should use motion, micro-motion, or remain static, and writes a bounded, copy-ready prompt that locks fragile design elements. ## Core Features & Use Cases - Motion Decision Framework: Classifies each image as motion, micro-motion, or static based on subject, physical cause, composition, and fragile elements, selecting at most one primary action and one linked response. - Prompt Construction with Lock Lists: Generates platform-ready prompts (e.g., for Jimeng or Seedance) with explicit fixed-element lists, negative constraints, range, timing, and a default restrained 4-second duration. - Failure Diagnosis and Repair: Rewrites failed prompts by identifying text, paper, camera, or layout drift and reducing active regions instead of adding more instructions. - Use Case: Given a folder of editorial posters, the Skill inventories the images, classifies each one individually, and outputs a separate Chinese image-to-video prompt per poster that keeps all typography and paper texture fixed. ## Quick Start Ask the AI to use gc-still-image-motion-director to analyze your attached poster, decide how it should move while locking text and paper, and return one copy-ready image-to-video prompt.
