What problem does it solve? AI-generated images and videos often look plausible at a glance but fail against the creative brief, brand constraints, or continuity requirements. This Skill provides a structured review method that judges the actual rendered artifact, applies must-pass quality gates, and proposes the smallest targeted repair instead of blind regeneration. ## Core Features & Use Cases - Rubric-based evaluation: Score images and videos across dimensions like brief fidelity, composition, motion coherence, and technical finish, with must-pass gates that block approval regardless of total score. - Scenario-specific gates: Apply tailored checks for logos, ecommerce product sets, identity/try-on work, UI designs, UGC, product films, and long-form clipping. - Evidence-based repair policy: Map failures to a causal layer (brief, reference, structure, capability, or stochastic) and change one variable per iteration while preserving what already works. - Use Case: After generating five product video variants, use this Skill to rank them against the brief, identify a temporal artifact in the best candidate, and get a single targeted regeneration instruction. ## Quick Start Use the review-content-quality skill to evaluate this generated product video against its brief and tell me the smallest fix needed.
