The Wan Video Generator skill empowers Claude to create professional-grade video content using the state-of-the-art Wan 2.6 models. It allows users to transform static images into dynamic animations (Image-to-Video) and maintain subject consistency across scenes using reference clips (Reference-to-Video). With support for up to 1080p resolution, multi-shot sequencing, and built-in LLM prompt expansion, this skill is ideal for creators needing consistent, high-fidelity AI video generation directly within their development workflow.
主な機能
01Multi-shot cinematic sequencing for complex narrative creation
02Automated prompt expansion for optimized visual quality and movement
03Image-to-Video generation with standard and high-speed flash models
0416 GitHub stars
05Reference-to-Video (R2V) for maintaining subject consistency
06Support for 720p and 1080p resolutions at various aspect ratios
ユースケース
01Create multi-shot storytelling videos with precise control over camera timing and shots
02Animate still character portraits or concept art into cinematic sequences
03Generate consistent B-roll footage by referencing existing character or object videos