Routes you across 10+ video models with one CLI so you don't have to pick between HappyHorse, Kling, Seedance, Veo, and the rest every time you need a clip. It handles text-to-video, image-to-video, and video extension, and the skill actually picks the right model for your intent: HappyHorse 1.0 for in-pass audio, Wan 2-7 for audio-driven lip sync, Veo 3-1 for physics-accurate motion, Kling 3.0 4K when you need hero shots. The prompting patterns and runcomfy invocations are all documented per model. Honest take: this is the skill you want if you're shipping video at any volume and don't want to remember which vendor does what.
npx skills add https://github.com/agentspace-so/runcomfy-agent-skills --skill ai-video-generation