What problem does it solve? Choosing the wrong video-edit model wastes iterations and credits. This Skill classifies the user's editing intent and routes it to the right RunComfy model — Wan 2.7 Edit-Video, Kling 2.6 Pro Motion Control, or Lucy Edit Restyle — with each model's documented prompting patterns built in. ## Core Features & Use Cases - Intent-based model routing: Matches requests like restyle, background swap, packaging swap, motion transfer, or outfit swap to the best-fit model in the RunComfy catalog. - Bundled prompting patterns: Ships per-model prompt templates (preservation-first phrasing, one edit direction per call) so edits succeed with fewer retries. - CLI-driven execution: Invokes runcomfy run <vendor>/<model>/<endpoint> with structured JSON input and downloads results to a local output directory. - Use Case: A marketer wants to swap the background of a talking-head product video while keeping the speaker's face and lip movement intact. The Skill routes to Wan 2.7 Edit-Video with audio_setting: "origin" and a preservation-first prompt. ## Quick Start Ask the agent to restyle your video's background to a modern office while preserving the speaker's face and motion, providing the source video URL.
