video-inpainting

Official

Remove objects and clean video regions.

Authorruncomfy-com
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Video inpainting removes unwanted regions across multiple frames—like watermarks, wires, or passing people—while preserving the rest of the clip and matching surrounding motion.

Core Features & Use Cases

  • Prompt-driven region edits across frames: Remove or replace an object that appears across many frames using spatial language (e.g., bottom-right corner).
  • Route selection for intent and consistency: Chooses between Wan 2-7 edit-video (default), Lucy Edit Restyle (identity-stable region-aware restyle), and Seedream 4-0 edit-sequential (frame-stack style) based on whether the user intent is prose-driven, identity-locked, or best treated per-frame.
  • CLI-first workflow: Executes edits via the RunComfy CLI and downloads results to an output directory.
  • Related best-path guidance: For pixel-precise, mask-propagation workflows, points to dedicated ComfyUI graphs instead of CLI endpoints.

Quick Start

Use the video-inpainting skill to remove a watermark from the bottom-right corner across all frames and preserve everything else exactly.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: video-inpainting
Download link: https://github.com/runcomfy-com/skills/archive/main.zip#video-inpainting

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.