face-swap

Replace faces in images or videos via RunComfy CLI routes.

5|2|Updated May 18, 2026
One-click install
npx skills add https://github.com/doany-ai/skills --skill face-swap-doany-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: face-swap
Source: https://github.com/doany-ai/skills/tree/main/face-swap
Command: npx skills add https://github.com/doany-ai/skills --skill face-swap-doany-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Face swaps enable substitution of a subject's face in images or videos using RunComfy's AI models, empowering creators to explore identity changes while preserving scene context.

Core Features & Use Cases

  • Identity-preserving face swaps across stills and videos using multiple routing options (Wan 2-2 Animate, Kling Motion Control Pro, GPT Image 2 Edit, Nano Banana Edit, Flux Kontext)
  • Preserve pose, lighting, and background while swapping faces to ensure natural results
  • Use cases include campaign visuals, actor substitutions with consent, or stylized character swaps in media production

Quick Start

Provide a target identity image or description and a source media, then run the RunComfy CLI to perform the face swap.

Frequently Asked Questions about face-swap

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform an AI face swap on a video while preserving motion?

Perform an AI face swap on a video by providing a target identity image and source media to the RunComfy CLI. The process preserves motion and scene context by routing through models like Wan 2-2 Animate or Kling Motion Control Pro.

Can I swap multiple faces in a batch of images using RunComfy?

Yes, you can swap multiple faces in a batch of images using RunComfy's CLI. The skill supports batch operations for still images, allowing you to process multiple targets efficiently while preserving original lighting and backgrounds.

What AI models are available for identity-preserving face swaps in images?

Available AI models for identity-preserving face swaps include GPT Image 2 Edit, Nano Banana Edit, and Flux Kontext. These routing options allow you to replace faces in still images while maintaining the original portrait lighting and pose.

Does the face swap process require separate source and target images?

Yes, the face swap process requires a target identity image or description and a source media file. Providing these inputs to the RunComfy CLI enables the AI to substitute the subject's face while preserving the original scene context.

What's the best way to ensure natural lighting and background preservation during a face swap?

To ensure natural lighting and background preservation during a face swap, use the RunComfy CLI with its supported routing options. The AI models are designed to maintain pose, lighting, and background context while substituting the subject's face.

Are there limitations when using AI face swaps for actor substitutions in media production?

AI face swaps for actor substitutions require explicit consent and are limited to identity changes within supported image and video formats. The process focuses on preserving scene context but should be used ethically for campaign visuals and stylized character swaps.