video-extend

Extend Veo video clips by generating continuation frames via RunComfy CLI.

12|2|Updated May 18, 2026
One-click install
npx skills add https://github.com/runcomfy-com/skills --skill video-extend
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: video-extend
Source: https://github.com/runcomfy-com/skills/tree/main/video-extend
Command: npx skills add https://github.com/runcomfy-com/skills --skill video-extend

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you extend a short generated video clip without losing continuity, so you can turn a brief Veo render into a longer scene or chain multiple shots into one coherent narrative.

Core Features & Use Cases

  • Extends existing Veo clips: continues an existing video past its per-call duration cap while preserving motion, lighting, and subject identity.
  • Chains shot-by-shot narratives: uses the output of one extend call as the next input to build longer sequences from a single seed clip.
  • Routes to the right Veo endpoint: selects between Veo 3-1 Extend for quality and Veo 3-1 Fast Extend for faster, lower-cost iteration.

Quick Start

Use the video-extend skill when you have a source Veo clip URL and want the camera and characters to continue exactly as described in your continuation prompt.

Frequently Asked Questions about video-extend

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

FAQPage Schema
How do I extend a short video clip without losing character identity or lighting continuity?

To extend a video clip without losing continuity, you generate additional frames from a source video URL using a continuation prompt. This preserves the original motion, lighting, and subject identity while lengthening the scene.

Can I chain multiple video shots together to build a longer narrative from a single Veo clip?

Yes, you can chain multiple video shots together by using the output of one extend call as the input for the next. This builds a longer, coherent multishot narrative from a single seed clip.

What is the best way to continue a Veo 3-1 generated video past its duration cap?

The best way to continue a Veo 3-1 video past its duration cap is routing the source video URL to Veo 3-1 Extend endpoints. You pass the source clip plus a continuation prompt to generate additional frames.

Does the video extension process support faster, lower-cost iterations for Veo clips?

Yes, faster and lower-cost iterations are supported by routing the source video URL to the Veo 3-1 Fast Extend endpoint instead of the standard Veo 3-1 Extend endpoint, trading some quality for speed.

How do I pass a source video and continuation prompt to lengthen a video using runcomfy?

To lengthen a video using runcomfy, you route the source video URL and continuation prompt to the Google Veo 3-1 extend-video endpoints via the runcomfy run CLI, downloading the result to an output directory.

When should I not use an AI video extension approach for my project?

You should not use AI video extension if your project requires exact frame-by-frame manual control or if your source video URL lacks the clear motion and lighting needed for the model to generate consistent additional frames.