skills.rest
runcomfy-com
runcomfy-comOFFICIAL
Explore all skills by this author →
Website𝕏@runcomfyaiGitHub

image-to-video

Route still images to RunComfy image-to-video models for animated video output.

12|2|Updated May 18, 2026
Search Tags:#multimodal#video generation#image-to-video#lip-sync#media automation#prompt routing#runcomfy
One-click install
npx skills add https://github.com/runcomfy-com/skills --skill image-to-video-runcomfy-com
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: image-to-video
Source: https://github.com/runcomfy-com/skills/tree/main/image-to-video
Command: npx skills add https://github.com/runcomfy-com/skills --skill image-to-video-runcomfy-com

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Converting a still image into a compelling video is hard because the “right” model depends on the user’s goal, such as portrait animation, product-style motion, lip-synced voiceover, or multimodal scene composition.

Core Features & Use Cases

  • Intent-routed i2v model selection: Automatically matches the user’s intent to the best available RunComfy image-to-video route and model (default portrait/product animation, custom-audio lip-sync, or multimodal image+video+audio).
  • Bundled prompting patterns per model: Uses the model-specific prompting approach to improve visual quality and reduce wasted iterations on the wrong setup.
  • Flexible input schemas for different workflows: Supports image_url for stills, audio_url for lip-synced talking-head results, and video_url/audio_url arrays for reference-driven multimodal clips.

Quick Start

Use the image-to-video skill to animate the file URL portrait.jpg into a short portrait video by matching natural facial motion and stable identity.

Frequently Asked Questions about image-to-video

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

FAQPage Schema
How do I convert a still image into a video with custom audio lip-sync?▼

To convert a still image into a lip-synced video, route the input image and a valid audio_url to the matching RunComfy image-to-video model endpoint. The model processes the multimodal inputs to generate an animated video output with synchronized voiceover.

How does intent-routed model selection work for image-to-video generation?▼

Intent-routed image-to-video generation automatically matches your goal, such as portrait animation or product motion, to the best available RunComfy model. It applies model-specific prompting patterns and input schemas to generate high-quality animated clips.

Can I use a reference video and reference audio for multimodal image-to-video composition?▼

Yes, multimodal image-to-video composition supports combining an input image with reference video_url and audio_url arrays. The RunComfy CLI executes the matching model endpoint with the appropriate JSON schema to generate reference-driven clips.

Do I need specific URLs and parameters to generate an image-to-video clip?▼

Yes, image-to-video generation requires valid image, audio, or video URLs alongside model-specific parameters like duration, aspect ratio, and prompts. You must execute the RunComfy CLI with the matching model endpoint and appropriate input JSON schema.

What is the best way to animate a portrait image into a short video?▼

The best way to animate a portrait image is to route the image_url to the default portrait animation RunComfy model. This approach applies bundled prompting patterns to match natural facial motion while maintaining stable identity output.

Related Skills

Explore More Design & Creative Skills→
runcomfy-comruncomfy-com
12

seedance-v2

Generate cinematic short-form videos with native lip-synced audio via RunComfy.

Official
Intermediate
agentspace-soagentspace-so
31

ai-avatar-video

Create avatar videos from audio or scripts via RunComfy model routing.

Official
Advanced
prime-skillsprime-skills
31

wan-2-7

Generate text-to-video clips with Wan 2.7 via the RunComfy Model API.

Official
Intermediate
prime-skillsprime-skills
31

image-to-video

Routes still-image animation requests to the appropriate RunComfy video generation model.

Official
Advanced