# Animate Still Images with the Right RunComfy Model

Choosing an image-to-video model can waste time and credits. This skill routes each request to a suitable RunComfy model and provides focused prompting guidance.

## Install

```bash
npx skillstore add skills-collective/image-to-video
```

## Metadata

- Status: approved
- Slug: skills-collective-image-to-video
- Skillstore revision: r2
- Version status: missing
- Tree hash: 9c50b2b32907fb02cb6ae3b02f8a7328ca6eaf4054b376100afa5591b5492007
- Author: skills-collective
- GitHub username: skills-collective
- License: MIT
- Repository: https://github.com/skills-collective/skills/tree/main/image-to-video/
- Ref: 9e952417e76879bc9d853e1b8b2cd6d6d8d4a1c2
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, network, filesystem
- Quality score: 50
- Quality tier: warning
- Public page: https://skillstore.pages.dev/skills/skills-collective-image-to-video
- Manifest: https://skillstore.pages.dev/api/skills/skills-collective-image-to-video/manifest

## Capabilities

- Routes portrait, product, and general animation requests to HappyHorse 1.0 I2V.
- Routes custom voiceover lip-sync requests to Wan 2.7 with an audio URL.
- Routes combined image, video, and audio reference workflows to Seedance 2.0 Pro.
- Provides model-specific duration, resolution, aspect ratio, seed, and media input guidance.
- Builds motion-focused prompts and invokes the selected model through the RunComfy CLI.

## Use Cases

- Create Portrait Motion: Animate a portrait with controlled facial movement, stable identity, and a simple camera move for short social videos.
- Build Product Reveals: Turn a product image into a short reveal with preserved packaging, smooth camera motion, and deliberate lighting changes.
- Produce Localized Spokesperson Clips: Create lip-synced variants by reusing a visual direction and changing the voiceover URL for each language.

## Prompt Templates

### Animate a Portrait

```
Animate this portrait from [image URL] with subtle breathing, blinking, and gentle camera drift. Create a five-second vertical clip with stable facial features.
```

### Create a Product Reveal

```
Animate [product image URL] with a slow orbit and rising rim light. Keep the packaging, logo, proportions, and background geometry unchanged.
```

### Lip-Sync a Voiceover

```
Create a vertical spokesperson clip using [audio URL]. Match the duration to the voiceover, keep the camera locked, and avoid subtitles or flicker.
```

### Combine Multimodal References

```
Use [image URL] for the subject, [video URL] for scene motion, and [audio URL] for voice tone. Create an eight-second coordinated clip.
```

## Limitations

- Requires the RunComfy CLI, a RunComfy account, and accessible source media URLs.
- Selects one model per call and does not blend routes or stitch separate clips.
- Clips are limited to each model's supported duration, resolution, aspect ratio, and media sizes.
- Prompts and media references are processed by third-party RunComfy infrastructure.

## Best Practices

- Lead with one clear motion or camera action and describe only what should change.
- State preservation requirements for faces, products, logos, costumes, and background geometry.
- Match duration to reference audio and reuse seeds when comparing controlled variants.

## Anti Patterns

- Do not combine several competing camera moves or character actions in one short clip.
- Do not restate every visible detail when motion, lighting, and preservation instructions are more useful.
- Do not submit sensitive, unlicensed, or inaccessible media URLs to third-party model services.

## Security Audit

- Audited at: 2026-07-24T01:35:14.778\+00:00
- Summary: Most static alerts are false positives caused by Markdown backticks, placeholder URLs, and standard configuration paths. Eight findings confirm intended installation or execution of external CLI commands. The skill also sends prompts and user media references to a third-party model service.

## Stats

- Views: 0
- Downloads: 1
- Favorites: 0
- Popularity score: 0