GitHub radar
Get your AI agent to animate art styles, not just shaky pans
This is a skill for your coding agent: it writes code that actually draws and moves a scene, in an art style or an explainer-video format. You get finished video without hand-drawing every frame.
huashu-art-motion is a skill for your coding agent. It writes code that truly draws and moves a scene in a chosen style, instead of just panning a still image.
When it helps
Useful if you make explainer videos or art-style clips through a coding agent. Skip it if you don't make video or don't use a coding agent.
Pros
- 35 ready art-style cards — from cave painting to Van Gogh
- 8 of 9 explainer formats include demo clips and a frame spec
- Runs on an ordinary laptop — no GPU needed
Cons
- Finance format is reference code only, no demo, needs your own characters
- Documentation is in Chinese
- Character shots still need a separate image-generation model
How to set it up — step by step
- 1Open your coding agent in your project folder.
- 2Send the agent the text from the block below.
- 3Tell it which art style or explainer format you want, plus your own script.
- 4If you need character shots, give the agent a key for an image-generation model.
- 5Review the finished video and ask the agent to refine details.
Text for your agent
Copy this and send it to your agent — Claude Code, Codex, any of them:
Install this animation skill for my agent from https://github.com/alchaincyf/huashu-art-motion — add it with `npx skills add alchaincyf/huashu-art-motion`, then follow SKILL.md to render a short test clip in one of the art styles based on my description, and show me the resulting video. If character frames need an image-generation model, ask me for the key.
A regular laptop — no GPU needed, just the uv Python tool, ffmpeg and Chromium (one-time install). Character frames need a separate image-generation model.
Open on GitHub▌ More finds