huashu-art-motion

huashu-art-motion

Catalogue stub — full skill: alchaincyf/huashu-art-motion

Decision tree

1. Is the full skill already installed?

   • Global: f=~/.agents/skills/huashu-art-motion/SKILL.md; [ -f "$f" ] && ! grep -q "^das:" "$f" && echo "installed" || echo "pointer"
   • Project: f=.agents/skills/huashu-art-motion/SKILL.md; [ -f "$f" ] && ! grep -q "^das:" "$f" && echo "installed" || echo "pointer"
   • installed → the skill is installed as huashu-art-motion, invoke it and proceed
   • pointer → go to step 2

2. Detect scope, then install:

   [ -e ~/.agents/skills/huashu-art-motion ] && echo "global" || echo "project"
   

Global:

npx skills add alchaincyf/huashu-art-motion --skill huashu-art-motion -g -y

Project:

npx skills add alchaincyf/huashu-art-motion --skill huashu-art-motion -y

Claude Code: send a command as a chat message starting with ! to run it without leaving the conversation.

Heads-up

• Requires: uv; ffmpeg; Playwright Chromium (first run: uv run --with playwright playwright install chromium); Skill text and references are in Chinese
• License: MIT; bundled stroke data is Arphic Public License, fonts are SIL OFL, and the Huashu character artwork and demo videos are demo-use only (not MIT)
• Large download (~154 MB): the skills CLI gives up cloning after 300 s; if it times out, retry with SKILLS_CLONE_TIMEOUT_MS=900000.

Invoke after install

• Skill name: huashu-art-motion
• Trigger phrases: "huashu art motion", "art style animation", "explainer animation grammar", "storyboard to animation video"

What it does

Turns the agent into director, storyboarder, animator, and editor, with code as the paintbrush. The pipeline writes a shot list first (one visual style per film, a colour palette, shot-by-shot nouns and verbs), lints it, renders each shot as a video clip, concatenates, runs frame QA, and only delivers on a green check. It ships a library of implemented art styles and transferable style recipes, plus explainer-animation grammars that can render clips from voice-over parameters. AI-generated frames, existing recordings, and voice cloning are optional. Needs uv, Playwright, and ffmpeg locally; the skill text and references are written in Chinese.

When NOT to use

• Magazine-grade typography and animated design prototypes → use huashu-design
• HTML-to-video rendering of existing web compositions → use hyperframes or remotion
• Interface micro-interactions and UI motion → use design-motion-principles or emilkowalski-skill