web-shader-extractor

name	web-shader-extractor
description	Extract a real WebGL / WebGPU / Canvas / shader effect from a live webpage (e.g. an animated gradient mesh, fluid/metaball, gaussian-splat scene, image distortion) and port it to a standalone, dependency-light JS project you can run locally or drop into an HTML artifact. Use when the user points at a site's visual effect and wants to reproduce/port/replay it, especially complex GPU effects our vanilla recipes don't cover.
triggers	["extract this shader / webgl effect","port this webgl background","reproduce this canvas effect","把这个网站的 webgl/shader 效果扒下来","复刻高斯泼溅 / gaussian splatting","gpu 背景效果 港口"]
od	{"mode":"prototype","surface":"web","platform":"desktop","category":"animation-motion","upstream":"https://github.com/lixiaolin94/skills","design_system":{"requires":false}}

web-shader-extractor (install-on-demand companion)

This is a thin wrapper. The upstream skill has no open license, so motion-anything cannot redistribute its code — it is fetched on demand instead. Everything below is motion-anything's own description; the implementation lives upstream.

What it does

Locks onto a page's rendering surface (WebGL/WebGPU/Canvas), records shader / resource / render-graph / timing / input evidence, builds an evidence-matched local baseline, then projectizes it into a standalone runnable. This is the right tool for the "复杂网页效果 (WebGL / 高斯泼溅)" need that our 2D compositor deliberately does not draw — see WEBGL.md.

How to invoke

1. If not already installed, fetch it (one-time, needs network):

   npx skills add https://github.com/lixiaolin94/skills --skill web-shader-extractor
   

2. Then follow that skill's flow against the target URL / effect.
3. Bring the extracted effect back as a background layer in an HTML artifact; motion-anything adds the surrounding tasteful motion.

Note for maintainers

If lixiaolin94/skills adds a permissive license, vendor it here properly (like skills/web-clone) so it ships offline with the product. Until then, keep this wrapper. Upstream: https://github.com/lixiaolin94/skills