Paper Collage Remotion

SKILL前端设计
前端设计 未标注权限 许可证不明确 已核验
安装这个技能

在 Remotion 中创建、修改或渲染分层剪纸/拼贴动画,适用于背景、人物、道具和前景装饰等视频。

一句话选型结论

按资料字段整理

Paper Collage Remotion是前端设计类的 Agent Skill,安装方式:ClawHub 市场,权限级别未标注,许可证不明确,有可直接复制的安装命令。技能本质是给模型读的说明书,效果取决于模型与客户端;装完先用一个小任务验证,再放进正式流程。

分类前端设计 安装方式ClawHub 市场 权限级别未标注权限 许可证许可证不明确
累计安装 未标注 本周 未标注
GitHub 星标 未标注 仓库 未标注
下载量 未标注 技能市场下载数
权限级别 未标注权限 安装前请确认风险范围
许可证 许可证不明确 未说明

安装

ClawHub 市场
在支持 Skills 的客户端里执行,装完问一句「你有哪些技能」验证
skillhub install user_9fe56694/paper-collage-remotion
  1. 1
    确认客户端支持

    确认你的客户端支持 Skills(Claude、Codex、Cursor 等),并升级到较新的版本。

  2. 2
    拿到技能文件

    执行上面的命令安装;也可以把仓库里的 SKILL.md 手动放进客户端的 skills 目录。

  3. 3
    刷新技能列表

    重新打开客户端,或刷新技能列表,让新放进来的技能被加载。

  4. 4
    验证是否生效

    问一句「你有哪些技能」,确认列表里出现它,再拿一个小任务试跑。

技能档案

前端设计
字段来自技能仓库与公开索引,中文说明由本站标注
作者 / 维护方user_9fe56694仓库 未标注 · 版本 未标注
宿主客户端未标注技能可被这些客户端加载
安装方式ClawHub 市场安装包 clawhub
权限级别未标注权限安装前请确认风险范围
许可证未说明许可证不明确,商用前请自行确认条款
仓库可用性available技能仓库当前是否可访问
最近核验未标注上游最后核验时间 · 数据更新 1791680581
正文体量7.5 KBSKILL.md 原文大小,越大说明说明书写得越细

典型使用场景

3 类
按分类整理的常见用法,具体以 SKILL.md 为准
出页面与组件

按描述生成页面结构和组件草稿。

调整样式与布局

改间距、配色与响应式布局。

生成设计说明

把设计稿整理成开发能接的说明。

选择建议

装之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 已做过站内核验,页面字段与仓库对得上。
  • 有可直接复制的安装命令,不用自己翻文档找入口。

先想清楚的地方

  • 没有标注权限级别,实际能做什么要装好后确认。
  • 许可证不明确,商用或二次分发前先到仓库确认条款。
  • 技能只是说明书,实际效果取决于模型与客户端,装完先用小任务试一遍。

更适合

  • 需要在前端设计场景里给模型补一套做法的人。
  • 愿意先读一遍 SKILL.md、确认它会让模型做什么的人。

不太适合

  • 对许可证要求很严、必须明确可商用的人。
  • 只想装完就自动变强、不打算看说明书的人。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
技能权限级别安装方式累计安装宿主客户端安装命令
Paper Collage Remotion当前 未标注权限 ClawHub 市场 未标注 未标注 有
前端设计 只读 GitHub 仓库 882.3 万 Codex、Claude、Cursor 有
Vercel 组合模式 联网访问 GitHub 仓库 385.9 万 Codex、Claude、Cursor 有
Vercel React native skills 前端 只读 GitHub 仓库 246.9 万 Codex、Claude、Cursor 有
adapt 搜索 浏览器控制 GitHub 仓库 202.7 万 Codex、Claude、Cursor 有
使用我的浏览器 浏览器控制 GitHub 仓库 184.8 万 Codex、Claude、Cursor 有

SKILL.md 全文

7.5 KB
模型实际读到的说明书原文,来自技能仓库

Paper Collage Remotion

Build depth from independently movable layers, not by applying camera motion to a single flattened illustration. Use Imagegen or another image model only for source assets; use Remotion as the timing, compositing, and rendering system.

For a new project, copy assets/remotion-starter/ into the project root. Fill script.json, add public assets at the declared paths, and run scripts/validate_project.py script.json public before opening Remotion Studio. Use examples/tang-dynasty/ only as a concrete visual reference, never as a required subject or style.

1. Establish the edit before generating assets

Turn the script into a shot list. For every shot, record: narrative beat, duration, focal subject, viewpoint, subject order, layer list, and caption/audio cue. Design around one dominant subject per shot.

Use this depth order unless the shot has a deliberate exception:

  1. background plate: paper texture, landscape, architecture
  2. distant layer: remote figures and small environment details
  3. middle layer: supporting characters and props
  4. primary layer: story-critical person or object
  5. foreground layer: kneeling figures, petals, tape, torn-paper edges
  6. captions and graphic accents

Do not put a main character into the background plate. Generate each character/prop as a complete, standalone layer; keep head, hands, feet, and held props inside the canvas.

2. Specify and prepare assets

Generate background plates without people. For each cutout, request a solid chroma background, a defined facing direction, no text/watermark/shadow, and a light paper outline. Prefer a source sheet only when it improves character consistency.

For a green-screen source, remove green before splitting. Tune tolerance against the actual source and inspect hair, sleeves, and thin props:

python "$CODEX_HOME/skills/paper-collage-remotion/scripts/key_to_alpha.py" \
  public/assets/source/close-six-green.png public/assets/source/close-six-alpha.png --tolerance 92

Then split the transparent sheet:

python "$CODEX_HOME/skills/paper-collage-remotion/scripts/split_sheet.py" \
  public/assets/source/close-six-alpha.png public/assets/layers close 6

It supports either --columns or --rows for non-square grids, --padding for a shared gutter, and --no-trim when every cell must retain the same dimensions. Inspect several outputs for clipped limbs and non-transparent backgrounds before animating.

Use an asset manifest in the project rather than scattering raw paths:

type LayerRole = "primary" | "secondary" | "tertiary" | "foreground";
type Layer = {src: string; role: LayerRole; x: number; y: number; width: number; delay: number; z: number};

Use the starter's script.json as the manifest. Keep all scene duration, copy, background paths, and layer values in that file; use code only for reusable rendering behavior. This makes the same animation system usable across subjects.

3. Compose a still first

Place all layers at their final positions without entrances. Make the primary subject visibly largest; decrease size, contrast, and motion with narrative importance. Verify that feet share a plausible ground line, faces and props remain unobscured, and foreground objects create intentional overlap.

Give cutouts a consistent outline and elevation shadow:

filter: "drop-shadow(4px 0 #f5eedc) drop-shadow(-4px 0 #f5eedc) drop-shadow(0 4px #f5eedc) drop-shadow(0 18px 9px rgba(20,15,12,.32))"

Reference all public assets through staticFile() and render images with Remotion's Img component. Keep all movement frame-driven with useCurrentFrame() and interpolate(); do not use CSS transitions or keyframe animations.

Ensure the project entry file calls registerRoot(RemotionRoot); pass that entry file (commonly src/index.ts) to the Remotion CLI, rather than a component file.

4. Animate by depth and narrative role

Stage the build: primary first, then supporting characters, then distant/foreground details. Offset every entrance; avoid a full cast appearing on the same frame. Use Sequence from={delay} for audio/entrance coordination.

Start from these visual ranges at 30 fps and tune to the shot, not to an arbitrary formula:

Role Entrance travel Rise Start scale
primary 78 px 55 px 0.86
secondary 58 px 38 px 0.90
tertiary 38 px 22 px 0.95
foreground 50 px 30 px 0.92

Use eased interpolation for an entrance and a small continuous bob after landing. Keep background drift and push-in near 1%; foreground may move slightly faster than the subject to make parallax readable. Keep positional values inline in the relevant style block when Studio editability matters.

const local = frame - layer.delay;
const roleMotion = {
  primary: {distance: 78, rise: 55, startScale: 0.86},
  secondary: {distance: 58, rise: 38, startScale: 0.90},
  tertiary: {distance: 38, rise: 22, startScale: 0.95},
  foreground: {distance: 50, rise: 30, startScale: 0.92},
};
const enter = interpolate(local, [0, 14], [0, 1], {
  easing: Easing.bezier(0.16, 1, 0.3, 1),
  extrapolateLeft: "clamp", extrapolateRight: "clamp",
});
const travel = roleMotion[layer.role].distance;
const bob = local > 14 ? Math.sin((local - 14) / 11) * 3 : 0;

<Img src={staticFile(layer.src)} style={{
  position: "absolute", left: layer.x, top: layer.y, width: layer.width, zIndex: layer.z,
  opacity: enter,
  translate: `${(1 - enter) * travel}px ${roleMotion[layer.role].rise * (1 - enter) + bob}px`,
  scale: interpolate(enter, [0, 1], [roleMotion[layer.role].startScale, 1]),
  filter: paperCutoutFilter,
}} />

5. Add audio and captions after picture timing works

Let narration duration set each scene duration. Use one caption track at the bottom; keep it away from feet and key props. Mix music below narration. Align impact sounds with primary entrances, whooshes with secondary entrances, and light ticks/pops with tertiary details. Use Audio from @remotion/media and staticFile() for local audio.

6. Preview and accept

Use Remotion Studio for timing and a still render for a fast layout check. Render the MP4 only after reviewing static composition and entrances.

npx remotion studio
npx remotion still <composition-id> --frame=150 --scale=0.25
npx remotion render <composition-id> out/final.mp4
ffprobe -v error -show_streams -show_format out/final.mp4
ffmpeg -ss 5 -i out/final.mp4 -frames:v 1 check-05.png

Check: no cut limbs; correct facing directions; clear size hierarchy; correct z-order; staggered entrances; captions unobstructed; sound onset on its intended frame; audio stream, dimensions, and duration valid.

7. Validate mechanically before review

Run the bundled validator before rendering. It checks the scene schema, asset paths, PNG alpha channels, layer bounds, required role motion, and expected duration.

python "$CODEX_HOME/skills/paper-collage-remotion/scripts/validate_project.py" script.json public

Treat warnings as review items: an intentional off-screen entrance is fine; a final position outside the frame, missing alpha, or absent asset is not.

常见问题

5 问
答案由本站依据条目字段整理
Paper Collage Remotion怎么安装?

在支持 Skills 的客户端里执行:skillhub install user_9fe56694/paper-collage-remotion。安装方式为ClawHub 市场。装完刷新技能列表或重开客户端。

Paper Collage Remotion支持哪些客户端?

资料没有标注宿主客户端,但只要客户端支持 Skills(Claude、Codex、Cursor 等),一般都能加载。

Paper Collage Remotion权限高吗,安全吗?

权限级别为未标注权限。安装前先看它会读取或写入哪些内容。

Paper Collage Remotion收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于技能本身和所用的模型或客户端,正式使用前按作者说明确认。

Paper Collage Remotion用什么许可证?

许可证标注为 未说明(许可证不明确)。商用或二次分发前请到仓库确认条款原文。

热门专题

按场景挑技能
全部专题 →

技能内容以作者仓库为准;分类、权限级别与许可证状态的中文说明为中文整理,最近核验 未标注。