Live2d Halfbody

从一张立绘或一段角色描述，做出能在 VTube Studio 里直播用的半身 Live2D 皮套，不用手动建模、不开 Cubism Editor：图像模型出图、See-through 拆层、代码补细节和绑定、自查后装进 VTS。默认带小角度转头、歪头、身体跟随、呼吸、眨眼和单眼眨眼、口型、四个表情、头发分段摆动（短发、过肩长发或双马尾，刘海可有可无，侧发可别在耳后）、挂件和耳坠摆动、眼镜、招手/摊手/指向三个手势和脸红。用于用户想新做一个皮套（有立绘、只有描述、或已有分层文件）、检查皮套素材、或调整已有角色的动作幅度时。Builds a half-body Live2D avatar for VTube Studio from one character illustration or a text description.

29 stars 7 forksImages, Video & Audio

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add bellisji233/live2d-halfbody-skill --skill live2d-halfbody

Manual install

git clone --depth 1 https://github.com/bellisji233/live2d-halfbody-skill.git /tmp/live2d-halfbody-skill
mkdir -p ~/.claude/skills
cp -r /tmp/live2d-halfbody-skill/. ~/.claude/skills/live2d-halfbody

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# 半身皮套：从一张立绘到 VTube Studio

给不想手动建模、不想调网格的人。一张立绘或一段角色描述进去，Agent 一步步做出能在 VTS 里直播用的半身 Live2D 皮套。用户只做三件事：

1. **确认立绘**（第 1 步）；
2. **同意把立绘传到 See-through**（第 3 步，Hugging Face，备用魔搭）；
3. **在 VTS 里看效果**，回复 `通过` 或 `不通过：<现象>`（第 5 步）。

其余由 Agent 完成，并在交给用户前自查。

## 跑一次就有的功能

| 功能 | 驱动 |
|---|---|
| 转头左右、上下（±30），歪头时身体跟着倾斜 | 摄像头 |
| 身体跟随：左右转、前后、呼吸起伏 | 摄像头、VTS 自动呼吸 |
| 眨眼、单眼眨眼（快闭上时换成画好的闭眼）；半闭眼时双眼皮淡出 | 摄像头（左右眼分开） |
| 眼珠跟随、眉毛上下 | 摄像头 |
| 张嘴说话、嘴角笑 | 摄像头 |
| 头发分段摆动：刘海（可选）、两侧（三段，从发根到发梢一段跟一段；可别在耳后）、后发或发髻；点头时上下弹。双马尾每条从发绳往下单独摆，同样分三段。过肩长发在肩膀以下跟身体走 | 物理 |
| 挂件（项链吊坠、耳坠）摆动 | 物理 |
| 眼镜跟着脸转 | 摄像头 |
| 表情：笑眼 N1、O 型嘴 N2、急眼 N3、困扰 N4，清除 N0 | 快捷键 |
| 手势：招手 N5、摊手 N6、指向 N7 | 快捷键（也配了摄像头手势，见“已知问题”） |
| 脸红 N8 | 快捷键 |

做不到：大角度转头（要侧脸素材）、全身、手臂跟着摄像头连续动（要分段手臂素材）。产出是 VTS 直接用的 `.moc3`，不是 Cubism Editor 的工程文件（`.cmo3`），Editor 打不开。精修级的效果（更细的物理、更多表情、商业级网格）要把 `layers-st/` 的图层导进 Cubism Editor 重新绑定。

## 文件夹

```
素材/角色名/                 立绘、提示词、变体、See-through 结果、图层
  A0-base.png                定稿立绘
  prompts/                   从 references/提示词/标准/ 复制，填好 [[…]]
  check/                     每张变体的对位检查图
  seethrough.psd             See-through 的原始结果
  seethrough/                拆成每层一张 PNG
  layers/                    自拆的图层：只作 See-through 合成的底，不单独交付
  layers-st/                 最终图层（See-through 合成，含手势和脸红）
  R0-rest.png                最终图层的静止合成，手势和脸红变体的底图
角色/角色名/
  character.json             位置、幅度、图层对照
  标注.png                    character.json 里的位置画在立绘上
  素材检查/  prepared/  build/runtime/  自查/  vts/
```

## 步骤

Read the full skill on GitHub

Details

Skill name: live2d-halfbody
Repository: bellisji233/live2d-halfbody-skill
License: MIT
Repo updated: 30 October 2026

FAQ

What does the Live2d Halfbody skill do?

How do I install the Live2d Halfbody skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add bellisji233/live2d-halfbody-skill --skill live2d-halfbody. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Live2d Halfbody popular?

It lives in bellisji233/live2d-halfbody-skill, which has 29 GitHub stars and 7 forks.

Related skills

Zackd Director

Anil-matcha/zack-d-films-ai-video-generator

Turn ONE curiosity-driven topic or question into a finished Zack D Films-style 3D animated short video, end to end on the MuAPI platform + local ffmpeg — curiosity-loop script, character consistency sheets, 3D keyframe rendering, motion clips, cloned voiceover, zoom-ins, transitions, and captions, all automated. Use this whenever the user wants a "Zack D Films style short", a 3D medical / anatomical / educational explainer short, a curiosity-driven short video, or wants to automate 3D animated short production with an API-backed media pipeline. Triggers: "zack d films short", "zack d style", "

Images, Video & Audio

Blender Claymation

angrypenguinpng/blender-claymation-skill

Make stop-motion claymation videos in Blender from a one-line idea. Builds a miniature clay set and puppet in Python, animates it "on twos" at 12 fps with hand-made boil, renders with Cycles, encodes an mp4, and presents it in a video player. Use when the user asks for a claymation, stop-motion, plasticine or clay animation, a clay character or short film made in Blender, or "an animation that looks like a real studio made it".

Images, Video & Audio

Read Anything

heyuxuan0209/read-anything

即时分析：任意链接（网页/博客/YouTube/B站/小宇宙播客/X）或本地录音（会议录音/语音备忘录）→ 抓取或本地转写 → 中文结构化解读（精读稿/快扫/访谈拆解/会议纪要/论文拆解/选题素材六种模板+自定义）→ 基于材料的多轮问答。触发：用户丢链接/音频文件/粘贴纪要并说"分析/解读/精读/这是什么/讲了啥/值得看吗/提选题"，或明确说"即时分析"。

Images, Video & Audio

FigFox Gen Skill

LawrenceRiver/FigFox-Gen-skill

Use when turning a scientific Methodology, with an optional reference image, into one evidence-grounded, human-editable final PNG figure.

Images, Video & Audio

Douban Sync

cosformula/douban-sync-skill

Export and sync Douban (豆瓣) book/movie/music/game collections to local CSV files (Obsidian-compatible). Use when the user wants to export their Douban reading/watching/listening/gaming history, set up incremental sync via RSS, or manage their Douban data locally.

Images, Video & Audio

Image to Prompt

jiemianduan/image-to-prompt

Analyze one or more reference images and reconstruct faithful, reusable image-generation prompts in Chinese and English. Use when users want to reverse-engineer, infer, restore, imitate, or write prompts from visual references.

Images, Video & Audio

Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.