design-system-extract
Skill1 starsChanged 2 months agoMITView on GitHub
从品牌规范、代码库、现有网站、Figma 或截图中提取颜色、排版、间距、圆角、阴影、动效和断点等设计 token,并输出符合项目格式的结构化 token 文件。用户要求“提取设计系统”“整理 token”“匹配现有品牌”或从参考物反推视觉规范时使用;无法从来源确认的值不得猜测。
1 other project uses a version of this.
Adopted by
Most recently taken up by
- Onevium/skillsword for word, first seen Sep 4, 2026
Other versions1 project uses a version of this.
Word for word · 1
- Onevium/skills★ 1100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository16 other files it gives its agents.
- narrative-video-directorSkill
- portrait-shoot-directorSkill
- accessibility-auditSkill
- ai-slop-checkSkill
- component-extractSkill
- discovery-questionsSkill
- frontend-aesthetic-directionSkill
- generate-variationsSkill
- hierarchy-rhythm-reviewSkill
- interaction-states-passSkill
- make-a-deckSkill
- make-a-prototypeSkill
- make-tweakableSkill
- polish-passSkill
- wireframeSkill
- code-quality-reviewSkill
The file
Files
提取设计系统 Token
从可验证来源提取现有视觉语言,让后续设计复用同一套规则。此 Skill 用于发现和结构化,不负责凭空创造品牌。
1. 确认来源
可能的来源包括主题文件、CSS 变量、Tailwind 配置、组件库、线上页面、截图、品牌 PDF 或 Figma。优先级通常为:明确的设计规范与 token 源码,其次是稳定组件实现,最后才是截图推断。
若来源未指定,先询问。记录每个值来自哪里;无法确认时标记缺口,不要补成“合理默认值”。
2. 按类别提取
- 颜色:品牌主色与强调色、语义色、中性色阶、页面和组件表面、边框与遮罩;记录值、来源名称和用途。
- 排版:字体家族与 fallback、字号、字重、行高、字距和已命名文本样式;只保留实际加载或明确定义的字重。
- 间距:真实使用的尺度,以及内边距、行内、区块或组件间距是否分层。
- 圆角与阴影:所有有效级别及完整 CSS 值,说明对应组件或层级。
- 其他:动效时长与缓动、z-index、断点、容器宽度、图标尺寸等来源确实存在的 token。
把近似重复、不同色调的中性色阶和离群值列为发现,不要自行合并。
3. 输出项目原生格式
优先匹配现有格式与命名:tokens.css、带类型导出的 tokens.ts、tokens.json、SCSS 变量或 Tailwind 扩展。新项目默认可使用 tokens.css,按品牌、语义、中性、表面、排版、间距、圆角、阴影和动效分组。
所有名称应表达语义或稳定层级,避免把一次性页面位置写进全局 token。保留来源可追溯性,但不要把本机绝对路径或私有信息写入公开文件。
4. 交付报告
列出使用的来源、已提取类别、无法确认的缺口、近似重复或离群值,以及需要用户决定的合并项。不要把未决项静默写成最终规范。
示例
输入:“从这个 React 项目的主题和组件里提取设计 token。”
**输出:**读取主题、CSS 变量与核心组件后生成项目命名风格一致的 tokens.ts;报告发现两套近似蓝色和三个离散圆角值,保留原值并要求用户确认是否收敛。
来源与许可
本 Skill 基于 Trystan Sarrade 的 claude-design-system-prompt 进行中文化与 Claude Skill 规范适配。上游与本改编均按 MIT License 发布,详见 LICENSE.txt。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

