项目近 7 天有 3 次提交，开发活跃。
Passes @google/design.md lint with 0 errors, 0 warnings
前端开发者
Frontend developers using AI coding assistants
目标用户 · 非用户评价Target user · not a testimonial
DESIGN/设计资源·UI-KIT
yuvrajangadsingh/brandmd
执行1条命令，从任何网站提取完整设计系统，零错误通过Google官方lint0 errors, 0 warnings — this design system extractor passes Google's linter clean. What's the catch?
把网站设计系统提取成DESIGN.md，让AI编码工具照着做，不再瞎猜界面样式。Extract any website's design system into a DESIGN.md file that passes Google's official linter, so AI coding tools stop guessing your UI.
怎么装：Install: npx brandmd https://stripe.com -o DESIGN.md
02 / 现在的位置02 / Why now
首次发现FIRST SEEN
03 OCT 202603 OCT 2026
增长GROWTH
历史不足Not enough history
只有 1 个快照，还不足以计算增长Only 1 snapshots so far — not enough to compute growth
讨论DISCUSSION
暂无None
还没有采集到外部讨论No external discussion collected yet
状态STATUS
持续活跃Active
质量 AQuality A
项目近 7 天有 3 次提交，开发活跃。Passes @google/design.md lint with 0 errors, 0 warnings
03 / 它能帮你做什么03 / What it helps you do
提取任意网站设计风格成规范文件
让AI照着设计写界面不瞎猜
检查设计文件是否符合官方规范
对比两个网站的设计差异
监测线上页面是否偏离设计规范
Extract design tokens from any public website
Output a DESIGN.md that passes Google's linter
Detect blocked pages and refuse garbage output
Generate CSS custom properties or Tailwind v4 theme
Compare two extractions to find design drift
04 / 谁在用04 / WHO IT’S FOR
项目近 7 天有 3 次提交，开发活跃。
Passes @google/design.md lint with 0 errors, 0 warnings
前端开发者
Frontend developers using AI coding assistants
目标用户 · 非用户评价Target user · not a testimonial
项目创建于 2026 年 3 月，最近提交在 2026 年 9 月，持续更新。
Detects blocked pages (Akamai, WAF 403s) and warns instead of emitting garbage
设计师
Design system maintainers
目标用户 · 非用户评价Target user · not a testimonial
已有 66 个 star，说明有一定关注度。
Supports dark mode, CSS custom properties, and Tailwind v4 output
独立开发者
Teams using Claude Code or Cursor
目标用户 · 非用户评价Target user · not a testimonial
项目近 7 天有 3 次提交，开发活跃。
Passes @google/design.md lint with 0 errors, 0 warnings
技术创业者
Developers rebuilding a site from a reference
目标用户 · 非用户评价Target user · not a testimonial
不适合谁Not for同样来自文档，不是我们的判断Also from the docs, not our judgement
05 / 安装05 / Install
$ README 里给出的安装方式。The installation method given in the README.
$ README 里给出的安装方式。The installation method given in the README.
两种装法选一个即可 —— 上面那条给命令行用户，下面那条给写代码的用户。Pick one of the two — the first is for command-line users, the second for people writing code.
运行 npx brandmd https://stripe.com -o DESIGN.md 即可直接使用。Run npx brandmd https://example.com -o DESIGN.md to extract a design system without installing anything.
06 / 限制与风险06 / Limits and risk
风险不是警告，是可信度的一部分。以下结论只基于文档静态扫描，我们不会执行项目里的任何代码。Risk here is evidence, not an alarm. These findings come from static scanning of the docs; we never execute a project’s code.
限制LIMITS
需检查Needs review
静态扫描STATIC SCAN
01控制浏览器Controls a browserreadme
Adds a "Dark Theme Overrides" section to the DESIGN.md with the dark color palette. Uses `prefers-color-scheme: dark` via Playwright, so it
02执行 Shell 命令Runs shell commandsSKILL.md
allowed-tools: Bash(npx brandmd *)
03读写本地文件Reads and writes local filesSKILL.md
allowed-tools: Write
静态扫描命中：该工具会控制浏览器（Playwright）、执行 Shell 命令（Bash）、读写本地文件（Write）。这些都是它提取网站设计所需的功能，但意味着运行它会下载并执行第三方代码，请在可信环境使用前审阅。The tool uses Playwright to browse websites, which could access arbitrary URLs. It also executes shell commands and writes files locally. Review the code before running on sensitive projects.
这里只做静态扫描：读 README、SKILL.md 和依赖清单，不执行代码。没有命中不代表安全。This is static scanning only: we read the README, SKILL.md and dependency list, and never execute code. No findings does not mean safe.
依赖：Dependencies:@google/genai, chroma-js, commander, playwright
{
"agents": [
{
"agent": "claude-code",
"evidence": "readme: …T-yellow.svg)](LICENSE) Stop Claude Code, Cursor, and Gemini CLI from…"
},
{
"agent": "cursor",
"evidence": "readme: …- **`--agent`**: also writes `.cursor/rules/brand.mdc` and `SKILL.md` to…"
},
{
"agent": "gemini-cli",
"evidence": "readme: …Stop Claude Code, Cursor, and Gemini CLI from guessing your UI. brandm…"
},
{
"agent": "vscode",
"evidence": "readme: …ncluding Claude Code, Cursor, VS Code/Copilot, Codex, and Gemini CL…"
}
],
"apiKey": "from a screenshot. Requires `GEMINI_API_KEY` (free at [aistudio.google.com/apikey](",
"docker": null,
"taxonomy": {
"scores": {
"design": 14,
"ai-tools": 3,
"dev-tools": 10,
"indie-web": 1,
"skill-agent": 6,
"productivity": 1
},
"primary": "design",
"secondary": "ui-kit"
},
"localRuntime": "sections. Default extraction runs locally with no API key. ```bash npx brandmd h",
"skillMdTotal": 1,
"skillMdErrors": [],
"treeTruncated": false,
"categoryScores": {
"data": 0,
"agent": 0,
"media": 0,
"design": 19,
"browser": 0,
"devtool": 4,
"security": 0,
"marketing": 1,
"productivity": 1
},
"classification": [
{
"signal": "1 SKILL.md with name + description",
"weight": 0.95
},
{
"signal": "README describes a Claude / Agent Skill",
"weight": 0.7
},
{
"signal": "description mentions AI",
"weight": 0.2
}
],
"scannedSources": [
"readme",
"SKILL.md"
],
"skillMdFetched": 1
}