Automated summary
Extracts a website's design system into a DESIGN.md file in Google Stitch format so AI agents can generate on-brand UI.
Strengths
- Assigns semantic roles to colors (background, text, accent, border) and breaks typography down per role (headings, body, buttons) with sizes, weights and line heights
- Offers several output formats: DESIGN.md, CSS custom properties, a Tailwind v4 @theme block, raw JSON, or a shareable HTML brand guide
- Merges tokens from multiple URLs and adds optional dark-mode (--dark) and beyond-CSS visual identity (--vision) sections
Limitations
- Scrapes a single page, so tokens reflect the supplied page rather than the whole site
- Color role assignment is heuristic (usage frequency and luminance) and should be reviewed by hand
- Requires a public URL: auth-gated or persistently bot-protected pages fail, and --vision needs a GEMINI_API_KEY
Use it when kicking off a frontend project that must match the visual identity of an existing website.
Skip it when design tokens already live in Figma or in code, or when the target site is private or bot-protected.
Security analysis
CautionThe skill is not destructive or exfiltrating by design, but it executes a remote package through npx and may use an API key, so it carries meaningful supply-chain and data-sharing risk. The declared tool scope is narrow and the documented behavior is legitimate, making this a caution rather than risky.
- •Runs a third-party npm package via `npx brandmd`, which executes remote code with network and filesystem access; package provenance/version is not pinned.
- •The optional `--vision` mode uses `GEMINI_API_KEY`; users should be aware that enabling it sends page/design context to a third-party Gemini API endpoint.
- •The skill fetches arbitrary user-supplied URLs and waits on Cloudflare JS challenges, so it should only be used on sites the user is authorized to inspect.
Examples
Extract the design system from https://stripe.com and write it to DESIGN.md, then build a pricing page that matches that brand.Run npx brandmd https://linear.app --dark --css so I get dark-mode tokens as CSS custom properties.Merge the tokens from our homepage, pricing page and blog into one Tailwind v4 @theme block.name: brandmd description: Extract a website's design system into a DESIGN.md file. Use when starting a new frontend project, rebuilding a site, or when the user wants AI-generated UI to match an existing brand. allowed-tools:
- "Bash(npx brandmd *)"
- "Read"
- "Write"
brandmd
You extract a website's design system into a DESIGN.md file that AI coding agents use for on-brand UI generation.
When to use
- User says "match this website's design" or "use the same style as [url]"
- Starting a new frontend project and the user has an existing site
- User wants design consistency across AI-generated components
- User asks to "extract design tokens" or "get the design system" from a URL
How to run
npx brandmd <url> -o DESIGN.md
Pass the most design-rich page (usually the homepage) for best results. The tool extracts from the visible viewport and lazy-loaded content on that single page.
What it extracts
The output follows Google Stitch's DESIGN.md format. Sections 1-5 are always present; 6-7 are conditional on flags.
- Visual Theme & Atmosphere - mood, shape language, depth
- Color Palette & Roles - colors with semantic roles (background, text, accent, border)
- Typography Rules - Primary font (role-aware: display > heading > body, skipping mono/fallback/icon fonts), per-role font breakdown (Headings, Display, Body, Buttons), all detected fonts with frequencies, size scale, weights, line heights, letter spacing
- Component Stylings - buttons, cards, inputs with radii, colors, shadows
- Layout Principles - spacing scale, base grid unit, border radii
- Dark Theme Overrides (with
--dark) - palette + atmosphere for dark mode - Visual Identity Beyond CSS (with
--vision) - illustration style, photography mood, copywriting voice, microcopy patterns. RequiresGEMINI_API_KEY(free tier: aistudio.google.com/apikey).
After extraction
- Read the generated DESIGN.md to confirm it looks right
- Summarize the key tokens for the user (primary colors, fonts, spacing scale)
- Use these tokens in all subsequent UI code generation
- Reference specific hex codes, font names, and spacing values from DESIGN.md when writing CSS/Tailwind/styled-components
Options
npx brandmd <url>- DESIGN.md to stdoutnpx brandmd <url> -o DESIGN.md- DESIGN.md to filenpx brandmd <url> --dark- also extract dark-mode palette (Section 6)npx brandmd <url> --vision- also extract illustration / photography / voice / microcopy via Gemini (Section 7). NeedsGEMINI_API_KEYenv var.npx brandmd <url> --css- CSS custom properties output instead of DESIGN.mdnpx brandmd <url> --tailwind- Tailwind v4@themeblock instead of DESIGN.mdnpx brandmd <url> --html- HTML brand guide (visual, shareable)npx brandmd <url> --json- raw tokens as JSON (programmatic use)- Multiple URLs merge tokens:
npx brandmd <url1> <url2> <url3>
Limitations
- Extracts from a single page, not the entire site (multi-URL merges token frequencies but the vision screenshot is the first page only)
- Can't read Figma tokens or design tool files
- Color role assignment is heuristic (based on usage frequency and luminance)
- Requires a publicly accessible URL (no auth-gated pages)
- Cloudflare-protected sites: brandmd waits up to 20s for the JS challenge to auto-resolve (
--cf-wait-msto tune). If the challenge persists, you get a clear error rather than garbage tokens. Hard-blocked sites (openai.com, perplexity.ai, discord.com, etc.) are correctly flagged as bot-protected and not bypassed.
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/extract-design-system-ms2s7bi2/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/extract-design-system-ms2s7bi2' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/extract-design-system-ms2s7bi2/download' --output '.skill-downloads/extract-design-system-ms2s7bi2/SKILL.md.part' && mv '.skill-downloads/extract-design-system-ms2s7bi2/SKILL.md.part' '.skill-downloads/extract-design-system-ms2s7bi2/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/extract-design-system-ms2s7bi2/SKILL.md' && test -s '.skill-downloads/extract-design-system-ms2s7bi2/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/brandmd' && cp -R '.skill-downloads/extract-design-system-ms2s7bi2' '.claude/skills/brandmd'Cursor
test -f '.skill-downloads/extract-design-system-ms2s7bi2/SKILL.md' && test -s '.skill-downloads/extract-design-system-ms2s7bi2/SKILL.md' && mkdir -p '.cursor/skills' && test ! -e '.cursor/skills/brandmd' && cp -R '.skill-downloads/extract-design-system-ms2s7bi2' '.cursor/skills/brandmd'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/extract-design-system-ms2s7bi2'openskills read 'extract-design-system-ms2s7bi2'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/extract-design-system-ms2s7bi2/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
React Hook Form Best Practices
Development
Master React Hook Form: avoid anti-patterns, manage subscriptions, and handle reset, dirty state, and controlled inputs correctly.
Emotion to Mantine Migration
Development
Migrate Emotion styled-components to Mantine components with style props and CSS modules. Use when converting .styled.tsx files or removing @emotion imports.
Scope Creep Detector
Development
Analyzes git diffs against a stated intent to detect scope creep, unrelated files, broad PRs, and oversized changes, then suggests keep, split, or justify actions.