brand-system
Skill6 starsChanged 22 days agoMITView on GitHub
Author a production-grade brand book + design system for a web product — a long-form BRAND.md (20 sections covering the core idea, logo system, color, typography, spacing, surfaces, motion, voice, accessibility, components, and Tailwind v4 @theme tokens), a printable BRAND.html + PDF sibling, and a matched tokens.css / tokens.json with primitive/semantic/component tiering. Use when the user asks to create, draft, author, scaffold, or update a brand book, brand guidelines, design system, design tokens, style guide, visual identity, or BRAND.md; or when a new product needs its first documented identity.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- whanoftheworld/graphic-skillstheir own version, first seen Oct 25, 2026
- timscheuerai/gtm-architecture-templatetheir own version, first seen Sep 14, 2026
- aamirqutub/claude-skillstheir own version, first seen Aug 9, 2026
- appautomaton/presentationtheir own version, first seen Jul 8, 2026
- firefly-events/plugin-hivetheir own version, first seen May 18, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- appautomaton/presentation★ 57same name
- timscheuerai/gtm-architecture-template★ 2same name
- firefly-events/plugin-hive★ 1same name
- aamirqutub/claude-skills★ 0same name
- whanoftheworld/graphic-skills★ 0same name
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 repository29 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- skills/prompt-engineer/references/claude.mdCLAUDE.md
- brand-assetsSkill
- codex-reviewSkill
- deck-to-videoSkill
- elevenlabs-sttSkill
- elevenlabs-ttsSkill
- excalidraw-diagramsSkill
- gdoc-syncSkill
- generating-video-clipsSkill
- gsheetsSkill
- gslides-syncSkill
- image-generationSkill
- namecheap-domainsSkill
- narrating-pptxSkill
- office-renderSkill
- pair-intelligenceSkill
- presentation-generatorSkill
- producing-animated-episodesSkill
- prompt-engineerSkill
- proton-passSkill
- reviving-sessionsSkill
- self-presenting-decksSkill
- tavily-extractSkill
- tech-design-docSkill
- viewing-videosSkill
- writing-plainlySkill
- writing-project-instructionsSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 98
Accent constant across light/dark: Only surface and text semantic tokens swap between themes.
- Very specific: line 110
WCAG 2.2 AA is the legal floor in 2026: Not APCA — APCA was pulled from WCAG 3 in July 2023 and has not returned.
- Very specific: line 116
RTL is a foundation, not an appendix: If the product ships in Hebrew or Arabic, use CSS logical properties (margin-inline-start, not …
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Brand System
Scaffolds a brand book + design system that sits at the root of a web
product's repo — long-form BRAND.md (the working reference, 20
sections) paired with BRAND.html (the printable stakeholder
distillation, ~7 A4 pages, rendered to BRAND.pdf via Chrome
headless) and a matched tokens.css for Tailwind v4 @theme.
This skill authors the document. Its sibling brand-assets produces the pixels (SVG favicons, PWA pack, apple-touch-icon). Call both in sequence for a complete brand rollout.
The five artifacts
| Artifact | What | How |
BRAND.md | Long-form 20-section working reference | templates/BRAND.md.tmpl via scripts/new-brand-book.sh |
BRAND.html | 7-page A4 printable distillation | templates/BRAND.html.tmpl via same script |
BRAND.pdf | Rendered via Chrome headless | scripts/render-pdf.sh |
tokens.css | Tailwind v4 @theme + :root + [data-theme="dark"] | templates/tokens.css.tmpl + scripts/extract-tokens.py |
tokens.json | W3C DTCG format for Style Dictionary / Tokens Studio | scripts/extract-tokens.py --format dtcg |
Workflow (decision tree)
- Starting fresh? — run the signature-moves interview first: copy templates/signature-interview.md.tmpl into the repo and answer every question before scaffolding. See reference/signature-moves.md for why each question matters. Without this, the output is a template; with it, it's a brand.
- Scaffolding a new book? — run
scripts/new-brand-book.sh with the
interview answers as flags. Emits
BRAND.md,BRAND.html,tokens.css,signature-interview.mdinto the output dir. - Drafting a specific chapter? — load the matching reference:
- §0 the idea / §2 primitive / §3 signature moves → reference/signature-moves.md
- §5 logo + favicon pack → reference/favicon-pack.md
- §6 colour → reference/color.md + reference/tokens.md
- §7 typography → reference/typography.md
- §10 surfaces & materials → reference/surfaces.md
- §11 motion → reference/motion.md
- §13 voice & tone → reference/voice-and-tone.md
- §14 accessibility → reference/accessibility.md
- §17 Tailwind v4
@theme→ reference/tailwind-v4.md - Hebrew / Arabic / RTL → reference/rtl.md
- Favicon + PWA pack? — do not generate assets in this skill.
reference/favicon-pack.md specs the
requirements (system-aware SVG favicon, opaque apple-touch-icon,
maskable PWA, OG light+dark pair). Hand the spec + a source SVG to
the brand-assets skill's
icon-pack.sh. - Contrast audit before shipping → scripts/audit-contrast.py. WCAG 2.2 AA on body (4.5:1), 3:1 on large / UI / focus rings. Runs per theme (light + dark). Paste the matrix into §14.
- Rendering the PDF →
scripts/render-pdf.sh. Chrome headless.
Produces
BRAND.pdffromBRAND.html. - Auditing an existing book → scripts/audit-outline.py. Validates the 20-section outline, flags missing decision log entries, warns if §2 signature primitive has <8 use-sites documented.
- Catching drift between BRAND.md and production CSS → scripts/check-consistency.py. Brand books fail when code silently diverges from them. Diffs colour tokens in BRAND.md against one or more target CSS files and exits non-zero on hex drift. Wire into CI.
- Enforcing the interview before scaffolding → pass
--require-interview path/to/signature-interview.mdto scripts/new-brand-book.sh. Refuses to scaffold if the interview still has unfilled{{PLACEHOLDER}},{{TODO}}, or> …answer lines. Strongly recommended.
House rules
- Require the anti-template moves before scaffolding. The interview is not optional. One invented proper noun, three falsifiable principles, three real don'ts from past mistakes, a 150-word voice sample in the brand's voice, and one signature move that deliberately breaks a best practice. Without these, the book is a template. See reference/signature-moves.md.
- Accent constant across light/dark. Only surface and text semantic tokens swap between themes. The one saturated brand colour is identical in both modes — that's what lets a signature primitive mean the same thing regardless of system preference. Encoded in templates/tokens.css.tmpl.
- Three-tier tokens per Nathan Curtis taxonomy. Primitive
(
--color-cream: #F3EAD3) → semantic (--bg: var(--color-cream)) → component (--btn-bg: var(--bg-elevated)). Primitives go in@theme, semantics in:root, dark-mode overrides in[data-theme="dark"]. See reference/tokens.md. - WCAG 2.2 AA is the legal floor in 2026. Not APCA — APCA was pulled from WCAG 3 in July 2023 and has not returned. AA means 4.5:1 body, 3:1 large/UI/focus, 24×24 target-size-minimum. Audited with scripts/audit-contrast.py; see reference/accessibility.md.
- RTL is a foundation, not an appendix. If the product ships in
Hebrew or Arabic, use CSS logical properties (
margin-inline-start, notmargin-left) from day one; the retrofit cost is 10–50× designing-for-it upfront. See reference/rtl.md. - The brand book is itself an instance of the brand. The
BRAND.htmltemplate uses the product's own palette + type — so the document looks like the brand on page 1. If the brand book's own styling doesn't pass the rules it preaches, it's wrong. - System-first, toggle-second for dark mode. Respect
prefers-color-schemeby default; persist explicit user overrides inlocalStorage. A visible dark-mode toggle in the nav is an anti-pattern. The system-aware SVG favicon rule in reference/favicon-pack.md follows the same logic.
Non-goals
- Generating logo artwork — that's image-generation.
- Mechanical asset pipelines (vectorize PNG → SVG, icon-pack generation, rasterize at size) — that's brand-assets.
- Building the actual components — §18 of
BRAND.mddocuments the specs; implement them in your own component layer — Anthropic's frontend-design skill is a good companion for that step. - Brand storytelling / marketing copywriting — §13 governs how to write, not what.
- Translating copy — the skill scaffolds RTL-aware structure but doesn't translate.
- Versioning the brand book as a product — update in place, commit with a clear message, one source of truth at the repo root.
Canonical 20-section outline (what BRAND.md.tmpl emits)
| # | Section | Core must-have |
| 0 | The idea | One-sentence core idea + the "strip everything but X" test |
| 1 | Reading the mark | 3–5 simultaneous readings of the logo |
| 2 | Signature primitive | The semantic token that appears across every surface — 8+ use sites |
| 3 | Signature moves | 5–7 things that identify the brand without the logo |
| 4 | Brand essence | What we are / are not / tone / positioning sentence |
| 5 | Logo system | Variants, clear space, min size, don'ts, favicon + PWA pack requirements |
| 6 | Colour system | Core + extended + neutrals + semantic tokens, 62/30/8 ratio, light/dark parity |
| 7 | Typography | Face, tabular nums, scale, measure, RTL moves, wordmark treatment |
| 8 | Iconography | Style, colour rules (accent never used), size tokens, RTL flipping |
| 9 | Spacing & layout | 8-point grid, containers, radii + concentric-radii rule |
| 10 | Surfaces & materials | Default + floating material, grain overlay, hero backgrounds |
| 11 | Motion | Principle, tokens, the signature animation, reduced-motion behaviour |
| 12 | Imagery & illustration | Photography rules, illustration style, anti-patterns |
| 13 | Voice & tone | Rules, tone matrix by emotional state, 3-levels-of-same-thought, RTL specifics |
| 14 | Accessibility | WCAG 2.2 AA contrast matrix, focus rings, reduced motion, RTL equivalence |
| 15 | Reference set | What shelf the brand sits on + what it's not |
| 16 | Anti-patterns | Specific "don't do this" list — concrete bad habits banned |
| 17 | Implementation — Tailwind v4 @theme | Full block, :root + [data-theme="dark"], body rules |
| 18 | Components | Buttons, links, inputs, cards, badges, nav, alerts, dialogs, prose, landing primitives |
| 19 | Migration plan | What's being retired; ✅/⚠️/❌ per token/pattern; dates |
| 20 | What this is not + Decision log | Non-goals + dated table of decisions with rationale |
Quick invocation
# Scaffold a new brand book at the repo root
~/.claude/skills/brand-system/scripts/new-brand-book.sh \
--product "Agentiko" \
--positioning "A real worker who lives inside WhatsApp." \
--palette-bg '#F3EAD3' \
--palette-fg '#0E1320' \
--palette-accent '#B85A3A' \
--signature-primitive "voice dot" \
--primary-font "Rubik" \
--locale "he" \
--output-dir .
# Audit contrast
~/.claude/skills/brand-system/scripts/audit-contrast.py \
--bg '#F3EAD3' --fg '#0E1320' --accent '#B85A3A'
# Render the printable
~/.claude/skills/brand-system/scripts/render-pdf.sh BRAND.html BRAND.pdf
# Later: check an existing book against the outline
~/.claude/skills/brand-system/scripts/audit-outline.py BRAND.mdRelated skills
- brand-assets — mechanical SVG/PNG pipelines; run after this skill to produce the favicon pack.
- image-generation — design logo/hero imagery before this skill runs.
- Anthropic's frontend-design — implement §18 components in production code.
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.

