Brand guidelines
Generates a brand kit from a company one-liner — positioning adjectives, a palette with hex codes, usage ratios and do/don'ts, a display+text type pairing with fallback stacks, a voice & tone table (say / never say), a logo brief, and a one-page BRAND.md brandbook other skills consume. Use when the user says "we need a brand", "create our visual identity", "pick our colors and fonts", "define our voice", or before designing anything for a company with no identity yet.From its SKILL.md
npx -y skills add alebgl77/claude-inc --skill brand-guidelinesAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 8 stars8 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
4.3 KB, ~1.0k tokens by cl100k_base, as published. Nobody here has run it
Brand Guidelines — Brand Keeper
"Build a brand kit"
When to use
- A company has a one-liner and nothing else — "we need a brand for this".
- Visuals exist but are incoherent — "our colors and fonts differ on every page; fix the source".
- Copy has no consistent voice — "define how we sound in the product and in emails".
- Upstream of sibling skills — run before
ui-ux-pro-maxorfrontend-designfor a new venture.
Workflow
- Extract positioning from the one-liner. Ask or infer the audience, the enemy (what the brand stands against — "enterprise bloat", "hidden fees"), and the price position. Distill into exactly three positioning adjectives plus one anti-adjective: "confident, warm, precise — never cute".
- Translate adjectives into a palette. One ground, one text, one accent, plus semantics — each with a hex code, a name that carries meaning ("Glacier", not "Blue 2"), a usage ratio (60/30/10), one do and one don't.
- Pair type. One display face + one text face with real contrast between them (serif/sans, geometric/humanist), the weights worth licensing, and a CSS fallback stack for each. State the pairing logic in a single sentence.
- Write the voice & tone table. Four traits, each with a verbatim say-this sentence and a verbatim never-say-this counter-example, all grounded in the positioning adjectives.
- Draft the logo brief — not the logo. Concept direction, construction (wordmark vs mark, geometric basis), clearspace rule, minimum size, and the misuse list: no gradients, no stretching, no drop shadows, no recoloring.
- Assemble BRAND.md, one page. Downstream skills must be able to consume it without asking a single follow-up question.
- Stress-test the kit. Apply it mentally to a button, an error message, and an invoice footer. If any feels off-brand, revise the kit — never the artifact.
Output format
# BRAND.md — <Company>
**One-liner:** <as given>
**Positioning:** <adj>, <adj>, <adj> — never <anti-adjective>.
## Palette — 60/30/10
| Name | Hex | Role | Ratio | Do | Don't |
|---|---|---|---|---|---|
| Paper | #FAF9F7 | ground | 60% | page and card backgrounds | text |
| Ink | #101828 | text | 30% | body copy, headings | large filled areas |
| Glacier | #2563EB | accent | 10% | CTAs, links, focus rings | full-bleed backgrounds |
## Type
- Display: <face> (<weights>) — fallback: <stack>
- Text: <face> (<weights>) — fallback: <stack>
- Pairing logic: <one sentence>
## Voice & tone
| Trait | Say | Never say |
|---|---|---|
| Precise | "Backups run every 10 minutes." | "Blazingly fast backups!" |
## Logo brief
<direction> / <construction> / clearspace: <rule> / min size: <px> / misuse: <list>
Quality bar
- Exactly three positioning adjectives plus one anti-adjective, all load-bearing
- Every palette entry has hex, role, ratio, one do and one don't — no orphan swatches
- Type pairing ships real fallback stacks and names the weights to license
- Voice table uses verbatim example sentences, never abstract descriptions
- BRAND.md fits one page and pre-answers the questions downstream skills would ask
- Text color on ground color passes WCAG AA
Example
Invocation: "Brand kit for 'Ledgerly — bookkeeping that closes your month in a day'."
Produces: BRAND.md: positioning "precise, calm, unstoppable — never playful"; a palette of
warm Paper ground (60%), Ink #101828 text (30%), Ledger Green #0E9F6E accent (10%) plus amber/red
semantics, each with do/don'ts; Fraunces display over Inter text ("editorial authority over
utilitarian clarity"); a four-row voice table (Say: "Your month is closed." Never: "Boom! Books
done!"); and a wordmark brief with tabular numerals as the identity hook, cap-height clearspace,
and a no-gradients-ever misuse list.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most marketing audience skills give in ~1.0k tokens
Counted across 690 of the 894 authors here whose files we hold, read 2026-08-07
- Apply Poppins font to headingsin 41 of 690, across 6 files
- Apply Lora font to body textin 41 of 690, across 6 files
- Use Arial fallback for headingsin 39 of 690, across 4 files
- Use Georgia fallback for body textin 39 of 690, across 4 files
- Maintain text hierarchy and formattingin 39 of 690, across 4 files
- Use accent colors for non-text shapesin 38 of 690, across 3 files
- Use RGB values for precise color matchingin 38 of 690, across 3 files
- Use brand colors for primary text and backgroundsin 36 of 690, across 1 file
- Read product marketing context file before asking questions, starting, or auditingin 35 of 690, across 23 files
- Use active voice instead of passive voicein 26 of 690, across 10 files
- Implement or generate appropriate JSON-LD structured datain 24 of 690, across 17 files
- Prioritize clarity over clevernessin 22 of 690, across 8 files
Said here and by no other author read
- extract three positioning adjectives and one anti-adjective
- create a palette with hex codes, ratios, and rules
- pair a display face and text face with fallback stacks
- write a voice and tone table
- draft a logo brief instead of a logo
- assemble a one-page BRAND.md
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.