Brand

Brand

This is the front door for any brand work, and it has two jobs:

1. Understand what exists, diagnose how much should change, and route. The depth of the work — a ground-up build, a refresh, or just better execution — is your diagnosis to make from the evidence. The user does not need to know the framework behind it.
2. Own the record. Get-or-make .agents/brand.md — the single source of truth for the brand's visual identity that every other design skill reads first, so the user never re-explains their colors, type, muses, or latitude.

> brand always runs. brand-strategy is the deep generative path it routes to — and when the brand is already sound, it routes away from strategy entirely. That's why they're separate skills: one runs without the other.

Step 1: Gather the Assets — Do the Legwork Yourself

Start here, always — and take this off the user's plate. Don't hand them a homework assignment ("send me your logo files, your colors, your social links, your decks…"). Get the one thing you actually need to begin — the website URL (or, failing that, the company name) — and go find the rest yourself.

From just the domain, gather autonomously:

• Crawl the key pages, not only the homepage — home, product/features, pricing, about, blog. Each surface uses the brand differently.
• Extract real values from the rendered pages — computed colors, font-family names, type scale, spacing, button/CTA treatment, radii, shadows.
• **Capture the logo — and make sure it's the current one. The source of truth for the live mark is the navigation bar / header of the rendered site**, nothing else. Inspect the actual `/ and pull the exact asset it uses — the , the inline , or the CSS background-image`. A plain text/markdown fetch strips this out, so read the raw HTML/DOM or screenshot the nav and confirm what you extracted matches what's on screen.
• Favicons, apple-touch-icons, manifest icons, OG images, social avatars, and press-kit downloads are frequently outdated — teams refresh the nav logo and forget the rest. Treat them as secondary: useful for collecting the other implementations (small, square, reversed), but reconcile every one against the nav logo. If they disagree, the live nav wins as "current," and the mismatch is itself a finding (stale assets to fix).
• You still want the full set — a mark's simplicity is judged by whether it's the same mark across all of them — but the nav logo is the reference, not the favicon. Don't let an old favicon or a stale press kit stand in for the current identity.
• Follow the footer. Almost every site links its own social profiles, app-store listings, and sometimes a press or brand page down there. Harvest those links and follow them.
• Check the common brand/press paths — /brand, /brand-assets, /press, /media-kit, /about/brand. Companies routinely publish logos, palettes, and usage rules there.
• **Screenshot and actually look at the rendered design — don't score from tokens alone. A brand's identity often lives in things a CSS/token dump or a product-UI Figma never expose: a recurring graphic motif (a radial, shape, or texture repeating across the hero, sections, and social), the accent colors used only in the hero or social art, and the logo's symbol/glyph**. Examine the hero, feature graphics, and social posts directly. Never conclude "no pattern" or "wordmark-only, no mark" from tokens — confirm with your eyes. These feed Pattern & Texture, Color, and Logo, and missing them is how a strong brand gets mis-scored as weak.
• Read the codebase if you're in it. A site repo's CSS / Tailwind config / design-token files are the most precise source there is — prefer them over anything inferred from a screenshot.
• Pull Figma if a link or connected file is available.
• Source the competitor set from the marketing context — don't freelance it. Identifying competitors by gut is hit-or-miss; pull the authoritative set instead, in this order:

1. .agents/product-marketing.md (from the marketing-skills library) — if it exists, use its Competitive Landscape: direct / secondary / indirect competitors, with why each falls short. The audit scores distinctiveness against the direct set; secondary/indirect inform the landscape but aren't the visual peer group.
2. The competitor-profiling skill — if there's no marketing context, run it (/competitor-profiling) to research and profile competitors properly, rather than guessing.
3. Autonomous discovery — fallback only: infer the direct competitors from the category, the positioning, and the brand's own comparison/alternative pages.

Then capture each direct competitor's identity across the nine audit categories — the comparison set the Step 2 audit scores against (see [references/brand-audit.md](references/brand-audit.md)) — and confirm the list with the user (below).

Confirm the competitor set before any scoring. Your idea of "direct competitors" is often wrong, and the entire distinctiveness verdict rides on it. Present the list you assembled — "Here's who I'm treating as your direct competitors: … Anyone to add, remove, or swap?" — and get the user's sign-off before running the Step 2 scorecard. Don't score against an unverified set.

Then show the user what you found, and ask only for the gaps you genuinely can't reach yourself:

• Private or internal material — a Figma you don't have access to, sales decks, print collateral, photography libraries not on the site
• The why behind decisions you can only see the what of
• Anything the Step 2 diagnosis hinges on that the assets don't answer

Build the asset inventory from all of it — what you gathered, what the user added, and what's still missing. Gaps are findings: they show where the brand is undefined, and how much exists is the first input to the diagnosis.

As you ingest (see [references/asset-intake.md](references/asset-intake.md)):

• Extract concrete values — hex codes, named typefaces, spacing, logo lockups, photo treatment — not impressions
• Watch how the brand behaves across mediums. The site may be buttoned-up while social runs loose — that contrast is data; it reveals what's truly fixed vs. what already flexes.
• Stated ≠ practiced. A style guide claims one thing; the live assets often show another. Capture both and flag the drift.

If little or nothing exists, that itself is the key finding — carry it into the diagnosis.

Step 2: Diagnose How Much Should Change

This is your diagnosis, not a question for the user — and you make it with a critical eye, not a generous one. A client has no idea what "Revolution / Evolution / Optimization" mean — never ask them to pick or lead with the jargon. Read the situation from the assets in Step 1 plus a few plain questions, then conclude. These are your internal scoping labels:

The verdict is a phased recommendation anchored to the client's appetite and timeline — not a single label derived from the scores. Ask, plainly, how much they want to change now vs. later, and whether they want incremental polish or a transformational break; that appetite isn't visible in the assets and it's the primary driver. A brand can score low and still be "Optimization now, deeper work later." Diagnose honestly (competitor similarity + scores) to shape the roadmap, then recommend a path — often sequenced:

• Optimization — the immediate design-fundamentals work on the existing assets (typography, hierarchy, contrast, layout, palette rationalization, consistency). The right starting point for any client moving incrementally — even a generic-foundation one — with deeper work teed up for later; also the terminal answer when assets are strong and only execution lags (the strong-assets-badly-built case). (Or the low-scoring, genuinely generic brand whose client wanted "optimization now, bigger later" → Optimization now, refine what exists, sequence creative-direction later.)
• Evolution — keep the equity, do generative build-out (creative direction, motif, deploy missing pieces) — the equity-in-a-drawer case.
• Revolution — rebuild the foundation: colliding/undifferentiated with little to keep (the lookalike case), or the client wants a transformational break and has the appetite for it.

Never override the client's incremental appetite with a diagnosis-driven verdict, and never argue against the depth they asked for. Be honest about what each phase buys: Optimization-now makes a generic brand tidier, not distinct — distinctiveness comes from the creative-direction phase. Don't sell polish as distinctiveness; don't refuse polish because distinctiveness needs more. The common shape is micro now (Optimization) → macro later (creative direction).

Optimization vs Evolution: in an Optimization you change only how existing assets are implemented (design fundamentals on what's there); in an Evolution you also create, define, or deploy pieces that are missing, unanchored, or sitting in a drawer.

At moderate similarity it's a genuine fork the client chooses by appetite: Optimization (content with the overlap → keep the assets, execute higher, shore the gaps) vs Revolution (want to break away → a wholesale shift, not incremental). Weigh their stated ambition; never assume it.

Hold a high bar — coherent is not the same as good (see design-principles → Coherent Is Not the Same as Good):

• Consistency, intentionality, and "not dated" are table stakes, not proof of a sound brand. A tidy brand is usually just fine — and "fine" is an Evolution, not an Optimization.
• Test distinctiveness, not just coherence. Look outward — scan the category, the direct competitors, and the prevailing design trend. A brand is only distinct relative to them. With the logo covered, would anyone know it was them?
• Name the default. A polished execution of the current category or AI-startup aesthetic — near-black canvas, high-contrast serif display, a single jewel-tone accent, parchment warmth, a tasteful mono — is generic, not sound. Reading as "deliberate" doesn't save it; everyone is deliberate in the same way. Flag the trend as a finding, not a strength.
• Assess the logo specifically with logo-design against the SAD bar — Simple (it must hold in plain black & white; if it leans on color or a gradient to read, that's a failure), Appropriate (to the emotional target), Distinct (against the competitor logos you gathered, not in isolation). A weak or generic mark is a real part of the diagnosis — don't wave it through because it's "their logo." And the mark is always on the table: if the creative direction is shifting significantly (Revolution or serious Evolution), the logo is in scope for change too — flag that rather than treating it as untouchable equity.
• Lead with the critique — but separate diagnosis from recommendation. Your value is catching "this is fine" dressed up as "this is good": a generic or competitor-colliding foundation is not a sound brand — don't declare it fundamentally fine. That's the honest diagnosis (the scores). It's separate from the recommendation, which is phased and appetite-anchored — you can diagnose a weak foundation and still recommend Optimization-now (fundamentals first), with the creative-direction work sequenced later. Don't confuse "the foundation is weak" (a scoring truth) with "so we must Revolutionize now" (a recommendation the client's appetite may not support).
• Don't be fooled by rank or drift. "We're #2 in the category" is not distinctiveness if the category is a lookalike cluster — best-in-cluster is still generic (this applies to the logo too: least-plain wordmark in a plain set ≠ distinct). And finding execution problems (stale press kit, half-finished font migration, fragmenting color) does not mean the foundation is sound — drift routinely sits on top of an undifferentiated foundation, so don't let it inflate the foundational scores. (It's fine for those execution fixes to become the Optimization-now phase of the roadmap — just don't mistake them for evidence the foundation is strong.)
• Implementation flaws are not weak assets. Look through a rough site. A distinctive pattern shown at blown-out contrast is a strong asset, badly executed — an Optimization signal — not a weak pattern; a great typeface buried in banded layouts is still a great typeface. Assess each asset's intrinsic quality and distinctiveness separately from how well the current site implements it. Slop-ish execution (banding, over-contrast, screenshot bleed) sitting on strong, distinct, deployed assets means fix the build (Optimization) — not a weak foundation.

Run the brand audit — don't eyeball it. Work through the nine categories ([references/brand-audit.md](references/brand-audit.md)) against the confirmed competitor set, with named comparisons (whose palette/mark/motif it resembles or collides with). Score on two axes: the foundational categories (Creative Direction, Color, Logo, Pattern) on distinctiveness against the field; the craft categories (Typography & Layout, Graphics, Icons, Application, Style Guide) on execution quality — where well-made, consistent work reaches 4–5 even with a common typeface or stock icons. Creative Direction does NOT cap the craft scores. Then read competitor similarity — the heaviest-weighted pivot — and let it plus the client's appetite drive the path (split fork at moderate similarity). Present the recommended path with rationale and the alternative's tradeoff. Write the audit into .agents/brand.md.

Run the AI-slop check ([../design-principles/references/ai-slop.md](../design-principles/references/ai-slop.md)) on the brand and the whole set. Tropes — a left accent rail, a purple-to-blue gradient, a cramped center container, no variety between section layouts, rounded-lg on everything — are findings that cap a dimension's score, however "clean" it looks. Name each one you find.

Diagnose from plain signals (ask only what the assets don't answer):

• Is the brand wrong/off-strategy, just dated, coherent but generic, or genuinely strong and distinct but poorly applied?
• Is it actually distinct in its category, or a tidy version of the current trend?
• What's real equity worth protecting, vs. what only looks finished?
• How much exists, and how good is it, honestly?
• What's the appetite and timeline? Do they want to change a lot now, or move incrementally — fundamentals now, bigger moves later? This drives the recommended path more than the scores do. Ask it; don't infer it — it isn't visible in the assets.

This also sets your mode: a genuinely strong brand → ingest and protect the equity; an okay-or-absent one → raise or generate it through discovery.

Reflect your conclusion back in plain terms, honestly — e.g., "Your brand is coherent and clearly intentional, but right now it looks a lot like the rest of the AI-tools category — serif-on-black with a single green accent is everywhere this year. That's not a broken brand, but it's not a distinct one yet. So I'd treat this as an evolution: keep the name and the bones, push the look somewhere only you could own." Don't flatter coherent-but-generic work into "sound." Use the REO word only if it helps; the honest read and the routing are what matter.

Present the full assessment — don't skip to the verdict

The per-category breakdown is the deliverable, not a byproduct. Always present it in the response — never collapse the audit into a one-line verdict or only write it to a file. Show, in this order:

1. The scorecard table — all nine categories with their scores, a blank "Your score" column for the user, plus the **Competitor si

…

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: zstvns
• Source: zstvns/design-skills
• License: MIT

Install and usage instructions live in the source repository linked above.