Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add zstvns/design-skills --skill brandgit clone --depth 1 https://github.com/zstvns/design-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zstvns/design-skills/brand)<a href="https://agentmods.dev/skills/zstvns/design-skills/brand"><img src="https://agentmods.dev/badge/skills/zstvns/design-skills/brand/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/zstvns/design-skills/brand"><img src="https://agentmods.dev/badge/skills/zstvns/design-skills/brand.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00141 | $0.05581 |
| Opus 5 | $0.00071 | $0.02790 |
| Sonnet 5 | $0.00028 | $0.01116 |
| Haiku 4.5 | $0.00014 | $0.00558 |
Grade C, and why
brand scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
**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 How it starts
The opening of the file, as written. The whole thing — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Brand
This is the front door for any brand work, and it has two jobs:
- 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.
- 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.
brandalways runs.brand-strategyis 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-familynames, 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
<header>/<nav>and pull the exact asset it uses — the<img src>, the inline<svg>, or the CSSbackground-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:
.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.- The
competitor-profilingskill — if there's no marketing context, run it (/competitor-profiling) to research and profile competitors properly, rather than guessing. - Autonomous discovery — fallback only: infer the direct competitors from the category, the positioning, and the brand's own comparison/alternative pages.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 178 lines · 141 tokens per session scan C cd1bf6c3feca
brand is a skill published in the GitHub repository zstvns/design-skills (10 stars, last pushed 20d ago), licensed MIT. It adds 141 tokens to every session and 5,581 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it C with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
new-template
Scaffolds a new visual (.tsx) in an existing BrandArtisan project, aligned with its brand.md guidelines. Use when the user wants to create a new image/cover/banner/OG in a project ("new visual", "add a cover for X", "create the OG for Y"). Always reads brands/ /brand.md and refuses if it is missing.
x-page
Creates the X / Twitter profile visuals (profile photo and header / banner) at X's official dimensions. Use when the user wants an "X/Twitter profile photo", a "Twitter header", an "X banner" or "the X avatar" for a project. Produces one or two .tsx templates at the sizes X imposes (profile 400x400, header 1500x500)…
brand-assets
Generates or regenerates a project's brand assets (logo, favicon and their variants) through the brand toolchain, faithfully. Use when the user wants to "regenerate the logo / the favicon", "add an icon or logo variant", "a new brand variant", or "set up a project's brand toolchain". Works in tools/ / reusing…
new-project
Initializes a new BrandArtisan visuals project by laying down its reference guidelines. Use when the user wants to create/add a new project, start a new brand or a new visual identity ("new project", "add project X", "set up brand Y"). Creates brands/ /brand.md (mandatory before any visual) and the templates/ / folder.
visual-research
This skill should be used when analyzing a competitor's brand identity, auditing visual brand elements, researching brand positioning, comparing brand aesthetics, evaluating competitor design language, or conducting systematic visual research on brand identity, logo systems, color palettes, or design patterns for…
ron-draper
Distill an already-built or near-launch product into one incisive brand kernel — the "reveal" already hiding in its name, story, or the product itself — then build the name, voice, visual identity, and a developer-ready handoff kit around it. This is LATE-STAGE branding: use it when there is a real, thought-through…