Derive a full UI palette from one or two brand colours — state variants, brand-tinted greys, semantic colours. Use when building or expanding a colour system.
日本語の概要は準備中です。原文の説明を表示しています。
Shape language across type, radius and icons carries brand personality. Use when starting a design system, picking icons or radius tokens, or reviewing visual consistency.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Visual shape communicates personality. A rounded corner says something different to the user than a sharp one — and that message arrives before they read a single word. The shapes in typography, border-radius, and iconography should tell a consistent story.
| Shape | Tone | Associated with |
|---|---|---|
| Rounded, pill-shaped | Friendly, approachable, playful, modern | Consumer apps, health, kids, lifestyle, social |
| Softly rounded (8–12px) | Professional, warm, accessible | SaaS, productivity, general B2B |
| Lightly rounded (2–4px) | Precise, structured, efficient | Enterprise tools, finance, data platforms |
| Sharp / no radius | Technical, serious, authoritative | Developer tools, security, industrial |
Every component — cards, inputs, modals, badges — should follow the same radius logic.
The perceived "roundedness" of an element changes with its scale. A radius that looks soft on a button may look sharp on a large container.
Before choosing a radius, look at the brand's existing materials:
Read the real shape language, don't infer it (dembrandt engine, optional). For a brand that already ships a product, extract its actual radius, type, and tone from the live site rather than guessing from the logo:
get_brand_identityandget_design_tokensreturn computed values off the DOM. Seeextract-design.
Typeface shapes carry the same tonal signals:
| Type style | Tone |
|---|---|
| Geometric sans (circular letterforms) | Modern, clean, slightly playful |
| Humanist sans (varied stroke widths) | Warm, readable, professional |
| Grotesque sans (neutral, utilitarian) | Serious, efficient, no-frills |
| Serif | Authoritative, established, editorial |
| Rounded sans | Friendly, approachable, informal |
| Monospace | Technical, developer-facing, precise |
The typeface and the border-radius should not contradict each other. A rounded, friendly typeface paired with sharp 0px corners creates visual dissonance.
Brand books often specify a display or heading font but leave body text underdefined — a single weight, no reading size, no fallback. This is common with luxury, fashion, or legacy brands where the brand identity was built for print, not screen.
When the brand book is insufficient for UI purposes, extend it deliberately:
When extension is justified:
How to extend:
Brand heading font (display, h1–h3): maintains identity
↓
Secondary body font (body, labels, UI copy): legibility and completeness
Pairing principles:
Do this sparingly. Two typefaces is a deliberate extension. Three typefaces is almost always too many. Document the decision and the rationale so future designers do not add a third.
Icon style must match the brand's shape language. Mixing icon styles — some thin, some bold, some filled, some outlined — breaks visual cohesion even when individual icons are correct.
| Icon style | Tone | Use when |
|---|---|---|
| Thin / outline (1–1.5px stroke) | Minimal, elegant, refined | Luxury, editorial, premium SaaS |
| Regular outline (2px stroke) | Balanced, professional | General SaaS, productivity tools |
| Bold / thick (2.5–3px stroke) | Strong, clear, accessible | Consumer apps, mobile-first, accessibility focus |
| Filled | Solid, confident, clear at small sizes | Dashboard indicators, status icons, mobile nav |
| Rounded corners on icon paths | Friendly, approachable | Consumer, lifestyle, health |
| Sharp corners on icon paths | Technical, precise | Developer tools, finance, data |
Rule: Use one icon library and one weight throughout. If mixing is unavoidable (e.g. a specialised icon not available in the chosen library), match stroke width and corner style manually.
Gear = settings, person = profile, magnifier = search, house = home, trash = delete, bell = notifications. Inventing an alternative here costs decoding for zero benefit. Search "icon [concept]" to check the convention before committing. Save design freedom for concepts with no established icon.
Icons let the eye skip the word — but only unambiguous ones, and only where scanning pays off (nav, status, row types). A vague or decorative icon adds work instead of saving it; an icon on every label is noise. Right icon, relevant place. See [[ui-density]] on reading as time.
Unicode glyphs (arrows, dots, chevrons) serve scaffolding that degrades gracefully. Anything that carries meaning is a shipped icon with a fixed rendering. A font-dependent glyph in a status column looks different on every platform.
A pencil, a trash can, a plus: these promise an action. Never spend one on a passive type marker. A row whose type icon is a pencil reads as editable whether or not it is.
Pin the icon set version. Sets redraw glyphs between releases, so an unpinned dependency changes the product's vocabulary on install. Across a product family the same concept uses the same glyph in every app.
At 16px a letter is about 3px wide and carries no identity. When a family of small marks must be told apart (file types, app tiles, favicons), separate them by hue first and by shape second.
All shape-bearing elements should follow the same visual logic:
| Element | Applies shape language via |
|---|---|
| Buttons | --radius-button |
| Cards | --radius-card (same or slightly larger than button) |
| Inputs | --radius-input (typically same as button) |
| Badges / tags | Can be more rounded than buttons — pill shape is common |
| Modals / drawers | --radius-modal (often larger, 12–16px) |
| Avatars | Always fully round (--radius-full) |
| Icons | Stroke weight and corner style match brand |
| Illustrations | Shape style consistent with icon style |
Find the guide's own scope statement before you enforce a single value from it. A manual written for print optimises for ink on paper, where a pale grey is legible and a licensed typeface is embedded once into a PDF. Neither survives a screen: the same grey can land near 2:1 against white, and the same typeface often cannot be served through a webfont pipeline at all.
Where the guide's medium and the work's medium differ, the guide is evidence, not law. Enforcing it literally produces an inaccessible product that is nonetheless on brand, and it discredits the guide with the engineers who then have to ignore it, which costs their compliance on everything else too.
Read an existing product's departures from the guide as findings before you read them as errors. A team that deviated on a text colour has usually already met the contrast problem you are about to re-create. The reverse error costs as much: treating every departure as drift throws away the only audit of the guide anyone has actually run.
The test that separates the two: can the departure be reproduced as a constraint on a screen? A colour changed because the original lands at 2:1, a typeface swapped because the licence does not cover webfonts, a radius altered because the original renders as a smudge at 16px: each is a finding, and the guide is what needs updating. A departure no constraint explains, where the original value would have worked, is drift, and the product is what needs correcting. Run the test per value rather than per product; most estates contain both.
Take the gaps back to the brand owner as findings. A guide that has never met a screen is not wrong, it is incomplete, and the fastest way to complete it is a list of the places it ran out. Contrast floors are in [[wcag-accessibility]].
Most brand work does not land on a clean codebase. It lands on an estate that includes a portal built in 2015 with whatever framework was current and an internal system old enough to vote — and those are the tools employees stare at all day, so the brand experience is at stake there too.
Two things make this tractable. First, the shape language is the last thing to arrive, not the first. A legacy application can carry the brand convincingly with nothing but the right logo, the brand hue, and the typeface; matching radius and icon style is rung 3 work and usually never worth it there. Second, decide per application how far up the ladder it goes — identity, chrome, tokens, components — rather than treating anything short of a rewrite as failure. The ladder and the per-application decision live in [[layout-paradigms-and-consistency]].
Where an old application does have variables — a Bootstrap or Sass build usually does — mapping brand colour and typeface onto them buys most of the perceived coherence for a fraction of the work. What you are buying is recognition, not fidelity: the user should know whose software this is within a second of it loading.
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Derive a full UI palette from one or two brand colours — state variants, brand-tinted greys, semantic colours. Use when building or expanding a colour system.
日本語の概要は準備中です。原文の説明を表示しています。
Top bar, app launcher, tenant and environment cue, status bar. Use when designing the persistent shell or navigation between apps behind one login.
日本語の概要は準備中です。原文の説明を表示しています。
Product visuals must show the real product with real data. Use when building landing pages, hero sections, screenshots or demo panels.
日本語の概要は準備中です。原文の説明を表示しています。
Rest, hover, active, focus, disabled and loading states, derived from the base colour. Use when designing buttons, links, inputs or any clickable component.
日本語の概要は準備中です。原文の説明を表示しています。
Rebuild a live web page 1:1 from measurement into Figma, Penpot or code, then verify it. Use when reproducing a page or checking a rebuild against the original.
日本語の概要は準備中です。原文の説明を表示しています。
Light, dark or both, and when a theme selector is worth it. Use when defining a base palette or deciding on dark mode support.
日本語の概要は準備中です。原文の説明を表示しています。