agentleFS
SkillMedevSkillMedev/brand-visual-identity/skills/typography-system/SKILL.md

typography-system

Skill1 starsChanged 39 days agoMITView on GitHub

Chooses and pairs typefaces and builds a modular type scale - ratio, named roles, line heights, measure limits, and responsive values - as a complete design-system type layer. Use when someone asks "what fonts should we use", "build our type scale", "why does our typography feel inconsistent", or is establishing the typographic voice for a brand or product. Do NOT use for animated or motion typography - use kinetic-typography instead; for page-level emphasis, spacing, and layout decisions, use visual-hierarchy; for assembling the full brand book, use brand-guidelines.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Follow the order: roles constrain the font choice, the ratio constrains the scale, and line height and measure are set per style - …

    line 14
  2. Distinctive:

    Use a modular scale from a 16px (1rem) base.

    line 33
  3. Distinctive:

    Do not choose fonts before roles - personality-first selection is how three-typeface systems happen.

    line 85
  4. Distinctive:

    Do not add a third typeface.

    line 86
  5. Distinctive:

    Do not name styles by pixel size.

    line 87
  6. Distinctive:

    Do not skip measure. Line height without a max-width still yields unreadable 120-character lines on wide screens.

    line 88
  7. Distinctive:

    Do not ship a scale without mobile values for Display and Heading steps - engineers will improvise them.

    line 89
  8. Distinctive:

    Every step has a role name, size, line height, weight, and mobile value.

    line 94

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Typography System

Typography carries more of a brand's personality than any other design element because it is always present - and a good type system is invisible when it works, painful when it doesn't. The costly mistake is picking fonts before defining roles: teams end up with three typefaces, forty ad-hoc sizes, and no way to tell a heading from a label. Roles and a scale come first; fonts fill them.

Operating procedure

Follow the order: roles constrain the font choice, the ratio constrains the scale, and line height and measure are set per style - reversing any of these produces a system that must be rebuilt.

Step 1: Gather inputs

  • Brand adjectives (3-5) and the product type: UI-heavy tool, editorial/content site, or display-heavy marketing. The product type picks the scale ratio later.
  • Target languages and scripts - glyph coverage eliminates typefaces early.
  • Platforms: web, native app, print. Each changes licensing and delivery constraints.
  • Font budget and any existing fonts under license. Label assumptions as assumptions.

Step 2: Define roles before choosing fonts

Roles: Display (hero headlines, campaign type), Heading (page titles, section headers), Body (long-form reading), UI (labels, captions, buttons), Mono (code, data). A brand needs two typefaces at most - one covering Display/Heading, one covering Body/UI. A third is a tax on every future designer. Strong pairings: one serif plus one sans-serif, or a distinctive display face with a neutral workhorse.

Step 3: Evaluate candidates on functional criteria

Beyond personality, a candidate must pass: weight range of at least Regular, Medium, Bold plus italics; legibility at small sizes (check 12-13px on a real screen); glyph and language coverage for target markets; web and app licensing; acceptable file size for web delivery. Prefer variable fonts for web - fewer requests and fine-grained weight control.

Step 4: Pick the scale ratio and build the scale

Use a modular scale from a 16px (1rem) base. Pick the ratio by product type:

  • 1.2 (Minor Third) - dense, data-heavy UIs where sizes must stay close together.
  • 1.25 (Major Third) - the default for UI-heavy products.
  • 1.333 (Perfect Fourth) - editorial products that need clear hierarchy in long reading.
  • 1.5 (Perfect Fifth) - display-heavy marketing layouts.

Name steps by role, never pixel size: body-sm, body-base, heading-md, display-lg. Pixel values change; role names stay stable.

Step 5: Set line height and measure per style

These are readability standards, not aesthetics:

  • Body: line height 1.4-1.6 (default 1.5); measure 45-75 characters, 60-75 ideal. Enforce measure with a max-width - it is the most neglected spec, and overlong lines destroy comprehension.
  • Headings: line height 1.1-1.3.
  • UI labels: line height 1.2-1.4; never set type below 12px.

Step 6: Define responsive values

Display sizes drop hard on mobile: a 72px desktop hero commonly becomes 40px. Document a mobile value for every Display and Heading step. Body and UI usually hold one size across breakpoints with minor tweaks. Hand breakpoint mechanics to responsive-spec.

Step 7: Document do/don't pairs and anti-patterns

For each role, write one sentence on what it is for and one on what it is not for, then list at least three anti-patterns: all-caps body text, centered body paragraphs, justified text without hyphenation, more than two weights in a single component.

Worked example

TYPE SCALE - ratio 1.25 (Major Third), base 16px / 1rem
Product: UI-heavy SaaS dashboard. Fonts: Inter (all roles), JetBrains Mono (code)

Token        Size   Line-height  Weight  Mobile  Notes
display-lg   61px   1.1          700     40px    Hero only; max 2 lines
display-sm   49px   1.1          700     34px    Section heroes
heading-lg   39px   1.2          600     31px    Page titles
heading-md   31px   1.2          600     25px    Section headers
heading-sm   25px   1.3          600     20px    Card titles
body-lg      20px   1.5          400     20px    Intro paragraphs; 60-75ch max-width
body-base    16px   1.5          400     16px    Default body; 60-75ch max-width
body-sm      13px   1.4          400     13px    Captions, metadata
ui-label     13px   1.2          500     13px    Buttons, form labels; never <12px

Don't: all-caps body text; centered paragraphs; justified text without
hyphenation; mixing >2 weights in one component.

Deliverable

Produce a type specification containing: the chosen typefaces with the functional-criteria rationale, the full scale table (token, size, line height, weight, mobile value), measure limits for every body style, do/don't sentences per role, and the anti-pattern list.

Do NOT

  • Do not choose fonts before roles - personality-first selection is how three-typeface systems happen.
  • Do not add a third typeface; every future design decision pays for it.
  • Do not name styles by pixel size; renaming ripples through every component when the scale shifts.
  • Do not skip measure. Line height without a max-width still yields unreadable 120-character lines on wide screens.
  • Do not ship a scale without mobile values for Display and Heading steps - engineers will improvise them.

Quality bar

  • Two typefaces maximum, each passing all five functional criteria.
  • Every step has a role name, size, line height, weight, and mobile value.
  • Body styles sit in the 1.4-1.6 line-height band with an enforced 45-75 character measure.
  • The ratio matches the product type and is stated in the doc, so future additions extend the scale instead of inventing sizes.

Pair with color-palette-builder for the color layer, responsive-spec for breakpoint behavior, and brand-guidelines to fold the system into full identity documentation.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.