SkillAgentSearch skills...

web-typography

Select, pair, and implement typefaces for web projects

Install / Use

npx skills add wondelai/skills --skill web-typography

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

88/100

Category

Design

Supported Platforms

Universal

Our assessment of web-typography

web-typography scores 88/100 on our quality scale, 117th of 260 Design skills we index (top 45%).

Its SKILL.md is 18 KB long, well organised into 16 sections with 7 code examples: a thorough specification that gives an agent plenty to work with.

With 2,248 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
30/30
Structure
20/20
Description
8/15
Adoption
14/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated about 2 months ago, so web-typography is actively maintained.
  • Our last check on 2026-11-01 found the source still online.
  • It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-10-02. Automated pattern scan on 2026-10-02. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

web-typography compared with similar skills

All 4 of these similar skills score higher than web-typography; compare them before choosing.

SkillScoreStarsUpdatedFormat
web-typography (this skill)by wondelai882.2k2mo agoSKILL.md
taste-skillby Leonxlnx100116.8k8d agoCLAUDE.md
Claude-Code-Game-Studiosby Donchitos10039.7k7d agoCLAUDE.md
x402by browser-use100121.0k3d agoSKILL.md
algorithmic-artby anthropics100177.9k55d agoSKILL.md

Frequently asked questions

How do I install web-typography?
Run npx skills add wondelai/skills --skill web-typography. The install tabs above show the steps for each supported agent.
Which AI agents does web-typography work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is web-typography safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is web-typography still maintained?
The repository was last updated about 2 months ago, so web-typography is actively maintained.

name: web-typography description: 'Select, pair, and implement typefaces for web projects. Use when the user mentions "font pairing", "which typeface", "line height", "responsive typography", "web font loading", "type hierarchy", "variable fonts", "FOUT/FOIT", "typographic scale", or "the text is hard to read". Also trigger when choosing between system fonts and web fonts, optimizing font-loading performance, or designing readable long-form content. Covers readability evaluation, CSS implementation, and performance optimization. For overall UI design systems, see refactoring-ui. For dramatic typographic experiences, see top-design.' license: MIT metadata: author: wondelai version: "1.5.0"

Web Typography

A practical guide to choosing, pairing, and implementing typefaces for the web. The best typography is invisible — it immerses readers in content rather than calling attention to itself.

Core Principle

Typography is the voice of your content. The typeface you choose sets tone before a single word is read — a legal site shouldn't feel playful; a children's app shouldn't feel corporate. Follow the "clear goblet" principle: typography should be like a crystal-clear wine glass, keeping focus on the wine (content), not the glass (type).

Scoring

Goal: 10/10. Score = number of the 10 Quick Diagnostic rows the implementation satisfies. Bands: 9-10 = body 16px+, measure under 75ch, line-height 1.4+, clear level contrast, font payload under 200KB, fallbacks set, survives 200% zoom; 5-6 = readable but missing measure control, fallbacks, or zoom resilience; <=3 = sub-16px body, no measure cap, FOIT, or unreadable hierarchy. Always state the current score and the specific diagnostic rows failing.

Two Contexts for Type

All typography falls into two categories:

| Context | Purpose | Priorities | |---------|---------|------------| | Type for a moment | Headlines, buttons, navigation, logos | Personality, impact, distinctiveness | | Type to live with | Body text, articles, documentation | Readability, comfort, endurance |

Workhorse typefaces excel at "type to live with" — versatile across sizes, weights, and contexts without drawing attention. Examples: Georgia, Source Sans, Freight Text, FF Meta.

Typography Framework

1. How We Read

Core concept: Understanding reading mechanics is the foundation for every typography decision. Eyes don't scan smoothly — they jump in bursts.

Why it works: Fighting these mechanics creates friction that drives readers away; aligning with them lets readers absorb content faster with less fatigue.

Key insights:

  • Saccades — eyes jump in 7-9 character bursts; line length and letter spacing directly affect saccade efficiency
  • Fixations — eyes pause briefly to absorb content; dense or poorly spaced text slows reading
  • Word shapes (bouma) — experienced readers recognize word silhouettes, not individual letters
  • Legibility vs. readability — legibility is whether characters can be distinguished (a typeface concern); readability is whether text can be comfortably read for extended periods (a typography concern: size, spacing, line length). A legible typeface can still be set unreadably

Product applications:

| Context | Application | Example | |---------|------------|---------| | Long-form content | Optimize for sustained comfort | 16-18px body, 1.5-1.7 line height, 45-75 char lines | | Dashboard UI | Optimize for rapid scanning | Distinct weight hierarchy, whitespace between data groups | | Mobile reading | Account for distance and lighting | Larger body (17-18px), higher contrast |

Copy patterns:

.prose {
  font-size: 1.125rem;     /* 18px */
  line-height: 1.6;
  max-width: 65ch;          /* ~45-75 characters */
}

See: references/typeface-anatomy.md when you need to name letterform parts (x-height, counter, aperture) or place a face in a classification system to justify a choice.

2. Evaluating Typefaces

Core concept: A typeface must pass technical, structural, and practical quality checks before it earns a place in a project. Beautiful specimens fail on screen.

Why it works: Screen rendering, variable bandwidth, and diverse devices impose constraints print never faced. Rigorous evaluation prevents costly mid-project typeface swaps.

Key insights:

  • Technical quality — consistent stroke weights, even visual color across text blocks, good kerning pairs (AV, To, Ty), complete character set, multiple weights (minimum: regular, bold, italic)
  • Structural assessment — generous x-height (better screen readability), open counters and apertures (a, e, c), distinct letterforms (Il1, O0, rn vs. m)
  • Practical needs — test at actual use sizes on target screens, check file size, verify the license
  • Real content testing — Lorem ipsum hides problems with character frequency, word length, and paragraph rhythm

Product applications:

| Context | Application | Example | |---------|------------|---------| | Body text selection | Prioritize x-height, open counters, even color | Source Serif Pro over Didot for long reads | | UI/System text | Prioritize small-size legibility and weight range | Inter or SF Pro for interface elements | | Multilingual product | Verify glyph coverage for target languages | Noto Sans for broad Unicode support |

Copy patterns:

/* Stress-test at every actual use size */
body { font-size: 16px; }
.caption { font-size: 0.75rem; }
h1 { font-size: 3rem; }

Ethical boundary: Always verify the web license before shipping a font — desktop/print licenses rarely cover web embedding, and unlicensed use creates real legal liability.

See: references/evaluating-typefaces.md when vetting a candidate face — full quality checklist, red-flags table, and the structural criteria to inspect.

3. Choosing Typefaces

Core concept: Start with purpose, not aesthetics. The content's tone, reading context, and duration should drive selection — not personal preference or trends.

Why it works: Purpose-driven choices feel inevitable rather than arbitrary, and survive stakeholder review because they can be justified with reasoning rather than taste.

Key insights:

  • Define the job first — body text, headlines, and UI elements may each need different faces
  • Match tone to content — a financial report needs a different voice than a bakery menu
  • Check the family — confirm needed weights, italics, and styles exist before committing
  • Safe starting points — body serif: Georgia, Source Serif Pro, Charter; body sans: system fonts, Source Sans Pro, Inter, IBM Plex Sans

Product applications:

| Context | Application | Example | |---------|------------|---------| | Content-heavy site | Workhorse serif or sans for sustained reading | Source Serif Pro or Charter for articles | | SaaS dashboard | Clean sans with strong tabular figures | Inter or IBM Plex Sans for data-rich UIs | | Accessibility-focused | Faces designed for maximum legibility | Atkinson Hyperlegible for vision-impaired users |

Copy patterns:

/* Web font with system fallback stack */
body {
  font-family: 'Source Sans Pro', -apple-system,
               BlinkMacSystemFont, 'Segoe UI', Roboto, sans-serif;
}

See: references/evaluating-typefaces.md when narrowing finalists or weighing free vs. paid faces — the side-by-side comparison method and quality-option shortlists.

4. Pairing Typefaces

Core concept: Successful pairings create clear contrast — faces should be obviously different, not confusingly similar. One to two typefaces maximum.

Why it works: Clear structural contrast (serif + sans, light + bold, humanist + geometric) gives each face a distinct role. Faces that are too similar create tension without purpose — readers sense something is "off" without knowing why.

Key insights:

  • Contrast types — structure (serif + sans), weight (light + regular), era (humanist + geometric), width (condensed + normal)
  • Same designer strategy — faces by one designer often share harmonizing DNA (FF Meta + FF Meta Serif)
  • Superfamilies — families designed to work together eliminate guesswork (Roboto + Roboto Slab)
  • Pairing failures — two near-identical faces, both faces competing for attention, one face overwhelming the other

Product applications:

| Context | Application | Example | |---------|------------|---------| | Editorial site | Serif headlines + sans body | Playfair Display + Source Sans Pro | | Documentation | Monospace code + sans prose from one family | IBM Plex Mono + IBM Plex Sans | | Minimal brand | Single family with weight variation | Inter at varying weights and sizes |

Copy patterns:

/* Classic serif + sans-serif pairing */
h1, h2, h3 { font-family: 'Playfair Display', Georgia, serif; }
body { font-family: 'Source Sans Pro', -apple-system, sans-serif; }

See: references/pairing-strategies.md when picking a second face — proven combinations, the contrast-type table, and same-designer/superfamily shortcuts.

5. Typographic Measurements

Core concept: Three measurements — font size, line length, and line height — form the foundation of comfortable reading. Getting these right matters more than typeface choice.

Why it works: These measurements govern how the eye tracks across and down text: optimal line length matches the saccade pattern, adequate line height prevents the eye from jumping to the wrong line on the return sweep, and sufficient size makes letterforms recognizable on screen.

Key insights:

  • Body size — 16px minimum; err larger (18px) for reading-heavy sites; mobile users hold phones farther than designers assume
  • Line length (measure) — 45-75 characters ideal, 66 optimal; enforce with ch units or max-width
  • Line height — 1.4-1.8 for body; longer lines need more; headlines need tighter (1.1-1.25)
  • Heading scale — consistent ratio (1.2-1.5) between levels creates hierarchy without extremes

Product applications:

| Context | Application | Example | |---------|------------|---------| | Blog / article | 65ch max-width, 1.6 line height | .prose { max-width: 65ch; line-height: 1.6; } | | Dashboard | Tighter line height for dense data | line-height: 1.3; for table cells and labels | | Landing page | Generous sizing for scanability | font-size: 1.25rem; line-height: 1.7; |

Copy patterns:

.prose {
  font-size: clamp(1rem, 0.95rem + 0.25vw, 1.125rem);
  line-height: 1.6;
  max-width: 65ch;
}
/* Line height by context */
h1, h2 { line-height: 1.1; }      /* headlines: 1.1-1.25 */
.ui-text { line-height: 1.35; }   /* UI: 1.3-1.4 */
.body-text { line-height: 1.6; }  /* body: 1.5-1.7 */

See: references/responsive-typography.md when writing the clamp() formulas — how to derive min/preferred/max and viewport-based measurement strategies.

6. Building Type Hierarchies

Core concept: Hierarchy tells readers what matters most. Create distinction through controlled variation in size, weight, and color — but don't pull all three levers at once.

Why it works: Deliberate, consistent differences between levels let readers grasp page structure at a glance; without hierarchy everything competes and nothing wins.

Key insights:

  • Three levers — size, weight, color; vary one or two between adjacent levels, never all three
  • The squint test — squinting at a page should still reveal the hierarchy
  • Consistent scale — a modular ratio (1.2-1.5) between heading levels creates rhythm; arbitrary sizes create noise
  • Don't skip levels — jumping H1 to H3 breaks the reader's mental model

Product applications:

| Context | Application | Example | |---------|------------|---

Truncated for display — read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars2.2k
CategoryDesign
Updated2mo ago
Forks228

Languages

Shell

Trust signals

100/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

No cautions