本文へ移動
cccskills
無料GitHub で公開

typography-scale-builder

Build a complete typographic system -- modular or fluid type scale, line-height and tracking per size, weight roles, and vertical rhythm -- exported as tokens, Tailwind config, and CSS. Pairs with font-pairing-suggester (which picks the fonts; this builds the system they live in).

インストール方法を見る

含まれるファイル(1)

  • SKILL.md3.8 KB

SKILL.md(原文)

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。

Typography Scale Builder

Typography is the design system's skeleton: most products that look "off" have inconsistent type long before they have bad colors. Build the scale as a system with roles and math, then make the codebase actually use it. Input: the chosen fonts (or run font-pairing-suggester first), the product type (marketing site, dense app UI, long-form editorial), and the target range of devices.

Workflow

  1. Audit what exists. Inventory every font-size, line-height, letter-spacing, and font-weight in the codebase. The typical finding -- 23 font sizes, 9 of them within 2px of each other -- is the case for the system. Map which are actually distinct roles vs. accumulated drift.
  2. Choose the scale math. Base size (16px body floor for marketing/editorial; 13-14px acceptable for dense app UI) and ratio -- modest for UI-heavy products (1.125-1.2, more usable steps, less drama), larger for marketing/editorial (1.25-1.333, real hierarchy). Fluid by default for marketing sites: clamp() per step so headings scale between viewport bounds without breakpoint jumps; fixed steps are fine for app UI.
  3. Set the companions per step -- the part single-number "scales" skip: line-height tightens as size grows (body ~1.5-1.6, headings down to 1.1-1.2 at display sizes); letter-spacing goes negative at display sizes (-1% to -2.5%) and slightly positive for small caps/labels; weight roles per font (which weight is body, emphasis, heading, display -- and which loaded weights get deleted because nothing uses them).
  4. Define the roles. Semantic tokens, not size names: display, h1-h4, body-lg, body, body-sm, caption, overline, code. Each role = size step + line-height + tracking + weight + family, as one token. Components consume roles; nobody composes raw steps.
  5. Vertical rhythm. Spacing-after rules per role (heading margins proportional to their size, paragraph spacing from body line-height), text-wrap: balance on headings, and measure limits (65-75ch for body prose) so line length stays readable at every viewport.
  6. Export and retrofit. typography-tokens.json, Tailwind fontSize config (size + line-height + tracking tuples per role; v4 @theme), typography.css with role classes, and TYPOGRAPHY.md. Then map the step-1 inventory onto roles and replace -- with a per-case flag on anything that does not cleanly map (it is either a missing role or drift to delete).

Rules

  • Roles over sizes everywhere: text-h2, never text-[28px] -- and no role means the component asks the system for one, not the developer for a number.
  • Every step ships with its line-height and tracking. A size without its companions is half a token.
  • The scale has as few steps as hierarchy needs -- 7-9 roles cover almost any product; 12+ means roles are duplicating.
  • Body text is the anchor: it gets sized for reading first, and the scale grows around it -- never shrink body to make headings look bigger.
  • Load only the weights the roles use. Every unused weight is dead bytes on first paint.
  • Fluid type gets floor and ceiling checks: the clamp() minimum must stay readable on a 320px phone and the max must not blow the layout on ultrawide.

Quick Commands

  • "Build a type system for this project" -- full workflow
  • "Audit my typography" -- step 1 only
  • "Make my headings fluid" -- fluid conversion of an existing scale
  • "Retrofit components" -- step 6 against an existing system

レビュー

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。

同じリポジトリのスキル

概要と使いどころ

Audit websites for accessibility issues and WCAG compliance. Use when checking accessibility, fixing a11y issues, or ensuring WCAG compliance.

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。

OneWave-AI/claude-skills3392026年10月2日 更新

Deploy a 2-layer parallel agent hierarchy for large, parallelizable work — big refactors, multi-file migrations, codebase-wide audits, bulk generation. A top-tier commander (Fable or Opus) orchestrates the swarms; the user picks a power level (Max Power / Heavy / Balanced / Economy) that sets the Opus/Sonnet/Haiku model mix per layer. Layer 1 is 3-50+ specialist agents, each with its own full context window; Layer 2 is 2+ sub-agents per member. Includes git safety, tiered sizing, a pre-deploy gate, phantom-completion checks, and multi-wave follow-up.

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。

OneWave-AI/claude-skills3392026年10月2日 更新

Use to test whether an AI agent, prompt, or skill actually works before or after it ships — build a small eval set, score runs, and catch regressions when the prompt or model changes. Trigger when the user asks whether a prompt is good, wants to compare models or versions, sees inconsistent agent output, or is about to put an agent in front of customers.

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。

OneWave-AI/claude-skills3392026年10月2日 更新

Deploys swarms of sub-agents for massive parallel data processing tasks. Unlike agent-army (which is for code changes), this is for DATA tasks -- processing 1000 documents, analyzing datasets, bulk content generation. Configurable swarm size, task distribution, result aggregation, progress tracking, and error recovery.

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。

OneWave-AI/claude-skills3392026年10月2日 更新

Designs and deploys custom agent teams for specific business workflows. Interactive discovery of business processes, then generates complete team configurations with specialized agent roles, tool access, communication protocols, and handoff rules.

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。

OneWave-AI/claude-skills3392026年10月2日 更新

Agent-to-Agent (A2A) communication protocol. Connect two or more Claude agents that pass messages, share context, delegate tasks, and collaborate. Implements structured handoffs, shared memory, and multi-agent conversations.

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。

OneWave-AI/claude-skills3392026年10月2日 更新

OneWave-AI のスキルをすべて見る

このスキルの問題を報告する