css-typography-ko
Skill0 starsChanged 33 days agoMITView on GitHub
Improve Korean web UI readability with CSS typography when designing, implementing, or reviewing Korean text, awkward line breaks, text hierarchy, spacing, or overflow. Use for Korean prose and mixed Korean/Latin interfaces; prose rewriting and native app typography are outside this skill's scope.
In this repository12 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- apple-app-icon-generatorSkill
- apple-bundle-id-guardrailsSkill
- break-it-downSkill
- docs-syncSkill
- flutter-flavorsSkill
- github-workflowSkill
- git-workflowSkill
- macos-dev-app-cleanupSkill
- macos-release-reinstallSkill
- prepare-release-githubSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
Use real Korean copy, including mixed Latin text and numbers.
- Distinctive: line 26
Line composition: Prefer Korean eojeol (어절, space-delimited units) boundaries for prose when the container permits.
- Distinctive: line 34
Use the project's CSS, modules, utilities, or tokens.
- Very specific: line 36
Add balance, pretty, or font-relative units as enhancements with a usable baseline.
- Distinctive: line 41
When a runnable page and browser tools are available, compare the affected components before and after using the same content and viewport.
- Distinctive: line 47
If runtime verification is unavailable, inspect the code and specify what still needs browser verification.
≈ “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.
Files
Korean CSS Typography
Make Korean text comfortable to read and easy to scan across screen sizes. Treat font choice, hierarchy, spacing, line length, and wrapping as one design decision, grounded in the actual content and existing visual identity. The primary outcome is readable text and reliable text layout; account for rendering cost when selecting enhancements.
Establish the reading context
- Identify the requested mode: design advice, implementation, or review. Implement when requested; report findings when reviewing.
- Inspect the affected components, typography tokens, font loading, language markup, and supported browsers. Reuse these conventions; add a shared token or utility only when the same decision recurs.
- Distinguish reading-heavy prose, scanning-heavy lists or tables, display headings, controls, and user-authored content. Establish the reader's task before changing density.
- Use real Korean copy, including mixed Latin text and numbers. Derive samples from the product; use the reference samples only when content is unavailable.
Make typography decisions
Work through the relevant decisions below. Read CSS decisions and examples when choosing values or resolving a specific rendering issue.
For wrapping issues, resolve four separate choices: word-break sets ordinary break opportunities, overflow-wrap allows emergency breaks, white-space controls whitespace preservation and wrapping mode, and text-wrap chooses the wrapping mode and composition style. Use keep-all as a starting point for Korean prose, balance for short headings, and pretty selectively for body copy. Preserve authored whitespace by its role, and choose nowrap only when the full label can fit. Evaluate these choices together with container sizing.
- Glyphs and weight: Confirm that the rendered font covers Hangul and that Korean, Latin, digits, and punctuation look coherent together. Check the fallback and available weights before replacing a font or increasing weight.
- Hierarchy: Define heading, body, label, and supporting-text roles using the existing scale. Combine size, weight, spacing, and placement; keep supporting text legible and distinguish links without relying solely on color.
- Reading rhythm: Tune font size, unitless line height, paragraph spacing, and text-column width together. Begin with normal tracking and adjust only after seeing the chosen Hangul font. Treat numerical starting points as heuristics, not accessibility requirements.
- Line composition: Prefer Korean eojeol (어절, space-delimited units) boundaries for prose when the container permits. Allow emergency breaks for an oversized unbroken string. Judge the rendered composition rather than expecting semantic phrase analysis or guaranteed removal of a short final line.
- Density and content preservation: Let controls and containers accommodate their labels. Preserve authored line breaks where they convey structure. Choose truncation only when the reader can retrieve the full content, especially for errors and instructions.
For each change, connect an observed or anticipated reading problem to the chosen adjustment. Keep established brand choices unless evidence shows they impair the requested task.
Apply CSS in context
- Scope rules to the affected roles and language. Use accurate
langmarkup and:lang(ko)where appropriate; Korean defaults are not a blanket policy for every CJK language. - Use the project's CSS, modules, utilities, or tokens. Preserve code blocks, editors, other locales, and intentional single-line or preformatted components.
- Resolve container constraints before hiding overflow: inspect flex/grid minimum sizing, available width, fixed heights, and padding. Smaller text is not the default fix for a crowded control.
- Add
balance,pretty, or font-relative units as enhancements with a usable baseline. Verify version-specific behavior against current official documentation when it affects the supported browser target. - Preserve wording and meaningful breaks. If readability requires editing copy or changing the layout beyond the request, explain the tradeoff and stay within the authorized scope.
Verify the rendered result
When a runnable page and browser tools are available, compare the affected components before and after using the same content and viewport. Use the environment's authorized browser tools; this skill requires no particular browser product.
Check representative narrow and wide layouts, including a 320 CSS-pixel viewport for ordinary flowing content, and enlarged text/zoom. Exercise long Korean terms, Latin identifiers or URLs, numbers with units, multiline labels, and authored line breaks as relevant. For font changes, compare loading/fallback and loaded states on the available target platforms.
Inspect computed styles and the rendered font when inherited rules or mixed glyphs are involved. Confirm no unintended clipping, overlap, horizontal scrolling, inaccessible truncation, or loss of content. Check text contrast and user spacing overrides for affected text; use the reference's accessibility criteria without claiming a full accessibility audit.
If runtime verification is unavailable, inspect the code and specify what still needs browser verification. Proposed CSS and a screenshot alone do not establish cross-browser behavior.
Deliver
Provide the requested design recommendation, scoped code change, or review findings. Briefly explain the reading problem, the decision, and the validation performed. State material unverified browser/font cases and remaining tradeoffs. An implementation is complete when the affected roles are updated and available checks pass; a review is complete when actionable findings identify the affected role or selector and its reading impact.
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
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.

