agentleFS
paulasilvatechpaulasilvatech/awesome-copilot-adventures/.github/skills/frontend-visual-system/SKILL.md

frontend-visual-system

Skill1 starsChanged 2 months agoMITView on GitHub

Define evidence-based typography, semantic color, layout, density, imagery, iconography, themes, and motion within an existing product system. Use this skill when frontend visual direction, design tokens, theming, composition, or visual-quality review is requested.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. About 0 are common, 15 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈15

    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
  1. Distinctive:

    Use a bounded semantic scale and support long content, localization, text resizing, dynamic type, and user font preferences where …

    line 24
  2. Distinctive:

    Avoid arbitrary per-component sizes, unpredictable viewport-width scaling, and unjustified letter spacing.

    line 25
  3. Distinctive:

    Read references/typography.md for the decision record.

    line 27
  4. Distinctive:

    Start from semantic roles for surfaces, text, borders, actions, focus, status, and data series.

    line 31
  5. Distinctive:

    Use cards for repeated comparable items, modals, or genuinely framed tools, not every section.

    line 40
  6. Distinctive:

    Prefer alignment, grouping, rhythm, contrast, and whitespace over ornamental containers.

    line 42
  7. Distinctive:

    Use icons and imagery with a clear information, brand, or state purpose.

    line 43
  8. Distinctive:

    Use motion to explain state, continuity, progress, hierarchy, or spatial relationships.

    line 49
  9. Distinctive:

    Prefer transform and opacity when appropriate, honor reduced motion, and provide non-motion comprehension.

    line 50
  10. Distinctive:

    Avoid scroll hijacking, custom cursors, continuous parallax, and decorative animation without an approved reason.

    line 51
  11. Distinctive:

    Use a trend only when it supports the primary job, fits the brand and trust context, works across required conditions, preserves hierarchy …

    line 57
  12. Very specific:

    Use assets/human-review-checklist.md and record evidence for pass, needs revision, blocked, or not applicable.

    line 59
  13. Distinctive:

    Do not create a brand identity or replace an approved design system.

    line 63
  14. Distinctive:

    Do not copy proprietary assets, exact layouts, text, or branding from reference products.

    line 64
  15. Distinctive:

    Do not treat Apple, Material, Fluent, or a current trend as a universal standard.

    line 65
  16. Distinctive:

    Do not approve contrast, runtime rendering, or motion behavior without applicable evidence.

    line 66

≈ “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

Frontend visual system

Translate product evidence and an established design system into a concrete visual grammar without substituting trends for information hierarchy.

When to invoke

  • "Define the visual direction for this product screen."
  • "Review our typography, color, spacing, and density."
  • "Create semantic frontend tokens without replacing our design system."
  • "Make this interface less generic while preserving the brand."
  • "Audit motion, themes, icons, and visual hierarchy."

Criteria

Typography

  • Reuse the brand type system when it exists.
  • Select roles by language coverage, readability, hierarchy, loading, and product character.
  • Use a bounded semantic scale and support long content, localization, text resizing, dynamic type, and user font preferences where applicable.
  • Avoid arbitrary per-component sizes, unpredictable viewport-width scaling, and unjustified letter spacing.

Read references/typography.md for the decision record.

Color and themes

  • Start from semantic roles for surfaces, text, borders, actions, focus, status, and data series.
  • Preserve meaning without color and check text, icons, controls, focus, charts, dark mode, high contrast, and forced colors.
  • Preserve brand colors through accessible pairings and fallbacks rather than flattening the product into one hue.

Read references/color-and-contrast.md.

Layout, density, and assets

  • Let the workflow choose density and composition.
  • Use cards for repeated comparable items, modals, or genuinely framed tools, not every section.
  • Establish stable dimensions for data regions, media, controls, and content that would otherwise shift.
  • Prefer alignment, grouping, rhythm, contrast, and whitespace over ornamental containers.
  • Use icons and imagery with a clear information, brand, or state purpose.

Read references/layout-and-density.md.

Motion and feedback

  • Use motion to explain state, continuity, progress, hierarchy, or spatial relationships.
  • Prefer transform and opacity when appropriate, honor reduced motion, and provide non-motion comprehension.
  • Avoid scroll hijacking, custom cursors, continuous parallax, and decorative animation without an approved reason.

Read references/motion.md.

Trend applicability gate

Use a trend only when it supports the primary job, fits the brand and trust context, works across required conditions, preserves hierarchy and budgets, integrates with the existing system, and beats a simpler established pattern.

Use assets/human-review-checklist.md and record evidence for pass, needs revision, blocked, or not applicable.

Limits

  • Do not create a brand identity or replace an approved design system.
  • Do not copy proprietary assets, exact layouts, text, or branding from reference products.
  • Do not treat Apple, Material, Fluent, or a current trend as a universal standard.
  • Do not approve contrast, runtime rendering, or motion behavior without applicable evidence.

Output template

## Visual system result
**Status:** ready | needs revision | blocked

### Evidence
| Source | Existing rule | Gap |
| --- | --- | --- |

### Visual contract
| Area | Semantic decision | Product rationale | Token/component impact |
| --- | --- | --- | --- |

### Runtime and accessibility evidence required
- <check>

### Rejected generic defaults
- <default and reason>

Quality gate

  • Decisions start from product and repository evidence.
  • Typography, semantic color, layout, density, assets, themes, and motion are considered when applicable.
  • Every addition fits or deliberately extends the existing system.
  • Hierarchy and product purpose justify decorative choices.
  • Long content, localization, zoom, contrast modes, and reduced motion have defined behavior.
  • The human review checklist has no unresolved blocked item.

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.