agentleFS
brayandiazcbrayandiazc/project-starter-template-en-ai/.claude/skills/design-system-audit/SKILL.md

design-system-audit

Skill2 starsChanged 2 months agoMITView on GitHub

Audits a view or component against the design system on what a script cannot check — interaction states, the four data states, native primitives versus reinvented components, and visual hierarchy. Use this when the user asks to review design-system compliance or whether a component follows the guidelines (e.g. "does this card follow our system?", "audit the settings view").

11 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use design-tokens and design-critique.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ilikescienceSkill

    design-tokens

    Expert in design tokens using the DTCG specification. Use this skill when users ask about design tokens, DTCG format, token validation, formatting, transformation, color spaces (sRGB, Display P3, OKLCH), references and aliasing, resolvers, theming with modifiers/contexts, multi-platform design systems, accessibility, or working with tools like jq, jsonata, and terrazzo. Helps with token file creation, resolver configuration, structure, naming conventions, and best practices.

    Also in 3 projects that use this

  • AnthropicSkill

    design-critique

    Get structured design feedback on usability, hierarchy, and consistency. Trigger with "review this design", "critique this mockup", "what do you think of this screen?", or when sharing a Figma link or screenshot for feedback at any stage from exploration to final polish.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 0 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 9 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read design/README.md (the system) and docs/conventions/ui.md (how the views are organized and where the brand assets are).

    line 16
  2. Very specific:

    Interaction states: default, hover, active and disabled.

    line 28
  3. Very specific:

    Do NOT redesign or restyle on your own initiative: you report and propose.

    line 46

“Common” is measured: kept by at least 6 in 10 of the 9 edited versions of this file we could read. “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

Audit the named UI against the design system in design/.

You do not review the colors. The CI Design system job (.github/scripts/check-design-tokens.sh) already fails on any hex, rgb(), palette utility (bg-blue-500) or arbitrary value (bg-[#0A7A9D]) in the views. If you think you see a raw color, it means the check does not cover that path: say so, and propose adding the folder to the script instead of becoming a human linter.

Your job is what no script can check:

  1. Read design/README.md (the system) and docs/conventions/ui.md (how the views are organized and where the brand assets are).
  2. Native primitive before a reinvented component. A dialog built with a <div> and position: fixed is a finding even if it looks fine: <dialog>, <details> and popover bring focus, keyboard and Escape solved by the browser. If there is a custom component where a primitive belonged, flag it with the concrete replacement.
  3. If the custom component is unavoidable, flag it and defer its accessibility (focus trap, Escape, aria-*, tabbing) to /accessibility-audit — that list lives there, in one place, on purpose.
  4. Interaction states: default, hover, active and disabled. (Visible focus and keyboard navigation are audited by /accessibility-audit.)
  5. The four data states, in every view that loads something: loading (a skeleton shaped like the content, not a generic spinner or "Loading…"), empty (icon + why it is empty + a CTA that orients), error (message + retry action) and success. The empty state is the one that most defines the product and the one most often skipped.
  6. Both themes: light and data-theme="dark" — the view must use the tokens in both and no piece may be stuck to a single theme. (Measuring AA contrast belongs to /accessibility-audit; design/preview.html brings it for the base tokens.)
  7. Hierarchy before decoration: can you tell what matters in the view without reading the text? Spacing and type weight before borders and shadows.

Report a list: each item Pass / Fail, with file:line and the concrete fix. Example: Modal.tsx:12 — <div role="dialog"> with no focus trap; use <dialog>.

Do NOT redesign or restyle on your own initiative: you report and propose. For a full accessibility review (measured contrast, screen readers), use accessibility-audit. design/ always wins.

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.