agentleFS
alfonsusenricoalfonsusenrico/pakem/skills/ui-ux-quality/SKILL.md

ui-ux-quality

Skill0 starsChanged 10 days agoMITView on GitHub

Use when creating, materially changing, or reviewing a user interface or interaction - deciding what the interface is for, navigation, layout, data display, visual design, frontend structure, accessibility, or choosing a design direction. Works from the product's purpose and verifies accessibility with evidence.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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. Very specific:

    Follow surfaces: decide what the product mostly is - for example a data dashboard, a public website, a web application, or a mix - record …

    line 21
  2. Distinctive:

    Follow frontend engineering: tokens first, shared primitives, size limits, purpose-built phone composition, formatter.

    line 35
  3. Distinctive:

    No captions or explanations that do not lead to an action or a decision.

    line 43

≈ “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

UI and UX Quality

1. Authority

  • The project's approved users, content, design system, tokens, and components come first. Read the project AGENTS.md (its Design surface:), its tokens file, and its design notes before changing anything visual.
  • Visual style - colour, type, shape, icons, motion, tone - is a project decision, recorded in the project's tokens file and design notes. The framework sets none.
  • Where the project has no design yet, choose a direction with the owner (section 3); never invent a style silently. After a direction is accepted: refine, don't restyle.
  • Where taste is uncertain, prefer research-backed defaults and cite the source.

2. Start from what the interface is for

Follow surfaces: decide what the product mostly is - for example a data dashboard, a public website, a web application, or a mix - record it as Design surface: in AGENTS.md, and apply that surface's goals, questions, and checks. The examples there are not a closed list; derive the same parts for anything else.

3. Choosing or changing a direction

For a new product direction or a major redesign, follow design direction: build two to four working alternatives, compare them in the browser with the owner, record the choice and the rejected options. Do not pick a direction from a written brief alone.

4. Building

Follow frontend engineering: tokens first, shared primitives, size limits, purpose-built phone composition, formatter.

5. Rules for every interface

  • Truthful content only: no invented metrics, outcomes, testimonials, or placeholder labels presented as real. Unknown is never shown as 0. Progress comes only from real counts.
  • No captions or explanations that do not lead to an action or a decision.
  • Copy in the audience's language; engineering identifiers follow the project's Language: line and its code conventions.
  • Light and dark themes (default:themes); the choice is persisted and follows the operating-system preference until the user changes it. Check contrast in each theme. A product with a single theme by design (a brand rule, a kiosk) overrides the label.
  • Design each theme for its own background: choose and check a light theme on light surfaces, never by inverting or reusing the dark one. Check every text, icon, mark and divider against the surface it sits on, in every theme, including content drawn by other programs (an embedded terminal, a third-party widget). Brand marks keep their colours.
  • Meaning is never carried by colour alone.
  • Phone layouts are designed for the phone, not stacked desktop columns.

6. Accessibility and verification

Target WCAG 2.2 AA and verify with accessibility and evidence. Never claim conformance without evidence that covers the pages, states, widths, and flows involved.

7. Review output

Report findings with file and line references, separated into code inspection, automated checks, browser checks, and assistive-technology checks. Do not invent metrics or evidence.

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.