agentsclimarketplace

Ui ux design principles

Skill databricks-solutions/consort/skills/ui-ux-design-principles

Experience-level engineering canon, the UI/UX counterpart to software-design-principles (code-level) and architectural-design-principles (system-level). Usability heuristics, visual hierarchy, accessibility, interaction and feedback, information architecture, design systems and tokens, and testable UI. Imported by the consort UX Designer (authoring design-guide.{md,json} + ia.md and the adherence gate) and by the Driver building UI. Use when: shaping a design guide or information architecture, reviewing a user-facing surface, choosing a UI framework, or making the UI testable.From its SKILL.md

Install
npx -y skills add databricks-solutions/consort --skill ui-ux-design-principles

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.

SKILL.md

5.7 KB, as published. Nobody here has run it

ui-ux-design-principles

Canon for decisions about the USER's experience. The third of the design-principles trio:

  • software-design-principles – how a unit of code is written (SOLID, DRY, clean code).
  • architectural-design-principles – how the system is shaped (layers, twelve-factor, fitness functions).
  • ui-ux-design-principles – how the product looks, behaves, and is navigated, kept testable.

The project's design-guide.{md,json} and ia.md are the per-project instantiation of this canon, the way architecture.json instantiates the layering rules. Markdown only, no scripts: consulted, not invoked. Like the architecture canon, it points each rule at how it's enforced (the E2E/Playwright layer, the design-adherence check against design-guide.json, a11y checks). A rule no test defends is advisory; one a check defends is part of the build.

When to use

  • The consort UX Designer imports it to author design-guide.{md,json} + ia.md and define the adherence gate.
  • The Driver imports it when building UI that must adhere to the design guide.
  • You're reviewing a user-facing change and need a shared vocabulary (heuristics, hierarchy, accessibility) rather than taste.
  • You're choosing how the UI is built and need the testability rules (framework choice, stable seams).

UX review checklist (mandatory before promote/merge)

Confirm each row before declaring the design done. A blank row is fine when scope justifies it; an unconsidered row is a smell.

PropertyThe ruleEnforced by
FeedbackEvery data-changing action shows success AND failure feedback; no silent failuresE2E asserts the visible confirmation
AccessibilityKeyboard-navigable, semantic HTML, sufficient contrast, labelled controlsa11y check (axe / equivalent) at the E2E layer
Visual hierarchyThe primary action is most prominent; the eye is guideddesign-adherence (tokens) + review
ConsistencyReuses design-guide tokens + standard components; one pattern per jobdesign-adherence against design-guide.json
Information architectureScreens, navigation, primary flows defined before stylingia.md present + E2E flows trace it
Testable seamsInteractive elements expose stable selectors (data-testid / role)E2E addresses elements by seam, not brittle CSS
Loading / latencyAny action over ~200ms shows a loading state; feedback causes no layout shiftE2E + design-adherence

A property with no owner or no check: resolve it before merging.

References

  • Usability heuristics – Nielsen's ten, condensed, each with its violation smell.
  • Visual hierarchy – type scale, spacing, color, contrast, alignment: guiding the eye.
  • Accessibility – WCAG essentials, semantic HTML, ARIA, keyboard, focus, contrast. Not optional.
  • Interaction and feedback – affordances, feedback rules, loading states, error prevention/recovery, progressive disclosure.
  • Information architecture – screens, navigation, flows, mental models. The source of ia.md.
  • Design systems and tokens – tokens as source of truth, consistency, component reuse, the design-guide.md / .json relationship.
  • Testable UI – testable frameworks, stable seams, deterministic rendering, boundary-layer rendering, design-adherence at the E2E layer.

Hard rules

  1. No silent failures, no unacknowledged success. Every data-changing action shows visible success and failure feedback; feedback never shifts layout.
  2. Accessibility is not optional. Keyboard-navigable, screen-reader-friendly, sufficient contrast, labelled controls, defended by an a11y check.
  3. Consistency over novelty. Reuse the design-guide tokens and standard components; one pattern per job. design-guide.json is the source of truth.
  4. Visual hierarchy guides the eye. The most important action is the most prominent; importance maps to weight, size, position.
  5. Information architecture before pixels. Screens, navigation, and flows are defined (ia.md) before styling.
  6. The UI is testable by construction. Modern framework, deterministic render, stable seams; rendering lives in the boundary layer. See testable-ui.
  7. Progressive disclosure. Show what the user needs now; defer complexity behind intent.

Composition

  • consort – UX Designer authors design-guide.{md,json} + ia.md and sets the adherence contract; Test Strategist writes E2E scenarios + a11y checks against it (see test-strategy); Driver builds UI that adheres.
  • architectural-design-principles – testable-ui shares the boundary-layer rule (templating is a boundary adapter) and the fitness-function model.
  • software-design-principles – clean-code naming applies to components and template partials too.

Keep looking

Skills are one crate of 614,327. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.