product-ui-design
Skill1 starsChanged 56 days agoApache-2.0View on GitHub
Design, implement, or review web, mobile, and desktop product interfaces from real product context and the repository's existing design language. Use for pages, flows, components, responsive behavior, interaction states, accessibility, and visual refinement. Reject generic template aesthetics and unverified visual-completion claims. Do not use for document, slide, or technical-diagram styling.
We split this skill into the rules it gives an agent
11 rules. About 0 are common, 10 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈10
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 8
Create interfaces that feel native to the product rather than generated from a fashionable prompt.
- Distinctive: line 13
Before choosing a style, inspect the relevant product and repository evidence: user task, content, navigation, nearby screens, tokens, …
- Very specific: line 26
Do not infer a visual identity from the words AI, SaaS, modern, or from the implementation framework.
- Distinctive: line 40
Before claiming an implementation complete, apply quality-gates.md.
- Distinctive: line 49
Use semantic controls, visible keyboard focus, meaningful labels, logical focus order, sufficient contrast, zoom-safe layouts, …
- Distinctive: line 56
Use hierarchy, proportion, rhythm, type, and content before decoration.
- Distinctive: line 60
Use icons from the repository's existing icon system.
- Distinctive: line 61
Do not invent metrics, testimonials, activity, users, certifications, or product capabilities to make a screen look complete.
- Distinctive: line 63
Keep microcopy specific to the action and consequence.
- Distinctive: line 65
Treat consistency as a system property, not identical cards, pills, spacing, and animation on every surface.
- Distinctive: line 70
When implementation is requested, render the actual interface at representative narrow and wide viewports and inspect it.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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
Product UI Design
Create interfaces that feel native to the product rather than generated from a fashionable prompt. Preserve the user's chosen stack, product direction, and existing design system.
Establish the design basis
Before choosing a style, inspect the relevant product and repository evidence: user task, content, navigation, nearby screens, tokens, components, assets, typography, supported themes, and target devices. Reuse established primitives and conventions unless the request is explicitly a redesign.
For a new product with no visual language, state a short design basis containing:
- primary user and task
- information-density and usage context
- content hierarchy and critical action
- one visual thesis tied to the product's subject or behavior
- accessibility, platform, brand, and performance constraints
- assumptions that remain reversible
Do not infer a visual identity from the words AI, SaaS, modern, or from the implementation
framework. A style catalog or generated design-system suggestion is input to evaluate, not a design
decision.
Work by mode
- Design: define the user flow and content hierarchy, then layout, components, tokens, and interaction behavior. Use realistic content shapes; label unknown copy and data as placeholders.
- Implement: detect the existing stack, extend its components and tokens, preserve working behavior, and keep the change inside the requested surface.
- Review: inspect the rendered interface and code. Report issues by user impact, evidence, location, and a concrete correction. Do not replace the product's taste with personal preference.
For visual direction and anti-pattern decisions, read anti-slop.md. Before claiming an implementation complete, apply quality-gates.md.
Design the complete experience
Model the normal path and every applicable state: loading, empty, partial, error, success, disabled, permission denied, offline or stale data, long content, localization, and destructive confirmation. Make the primary task obvious without turning every action into a prominent button or every content group into a card.
Use semantic controls, visible keyboard focus, meaningful labels, logical focus order, sufficient contrast, zoom-safe layouts, reduced-motion behavior, and non-color status cues. On touch surfaces, respect platform navigation, safe areas, and adequately sized targets. Prefer motion that explains a state or spatial change; omit motion that merely advertises polish.
Preserve authorship
- Use hierarchy, proportion, rhythm, type, and content before decoration.
- Give a screen one dominant visual idea. Supporting elements should not compete with it.
- Derive distinctive details from product content, interaction, or brand assets—not arbitrary blobs, gradients, glows, mascots, or geometric ornaments.
- Use icons from the repository's existing icon system. Do not substitute emoji or mix icon families.
- Do not invent metrics, testimonials, activity, users, certifications, or product capabilities to make a screen look complete.
- Keep microcopy specific to the action and consequence. Remove promotional filler from operational interfaces.
- Treat consistency as a system property, not identical cards, pills, spacing, and animation on every surface.
Completion
When implementation is requested, render the actual interface at representative narrow and wide viewports and inspect it. Exercise keyboard navigation and the important interaction states when the environment permits. Fix visible defects, then re-render. A successful build or screenshot command does not prove visual quality.
Report changed paths, the design basis, reused and introduced tokens/components, states verified,
viewport and accessibility checks, and unresolved assumptions. If the interface could not be
rendered or inspected, say visual verification: NOT_RUN and do not claim the UI is finished.
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- JarodFroneman/jstack★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository65 other files it gives its agents.
- agent-evalSkill
- architecture-review-workflowSkill
- backend-code-review-workflowSkill
- code-authoring-routerSkill
- code-diagram-extractSkill
- codex-agent-runtimeSkill
- design-pattern-applySkill
- diagram-stylerSkill
- document-designerSkill
- e2e-evidence-reportSkill
- e2e-report-rendererSkill
- e2e-runnerSkill
- e2e-scenario-authorSkill
- e2e-test-workflowSkill
- entitySkill
- enumSkill
- explore-dead-codeSkill
- flywaySkill
- git-flowSkill
- harness-proposeSkill
- harness-refineSkill
- java-spring-review-workflowSkill
- knowledge-auditSkill
- korean-dev-writerSkill
- kotlin-code-authoringSkill
- kotlin-spring-reviewSkill
- kotlin-spring-review-workflowSkill
- learnSkill
- mapperSkill
- markdown-publisherSkill
- mermaid-diagramsSkill
- mssql-code-authoringSkill
- nl-sql-querySkill
- notion-domain-contextSkill
- oop-design-reviewSkill
- opensearch-code-authoringSkill
- opensearch-optimizationSkill
- opensearch-query-reviewSkill
- opensearch-review-workflowSkill
- opensearch-schema-reviewSkill
- pdf-qaSkill
- playwright-e2eSkill
- postgres-code-authoringSkill
- postgres-query-reviewSkill
- postgres-review-workflowSkill
- postgres-risk-checkSkill
- postgres-schema-designSkill
- product-requirementsSkill
- refactoring-catalogSkill
- repositorySkill
- requestSkill
- responseSkill
- retroSkill
- schema-doc-extractSkill
- security-scan-workflowSkill
- serviceSkill
- skill-updaterSkill
- start-taskSkill
- test-authoringSkill
- test-quality-refactoring-workflowSkill
- The first 60 shown.
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.

