uinaf/uinaf-design-system
Apply the uinaf brand identity to anything that ships under the uinaf name — web interfaces, blog posts, changelogs, documentation, READMEs, slides, OG / social images, email, terminal banners, app or product UI starting points. Covers voice, design tokens, components, motion, and brand assets, with a Tailwind v4 path for web work. Use when producing or restyling any uinaf-branded artefact. Skip for non-uinaf work; this is opinionated brand guidance, not a generic UI kit.
96
100%
Does it follow best practices?
Impact
91%
1.56xAverage score across 5 eval scenarios
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly defines its scope (uinaf brand identity), enumerates specific artifact types and coverage areas, provides explicit 'Use when' and 'Skip when' guidance, and distinguishes itself from generic UI/design skills. The description is comprehensive yet concise, uses third-person voice throughout, and includes natural trigger terms that would help Claude select it appropriately.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and artifact types: web interfaces, blog posts, changelogs, documentation, READMEs, slides, OG/social images, email, terminal banners, app/product UI. Also specifies what it covers: voice, design tokens, components, motion, brand assets, and a Tailwind v4 path. | 3 / 3 |
Completeness | Clearly answers both 'what' (apply uinaf brand identity across various artifact types, covering voice, design tokens, components, motion, assets) and 'when' with an explicit 'Use when producing or restyling any uinaf-branded artefact' clause, plus a helpful exclusion: 'Skip for non-uinaf work.' | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms a user would say: 'uinaf' (the brand name itself is the primary trigger), plus artifact types like 'blog posts', 'documentation', 'READMEs', 'slides', 'email', 'terminal banners', 'UI', 'brand', 'Tailwind'. The brand name 'uinaf' is repeated multiple times ensuring strong matching. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive due to being scoped to a specific brand ('uinaf'). The explicit exclusion 'Skip for non-uinaf work; this is opinionated brand guidance, not a generic UI kit' further reduces conflict risk with generic design or UI skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is an exceptionally well-crafted skill file. It delivers dense, actionable brand guidance with zero fluff, maintains a clear workflow with validation, and uses progressive disclosure masterfully — keeping universal hard rules inline while routing task-specific detail to clearly labeled reference files. The CDN URLs, CSS class names, and specific constraints (6px max radius, no emoji, specific Unicode characters) make it immediately executable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every section earns its place. Hard rules are terse and specific, no concepts are over-explained, and Claude's intelligence is respected throughout. The skill avoids explaining what a design system is or why brand consistency matters — it just delivers the rules. | 3 / 3 |
Actionability | Concrete and specific throughout: exact typeface names, exact CDN URLs, exact CSS class names, pixel-level radius caps, specific Unicode characters for iconography, and named color roles. Copy rules give explicit anti-patterns ('empower', 'unlock', 'synergies'). The bundled CSS file and CDN paths are copy-paste ready. | 3 / 3 |
Workflow Clarity | The 5-step workflow is clearly sequenced: identify surface → load references → pull assets → write copy → verify against hard rules. Step 5 is an explicit validation checkpoint. The 'Handoffs' section provides clear next-action routing for verification, review, and docs updates. | 3 / 3 |
Progressive Disclosure | Exemplary progressive disclosure: the SKILL.md is a concise overview with hard rules inline (since they apply universally) and all detailed guidance pushed to clearly labeled, one-level-deep references organized by task. The 'Read by task' section is a well-signaled navigation index with specific file paths and scope descriptions. | 3 / 3 |
Total | 12 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
Reviewed
Table of Contents