Apply Stave's desktop-first design system when a task changes UI, layout, theme, dialogs, sidebars, empty states, settings, prompt input, or other visual UX in this repo. Use for prompts like "디자인", "UI", "redesign", "polish", "sidebar", "dialog", "settings", or whenever a new interface pattern is introduced. Always use existing shadcn components and the radix-vega preset first — never hand-roll a control that shadcn already provides.
72
88%
Does it follow best practices?
Impact
—
No eval scenarios have been run
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 a strong skill description that clearly defines its scope (Stave's design system for UI changes), provides excellent trigger terms in both English and Korean, and includes explicit 'Use for/when' guidance. The implementation constraints (shadcn components, radix-vega preset) add further specificity and help distinguish it from generic UI skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and contexts: 'changes UI, layout, theme, dialogs, sidebars, empty states, settings, prompt input, or other visual UX'. Also specifies concrete implementation constraints like using shadcn components and radix-vega preset. | 3 / 3 |
Completeness | Clearly answers both 'what' (apply Stave's desktop-first design system for UI/layout/theme changes) and 'when' (explicit 'Use for prompts like...' clause with specific trigger terms and situational triggers like introducing new interface patterns). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including both English and Korean: '디자인', 'UI', 'redesign', 'polish', 'sidebar', 'dialog', 'settings'. Also mentions broader triggers like 'whenever a new interface pattern is introduced'. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — scoped to a specific repo's design system ('Stave's desktop-first design system'), specific tooling (shadcn, radix-vega preset), and specific UI domains. Unlikely to conflict with other skills due to the narrow, well-defined niche. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, project-specific design system skill that provides highly actionable guidance with concrete component mappings, CLI commands, file paths, and a clear implementation workflow with validation steps. Its main weakness is moderate verbosity — several rules are stated multiple times across different sections (component-first rule, no hand-rolling, shadcn preference), and some content could be extracted to referenced files to improve progressive disclosure. Overall it would serve Claude well in making correct UI decisions for this specific codebase.
Suggestions
Consolidate the repeated component-first/no-hand-rolling guidance — it appears in the Component-First Rule section, the Do/Don't lists, and the QA Checklist. State it definitively once and reference it elsewhere.
Consider extracting the detailed component inventory table and Glass/Depth guidelines into separate referenced files (e.g., `docs/ui/component-inventory.md`) to keep SKILL.md as a leaner overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly well-organized and most content is project-specific, but there's some redundancy — the 'Do/Don't' section largely restates rules already covered in earlier sections (e.g., the component-first rule is repeated in multiple places). The 'Mission' paragraph includes some generic design philosophy that Claude already understands. Could be tightened by ~20-30%. | 2 / 3 |
Actionability | Highly actionable: provides a concrete component inventory table, specific decision rules (e.g., 'Need a boolean toggle? → Use Switch'), an exact CLI command for generating missing components (`bunx --bun shadcn@latest add <component> --yes`), specific file paths to check, and a clear QA checklist. Guidance is specific to this repo's actual setup. | 3 / 3 |
Workflow Clarity | The Implementation Workflow section provides a clear 8-step sequence with validation checkpoints (check both themes, check at multiple widths, update docs). The component-first rule acts as a gate before any UI work begins. The QA Checklist provides explicit verification steps covering themes, keyboard access, barrel exports, and documentation updates. | 3 / 3 |
Progressive Disclosure | The skill references external files like `docs/ui/shadcn-preset.md`, `src/globals.css`, and `components.json` for deeper context, which is good. However, the SKILL.md itself is quite long (~180 lines of substantive content) with significant inline detail that could be split into referenced files (e.g., the full component inventory table, the glass/depth guidelines). No bundle files are provided to verify references, but the structure is reasonably organized with clear section headers. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
e275ad5
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.