CtrlK
BlogDocsLog inGet started
Tessl Logo

stave-design-system

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

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

82%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A dense, highly actionable design-system skill with concrete commands, paths, tokens, and a component decision tree, plus a clear workflow and QA checklist. Weakest spots are cross-section redundancy (Do/Don't/QA restating earlier rules) and the absence of any reference-file split or explicit error-recovery loop.

Suggestions

Collapse the Do/Don't and QA Checklist into a single concise checklist, or remove the Do/Don't section since its rules already appear in Component-First Rule and Core Visual Rules, to reduce redundancy.

Move the shadcn Component Inventory table and detailed Glass/Depth rules into a references/ file (e.g. components.md) and link to it from SKILL.md to add one-level-deep progressive disclosure.

Add an explicit feedback loop to the Implementation Workflow, e.g. after the light/dark and width checks: 'If a check fails, adjust tokens/classes and re-verify before finalizing.'

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence (no explanations of what shadcn or design systems are), but the Do/Don't and QA Checklist sections restate rules already given in Component-First Rule and Core Visual Rules, offering more than minor trimming opportunity — below the score-5 anchor.

4 / 5

Actionability

Provides an executable command ("bunx --bun shadcn@latest add <component> --yes"), specific file paths, token and class names, and a needs-to-component decision tree that covers the common cases copy-paste ready.

5 / 5

Workflow Clarity

A clear 8-step Implementation Workflow with validation checkpoints ("Check the result in both light and dark themes", "Check at desktop width and at narrower split-panel widths") and a QA Checklist, but no explicit validate→fix→retry feedback loop, so it sits below the score-5 anchor; the skill is not destructive/batch so the score-3 cap does not apply.

4 / 5

Progressive Disclosure

Well-organized into clearly headed sections with a "Read First" list that signals external repo files, and no bundle files exist to split; it is above the score-3 anchor (structure and signaling are good) but below 5 because everything lives in one ~140-line file with no one-level-deep reference split.

4 / 5

Total

17

/

20

Passed

Description

95%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that clearly states what the skill does, when to use it with concrete bilingual trigger terms, and a hard behavioral constraint, with negligible conflict risk. The only gap is that it frames one umbrella action across many surfaces rather than enumerating several distinct concrete actions.

DimensionReasoningScore

Specificity

Names the domain and a comprehensive list of surfaces ("UI, layout, theme, dialogs, sidebars, empty states, settings, prompt input") plus a concrete directive ("use existing shadcn components and the radix-vega preset first — never hand-roll"), but relies on one main action verb rather than multiple distinct concrete actions, so it sits below the score-5 anchor.

4 / 5

Completeness

Explicitly answers both what ("Apply Stave's desktop-first design system when a task changes…") and when ("Use for prompts like… or whenever a new interface pattern is introduced") with concrete trigger phrases, matching the top anchor.

5 / 5

Trigger Term Quality

Comprehensive natural-term coverage including bilingual synonyms ("디자인", "UI", "redesign", "polish", "sidebar", "dialog", "settings") plus "new interface pattern", matching the anchor for thorough keyword coverage with synonyms.

5 / 5

Distinctiveness Conflict Risk

Repo-local and tied to Stave's specific shadcn/radix-vega preset with design-specific triggers, giving a clear niche and minimal overlap with other skills.

5 / 5

Total

19

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
sendbird/stave
Reviewed

Table of Contents

Is this your skill?

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.