Final visual polish for an existing UI without redesigning it. Use after structure is clear, when asked to improve spacing, alignment, text hierarchy, readability, shadows, highlights, effects, action placement, or overall production feel. For screen structure, layer naming, code handoff, preview scenes, or animation planning, use pencil-ui-structure first.
68
81%
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 an excellent skill description that clearly defines its scope as visual polish for existing UIs, provides rich natural trigger terms covering spacing, alignment, shadows, and more, and explicitly delineates boundaries with a related skill. The description is concise yet comprehensive, uses third person voice appropriately, and would enable Claude to confidently select this skill in the right context.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: improve spacing, alignment, text hierarchy, readability, shadows, highlights, effects, action placement, and overall production feel. Also clearly delineates what it does NOT do by referencing the sister skill. | 3 / 3 |
Completeness | Clearly answers both 'what' (final visual polish for existing UI) and 'when' (use after structure is clear, when asked to improve specific visual properties). Also includes explicit negative triggers directing to another skill, which strengthens the 'when' guidance. | 3 / 3 |
Trigger Term Quality | Includes highly natural trigger terms users would say: 'spacing', 'alignment', 'text hierarchy', 'readability', 'shadows', 'highlights', 'effects', 'action placement', 'production feel', 'visual polish'. These are terms designers and developers naturally use when requesting UI refinement. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive by explicitly scoping itself to visual polish only (not structure), and directly referencing the complementary 'pencil-ui-structure' skill with clear boundary conditions. This makes it very unlikely to conflict with related UI skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
62%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-thought-out skill with strong workflow clarity and specific numeric guidance for visual polish operations. Its main weaknesses are verbosity from repeated 'don't redesign' constraints across multiple sections, and the lack of executable code/command examples despite being a tool-oriented skill. The content would benefit from consolidating redundant constraint language and splitting detailed reference material (presets, effects) into separate files.
Suggestions
Consolidate the repeated 'don't redesign' / 'preserve existing design' constraints into a single authoritative section rather than restating them across Goals, Non-Goals, Default Stance, Decision Rules, Anti-Patterns, and Final Verification.
Add concrete executable examples — e.g., a before/after node property diff, a specific Pencil tool command sequence, or a CSS snippet showing a shadow fix — to increase actionability from descriptive to copy-paste ready.
Extract the Pencil Starting Presets and Effects Best Practices into separate reference files (e.g., EFFECTS.md, PENCIL-PRESETS.md) and link to them from the main skill to improve progressive disclosure and reduce the monolithic length.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is thorough and mostly well-organized, but it's quite verbose at ~300+ lines. Several sections repeat the same core principle ('don't redesign', 'preserve existing design') across Goals, Non-Goals, Default Stance, Polish Decision Rules, Anti-Patterns, and Final Verification. The Pencil Starting Presets section includes detailed numeric values that earn their place, but the repeated constraint language could be significantly tightened. | 2 / 3 |
Actionability | The skill provides concrete heuristics, specific numeric ranges for shadows/highlights/opacity, and clear decision rules. However, it lacks executable code examples or copy-paste-ready commands. The Pencil workflow steps are procedural but not tied to specific tool commands or API calls. The guidance is specific enough to act on but falls short of fully executable. | 2 / 3 |
Workflow Clarity | The Review Order provides a clear 9-step sequence. The Polish Decision Rules establish an explicit priority order (1-6). The Pencil Workflow includes before/after screenshot validation checkpoints and a concrete verification question. The Final Verification section serves as a comprehensive checklist. The Change Budget system (Tiny/Normal/System) provides clear scope control with feedback guidance. | 3 / 3 |
Progressive Disclosure | The content is well-structured with clear headers and logical sections, but it's entirely monolithic — all content lives in one file with no references to supporting documents. The Effects Best Practices, Pencil Starting Presets, and Anti-Patterns sections could reasonably be split into separate reference files. However, since no bundle files exist, there's no opportunity for external references, and the internal organization is decent. | 2 / 3 |
Total | 9 / 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.
64f57c5
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.