Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, well-organized design standards skill that covers the right topics with specific, measurable criteria. Its main weaknesses are moderate verbosity (explaining concepts Claude already understands, like visual hierarchy principles) and a workflow section that lacks explicit feedback loops for failure cases. The progressive disclosure is excellent with clear references to supporting files and sibling skills.
Suggestions
Tighten sections 3 (Visual Hierarchy) and 4 (Spacing and Rhythm) by removing explanations of basic design concepts Claude already knows — keep only the specific rules, values, and failure patterns.
Add explicit feedback loops to the workflow: e.g., 'If contrast check fails → adjust color token → re-check → update token file' rather than just listing 'Run contrast checks' as a step.
Consider adding a concrete example of a design review output (even a brief one) showing how the 6 standards map to actual review notes, making the output format section more actionable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is generally well-structured but includes some content Claude already knows (e.g., explaining what visual hierarchy is, what proximity principle means, explaining that contrast is 'not a preference'). The 'common failures' lists are useful but add significant length. Some sections like the hierarchy explanation could be tightened considerably. | 2 / 3 |
Actionability | The skill provides concrete values (specific pixel sizes, contrast ratios, spacing scales, viewport widths) which is good, but lacks executable code examples. For a stack-agnostic design skill this is somewhat justified, but the guidance remains at the level of rules and principles rather than copy-paste-ready implementations. The token definitions are specific but presented as lists rather than usable templates (the template is deferred to a reference file). | 2 / 3 |
Workflow Clarity | The 7-step workflow is clearly sequenced and includes validation steps (contrast checks, viewport testing, pre-ship checklist). However, there's no explicit feedback loop for when contrast checks fail or viewport testing reveals issues — it's implied but not stated as 'if X fails, do Y, then re-check.' For a skill involving potentially destructive design decisions, the lack of explicit error recovery caps this at 2. | 2 / 3 |
Progressive Disclosure | The skill has a clear overview structure with well-signaled one-level-deep references to three specific files (design-tokens-template.md, preship-checklist.md, tailwind-patterns.md). It also clearly delineates when to use companion skills (brand-identity, brand-style-guide, design-system). Navigation is straightforward and content is appropriately split between the main file and references. | 3 / 3 |
Total | 9 / 12 Passed |