Content
67%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured design guidelines skill that efficiently organizes platform-specific guidance with a clear reference index and decision framework. Its main weakness is the lack of concrete, executable examples — no code snippets, no sample output, and no specific API references despite mentioning SwiftUI/UIKit/AppKit. The progressive disclosure structure is strong with the reference table, though bundle files couldn't be verified.
Suggestions
Add a concrete example showing the output format — e.g., a sample platform differences table and implementation notes for a real scenario like 'navigation for a music app on iOS vs macOS'.
Include at least one executable SwiftUI code snippet demonstrating a platform-adaptive pattern (e.g., NavigationSplitView on iPad vs TabView on iPhone).
Remove the vacuous 'When to Use' section — it adds no information and wastes tokens.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is mostly efficient and well-structured. The platform summaries are concise and informative without over-explaining. Minor verbosity in the 'Questions to Ask' and 'When to Use' sections, and the footer/branding line adds no value for Claude. The 'When to Use' section is particularly vacuous. | 4 / 5 |
Actionability | The skill provides a clear decision framework and structured output format, but lacks concrete executable examples. There are no code snippets, specific API calls, or copy-paste ready design specifications. The guidance is directional ('use sidebars and multi-column layouts') rather than specific (e.g., showing SwiftUI code for a sidebar navigation pattern). For a design guidelines skill, this is somewhat expected, but the output format section only describes what to produce without showing an example. | 3 / 5 |
Workflow Clarity | The decision framework provides a clear 5-step sequence for platform design decisions. The 'Questions to Ask' section establishes good information-gathering steps. However, there are no validation checkpoints or feedback loops — e.g., no step to verify designs against HIG compliance or test across platforms. Since this isn't a destructive/batch operation, the lack of validation doesn't cap the score, but the workflow could benefit from a review/verification step. | 4 / 5 |
Progressive Disclosure | Excellent reference index table with clear one-level-deep references to platform-specific files, well-signaled with topic and key content columns. Related skills are clearly listed. However, no bundle files were provided, so we cannot verify the referenced files exist. The skill also references `.claude/apple-design-context.md` without explaining its structure. Minor gap: the 'Related Skills' section references other skills but doesn't clarify their relationship format. | 4 / 5 |
Total | 15 / 20 Passed |