Content
65%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is a lean, well-structured overview with excellent progressive disclosure to three real reference files, and the HIG principles are appropriately terse. Weaknesses are the absence of an explicit workflow sequence, soft actionability (several directives lack concrete APIs, thresholds, or examples), and a boilerplate 'When to Use' section that adds pure token cost.
Suggestions
Make the operating sequence explicit (e.g., '1. Check .claude/apple-design-context.md 2. Ask unresolved questions 3. Consult the relevant reference 4. Produce output in the format below') to raise workflow clarity.
Add concrete specifics to soft directives — e.g., name the actual APIs (ProgressView, HUD, statusBarStyle/preferredStatusBarStyle) and give a duration threshold table instead of 'a second or two'.
Delete the boilerplate 'When to Use' section ('This skill is applicable to execute the workflow or actions described in the overview.'), which states nothing actionable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The 14 principles are one-line directives with minimal rationale and assume design-context knowledge (e.g., 'Don't stack multiple indicators. Aggregate simultaneous operations into one representation'), fitting 'Efficient; minor instances of over-explanation that could be trimmed' — the trimmable parts being the contentless 'When to Use' boilerplate ('This skill is applicable to execute the workflow or actions described in the overview.') and the footer. It is not level 5 because of that pure-padding section, and not level 3 because the core content has no unnecessary concept explanations. | 4 / 5 |
Actionability | Some guidance is concrete (ring colors 'Red (Move), green (Exercise), blue (Stand)', 'Use HealthKit APIs', 'Show progress for operations longer than a second or two'), but much is directive without specifics — 'Match status bar style to your content. Light or dark for adequate contrast' names no API or decision rule, and no concrete examples, thresholds, or code-level specifics appear, matching 'Some concrete guidance but incomplete; missing key details' rather than the mostly-executable level-4 anchor. | 3 / 5 |
Workflow Clarity | A rough sequence is implied (check .claude/apple-design-context.md, ask the four questions, apply principles, produce the four-part Output Format), but it is never presented as an ordered flow and has no validation checkpoints, matching 'Steps listed but validation gaps; sequence present but checkpoints missing or implicit'. It is not level 2 because the components of a flow do exist in writing, and not level 4 because the sequence must be inferred from scattered sections. | 3 / 5 |
Progressive Disclosure | The body is a genuine overview (distilled principles, output format, questions) with a well-signaled one-level-deep Reference Index table mapping each of references/progress-indicators.md, references/status-bars.md, and references/activity-rings.md to topic and key content — all three files exist in the bundle — matching 'Clear overview with well-signaled one-level-deep references; content appropriately split; easy navigation'. | 5 / 5 |
Total | 15 / 20 Passed |