Content
82%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.
A strong, highly actionable skill body: real verified tool output, concrete fix guidance, clear modes and embedded validation. The main defects are a dangling reference to a nonexistent audit template and inline time-sensitive version/date enumeration that belongs in a reference or a changelog-style section.
Suggestions
Create templates/hig-audit-template.md (or remove the reference) — Mode 2 instructs 'fill in templates/hig-audit-template.md' but the file is absent from the bundle, leaving the audit workflow with a dead-end step.
Move the WWDC25 announcement date and the iOS/iPadOS/macOS/watchOS/tvOS/visionOS 26 version enumeration out of Core Design Principles into references/visual-design.md or a versioned-changelog section, keeping the body evergreen (the 'verify against the live HIG pages' note already covers drift).
Make the audit feedback loop explicit in Mode 2: apply fixes, re-run `hig_checker.py batch`, and confirm the score reaches the 90-100 ship band before delivering the report.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and avoids explaining concepts Claude already knows, but the Core Design Principles section inlines time-sensitive version enumeration ("announced at WWDC25, June 2025; shipped Sept 2025 across iOS 26, iPadOS 26, macOS Tahoe, watchOS 26, tvOS 26, visionOS 26") which the guidelines penalize when not placed in a deprecated/old-patterns section. This is the only real over-explanation, so it sits at 'minor instances that could be trimmed' rather than the mostly-inefficient level below. | 4 / 5 |
Actionability | Fully executable, copy-paste-ready guidance: the three documented subcommands run as written (verified: `hig_checker.py contrast "#8E8E93" "#FFFFFF"` -> "Contrast Ratio: 3.26 [FAILED]"; `batch audit.json` -> the exact worked-example JSON shown), plus a concrete input schema and specific fixes ("Use `.secondaryLabel` (semantic color) instead of hardcoded `#8E8E93`", "expand the hit region to 44x44 with padding/`contentShape`"). The worked example covers the common audit case end-to-end, matching the top anchor. | 5 / 5 |
Workflow Clarity | The two modes are clearly sequenced and validation is embedded in the tool (FAILED flags, scorecard thresholds "90-100 = ship, 70-80 = fix before release, below 70 = systematic rework") with a manual re-test checkpoint ("re-test caption contrast against the *busiest* underlying region and with Reduce Transparency on"). Not 5 because Mode 2's step "fill in `templates/hig-audit-template.md`" points to a file that does not exist in the bundle, and the fix -> re-run -> confirm loop is implied rather than explicit; the batch operation is read-only measurement, so the destructive-batch cap at 3 does not apply. | 4 / 5 |
Progressive Disclosure | Good structure with well-signaled, one-level-deep references that all exist and are topic-matched (`references/visual-design.md`, `references/accessibility.md`, `references/platform-specifics.md`, `scripts/hig_checker.py`), and reference files contain no nested references. Not 5 because the body references `templates/hig-audit-template.md` in Mode 2 but no templates/ directory exists in the bundle — a dangling path that breaks navigation for a core workflow component. | 4 / 5 |
Total | 17 / 20 Passed |