Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Reviews SwiftUI code for best practices on modern APIs, maintainability, and performance. This skill should be used when the user asks to review SwiftUI code, check for deprecated iOS/macOS APIs, validate data flow patterns, or audit accessibility compliance in Swift projects.
.claude/skills/fradser-swiftui-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 34% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 95% | 0% |
Review Swift and SwiftUI code for correctness, modern API usage, and adherence to project conventions. Report only genuine problems - do not nitpick or invent issues.
Review process:
references/api.md.references/views.md.references/data.md.references/navigation.md.references/design.md.references/accessibility.md.references/performance.md.references/swift.md.references/hygiene.md.references/clean-architecture.md.If doing a partial review, load only the relevant reference files.
CRITICAL: Load and apply every applicable reference file (references/*.md) before reporting findings. Rules marked MUST or Never in those files are non-negotiable — flag every violation found, and skip a file only when it has no issues.
Organize findings by file. For each issue:
foregroundStyle() instead of foregroundColor()").Skip files with no issues. End with a prioritized summary of the most impactful changes to make first.
Example output:
Line 12: Use foregroundStyle() instead of foregroundColor().
swift// Before Text("Hello").foregroundColor(.red) // After Text("Hello").foregroundStyle(.red)
Line 24: Icon-only button is bad for VoiceOver - add a text label.
swift// Before Button(action: addUser) { Image(systemName: "plus") } // After Button("Add User", systemImage: "plus", action: addUser)
Line 31: Avoid Binding(get:set:) in view body - use @State with onChange() instead.
swift// Before TextField("Username", text: Binding( get: { model.username }, set: { model.username = $0; model.save() } )) // After TextField("Username", text: $model.username) .onChange(of: model.username) { model.save() }
foregroundColor() on line 12 should be foregroundStyle().End of example.
references/accessibility.md - Dynamic Type, VoiceOver, Reduce Motion, and other accessibility requirements.references/api.md - updating code for modern API, and the deprecated code it replaces.references/clean-architecture.md - Clean Architecture patterns, layer separation, MVVM, dependency injection, and modern SwiftUI architecture.references/design.md - guidance for building accessible apps that meet Apple’s Human Interface Guidelines.references/hygiene.md - making code compile cleanly and be maintainable in the long term.references/navigation.md - navigation using NavigationStack/NavigationSplitView, plus alerts, confirmation dialogs, and sheets.references/performance.md - optimizing SwiftUI code for maximum performance.references/data.md - data flow, shared state, and property wrappers.references/swift.md - tips on writing modern Swift code, including using Swift Concurrency effectively.references/views.md - view structure, composition, and animation.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 19,420 | 23,989 | +24% | 1 | 1 | 0% | 3,356 | 2,824 | -16% | 0 | 0 | — |
case-02 | pass→pass | 13,648 | 15,240 | +12% | 1 | 1 | 0% | 2,843 | 3,796 | +34% | 0 | 0 | — |
case-03 | pass→pass | 11,334 | 20,578 | +82% | 1 | 1 | 0% | 2,390 | 4,654 | +95% | 0 | 0 | — |
case-04 | pass→pass | 12,332 | 12,916 | +5% | 1 | 1 | 0% | 2,256 | 3,361 | +49% | 0 | 0 | — |
case-05 | fail→pass | 14,378 | 7,984 | -44% | 1 | 1 | 0% | 2,482 | 2,687 | +8% | 0 | 0 | — |
case-22 | fail→fail | 6,195 | 7,311 | +18% | 1 | 1 | 0% | 1,131 | 2,186 | +93% | 0 | 0 | — |
case-06 | pass→pass | 5,459 | 4,171 | -24% | 1 | 1 | 0% | 1,061 | 1,858 | +75% | 0 | 0 | — |
case-07 | pass→pass | 13,087 | 11,553 | -12% | 1 | 1 | 0% | 2,322 | 3,300 | +42% | 0 | 0 | — |
case-08 | fail→pass | 11,832 | 13,996 | +18% | 1 | 1 | 0% | 2,298 | 3,435 | +49% | 0 | 0 | — |
case-09 | pass→pass | 10,203 | 9,055 | -11% | 1 | 1 | 0% | 2,068 | 2,690 | +30% | 0 | 0 | — |
case-10 | pass→pass | 9,825 | 8,292 | -16% | 1 | 1 | 0% | 1,922 | 2,782 | +45% | 0 | 0 | — |
case-11 | pass→pass | 12,530 | 11,185 | -11% | 1 | 1 | 0% | 2,067 | 2,883 | +39% | 0 | 0 | — |
case-12 | pass→pass | 13,638 | 10,733 | -21% | 1 | 1 | 0% | 2,406 | 2,970 | +23% | 0 | 0 | — |
case-13 | pass→pass | 15,785 | 13,999 | -11% | 1 | 1 | 0% | 2,879 | 3,806 | +32% | 0 | 0 | — |
case-14 | pass→pass | 4,559 | 4,246 | -7% | 1 | 1 | 0% | 866 | 1,901 | +120% | 0 | 0 | — |
case-15 | pass→pass | 13,469 | 12,741 | -5% | 1 | 1 | 0% | 2,395 | 3,538 | +48% | 0 | 0 | — |
case-16 | pass→pass | 17,375 | 6,367 | -63% | 1 | 1 | 0% | 1,963 | 2,147 | +9% | 0 | 0 | — |
case-17 | pass→pass | 9,854 | 7,751 | -21% | 1 | 1 | 0% | 1,932 | 2,722 | +41% | 0 | 0 | — |
case-18 | pass→pass | 4,112 | 6,407 | +56% | 1 | 1 | 0% | 832 | 2,282 | +174% | 0 | 0 | — |
case-19 | pass→pass | 9,134 | 10,656 | +17% | 1 | 1 | 0% | 1,811 | 2,660 | +47% | 0 | 0 | — |
case-20 | pass→pass | 9,066 | 7,223 | -20% | 1 | 1 | 0% | 1,775 | 2,404 | +35% | 0 | 0 | — |
case-21 | pass→pass | 14,530 | 6,862 | -53% | 1 | 1 | 0% | 2,485 | 2,501 | +1% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.