Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate XCTest UI tests for macOS applications with accessibility identifiers and page object patterns
.claude/skills/a5c-ai-xctest-ui-test-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 25% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 34% | 0% |
Generate XCTest UI tests for macOS applications. This skill creates UI test classes with accessibility identifiers, page object patterns, and proper test organization.
json{ "type": "object", "properties": { "projectPath": { "type": "string" }, "targetViews": { "type": "array" }, "usePageObjects": { "type": "boolean", "default": true }, "generateAccessibilityIds": { "type": "boolean", "default": true } }, "required": ["projectPath"] }
swiftimport XCTest final class MainViewUITests: XCTestCase { var app: XCUIApplication! override func setUpWithError() throws { continueAfterFailure = false app = XCUIApplication() app.launchArguments = ["--uitesting"] app.launch() } func testMainViewLoads() throws { let mainView = app.windows["MainWindow"] XCTAssertTrue(mainView.waitForExistence(timeout: 5)) let titleLabel = mainView.staticTexts["welcomeLabel"] XCTAssertTrue(titleLabel.exists) XCTAssertEqual(titleLabel.label, "Welcome") } func testNavigationToSettings() throws { app.menuItems["Preferences…"].click() let settingsWindow = app.windows["SettingsWindow"] XCTAssertTrue(settingsWindow.waitForExistence(timeout: 2)) } }
swiftui-view-generatordesktop-ui-testing process| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 7,837 | 7,299 | -7% | 1 | 1 | 0% | 1,531 | 1,523 | -1% | 0 | 0 | — |
case-02 | fail→pass | 9,512 | 7,156 | -25% | 1 | 1 | 0% | 1,786 | 1,809 | +1% | 0 | 0 | — |
case-03 | pass→pass | 4,792 | 3,498 | -27% | 1 | 1 | 0% | 894 | 1,115 | +25% | 0 | 0 | — |
case-04 | fail→pass | 10,483 | 9,878 | -6% | 1 | 1 | 0% | 2,064 | 2,342 | +13% | 0 | 0 | — |
case-05 | pass→pass | 3,923 | 2,919 | -26% | 1 | 1 | 0% | 673 | 901 | +34% | 0 | 0 | — |
case-06 | pass→pass | 14,494 | 14,367 | -1% | 1 | 1 | 0% | 2,694 | 3,460 | +28% | 0 | 0 | — |
case-07 | pass→pass | 5,085 | 3,166 | -38% | 1 | 1 | 0% | 945 | 1,010 | +7% | 0 | 0 | — |
case-08 | pass→pass | 7,362 | 6,294 | -15% | 1 | 1 | 0% | 1,427 | 1,302 | -9% | 0 | 0 | — |
case-09 | pass→pass | 4,054 | 2,446 | -40% | 1 | 1 | 0% | 647 | 827 | +28% | 0 | 0 | — |
case-10 | pass→pass | 3,835 | 1,944 | -49% | 1 | 1 | 0% | 670 | 660 | -1% | 0 | 0 | — |
case-11 | pass→pass | 6,135 | 4,057 | -34% | 1 | 1 | 0% | 1,130 | 1,140 | +1% | 0 | 0 | — |
case-12 | pass→pass | 4,340 | 2,374 | -45% | 1 | 1 | 0% | 834 | 858 | +3% | 0 | 0 | — |
case-13 | pass→pass | 6,482 | 4,345 | -33% | 1 | 1 | 0% | 1,108 | 1,171 | +6% | 0 | 0 | — |
case-14 | pass→pass | 3,399 | 2,646 | -22% | 1 | 1 | 0% | 717 | 927 | +29% | 0 | 0 | — |
case-15 | pass→pass | 3,481 | 1,521 | -56% | 1 | 1 | 0% | 511 | 673 | +32% | 0 | 0 | — |
case-16 | fail→pass | 10,743 | 2,039 | -81% | 1 | 1 | 0% | 1,831 | 779 | -57% | 0 | 0 | — |
case-17 | pass→pass | 9,185 | 1,960 | -79% | 1 | 1 | 0% | 1,254 | 736 | -41% | 0 | 0 | — |
case-18 | pass→pass | 10,813 | 2,887 | -73% | 1 | 1 | 0% | 1,748 | 929 | -47% | 0 | 0 | — |
case-19 | pass→pass | 8,998 | 6,270 | -30% | 1 | 1 | 0% | 1,777 | 1,720 | -3% | 0 | 0 | — |
case-20 | pass→pass | 21,755 | 5,506 | -75% | 1 | 1 | 0% | 1,257 | 1,473 | +17% | 0 | 0 | — |
case-21 | pass→pass | 3,547 | 3,270 | -8% | 1 | 1 | 0% | 652 | 1,091 | +67% | 0 | 0 | — |
case-22 | pass→pass | 12,381 | 7,973 | -36% | 1 | 1 | 0% | 2,000 | 1,940 | -3% | 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.