Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Appium mobile testing framework for iOS and Android automation
.claude/skills/a5c-ai-appium-mobile-testing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -65% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 0% | 0% |
This skill provides expert-level capabilities for Appium-based mobile testing, enabling iOS and Android automation for native, hybrid, and web applications.
mobile-testing.js - Mobile test implementationcross-browser-testing.js - Mobile browser testinge2e-test-suite.js - Mobile E2E scenariosappium - Appium serverwebdriverio - WebDriver clientjavascript{ kind: 'skill', skill: { name: 'appium-mobile', context: { action: 'execute-tests', platform: 'iOS', deviceName: 'iPhone 14', app: './app/MyApp.ipa', automationName: 'XCUITest' } } }
The skill supports local emulators/simulators and cloud device farms for testing across multiple devices.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 5,944 | 12,215 | +106% | 1 | 1 | 0% | 1,349 | 1,206 | -11% | 0 | 0 | — |
case-02 | pass→pass | 10,269 | 4,526 | -56% | 1 | 1 | 0% | 1,877 | 1,240 | -34% | 0 | 0 | — |
case-03 | fail→pass | 5,848 | 3,809 | -35% | 1 | 1 | 0% | 1,021 | 659 | -35% | 0 | 0 | — |
case-04 | fail→pass | 11,937 | 1,749 | -85% | 1 | 1 | 0% | 2,097 | 737 | -65% | 0 | 0 | — |
case-05 | fail→pass | 9,014 | 1,587 | -82% | 1 | 1 | 0% | 1,614 | 690 | -57% | 0 | 0 | — |
case-06 | pass→pass | 5,478 | 1,822 | -67% | 1 | 1 | 0% | 863 | 726 | -16% | 0 | 0 | — |
case-07 | pass→pass | 7,127 | 2,243 | -69% | 1 | 1 | 0% | 1,124 | 845 | -25% | 0 | 0 | — |
case-08 | fail→fail | 8,407 | 9,165 | +9% | 1 | 1 | 0% | 1,267 | 1,912 | +51% | 0 | 0 | — |
case-09 | pass→pass | 10,463 | 6,379 | -39% | 1 | 1 | 0% | 1,731 | 1,404 | -19% | 0 | 0 | — |
case-10 | pass→pass | 9,862 | 3,251 | -67% | 1 | 1 | 0% | 1,773 | 990 | -44% | 0 | 0 | — |
case-11 | pass→pass | 10,247 | 8,771 | -14% | 1 | 1 | 0% | 1,745 | 2,072 | +19% | 0 | 0 | — |
case-12 | pass→pass | 19,075 | 19,242 | +1% | 1 | 1 | 0% | 3,227 | 3,841 | +19% | 0 | 0 | — |
case-13 | pass→pass | 9,530 | 4,223 | -56% | 1 | 1 | 0% | 1,585 | 1,248 | -21% | 0 | 0 | — |
case-14 | pass→pass | 11,638 | 9,846 | -15% | 1 | 1 | 0% | 1,779 | 2,433 | +37% | 0 | 0 | — |
case-15 | fail→pass | 8,806 | 6,796 | -23% | 1 | 1 | 0% | 1,476 | 1,482 | +0% | 0 | 0 | — |
case-16 | pass→pass | 13,846 | 7,178 | -48% | 1 | 1 | 0% | 1,916 | 1,760 | -8% | 0 | 0 | — |
case-17 | pass→pass | 7,137 | 5,898 | -17% | 1 | 1 | 0% | 1,385 | 1,551 | +12% | 0 | 0 | — |
case-18 | pass→pass | 3,645 | 1,588 | -56% | 1 | 1 | 0% | 601 | 691 | +15% | 0 | 0 | — |
case-19 | fail→pass | 6,577 | 2,424 | -63% | 1 | 1 | 0% | 1,240 | 948 | -24% | 0 | 0 | — |
case-20 | fail→fail | 14,460 | 9,162 | -37% | 1 | 1 | 0% | 2,449 | 2,195 | -10% | 0 | 0 | — |
case-21 | fail→fail | 14,017 | 15,138 | +8% | 1 | 1 | 0% | 2,586 | 3,168 | +23% | 0 | 0 | — |
case-22 | fail→fail | 16,315 | 18,195 | +12% | 1 | 1 | 0% | 3,078 | 3,324 | +8% | 0 | 0 | — |
case-23 | fail→fail | 8,409 | 4,375 | -48% | 1 | 1 | 0% | 1,314 | 1,173 | -11% | 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. 23 cases were attempted. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 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.