Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Cucumber/Gherkin BDD testing for behavior-driven development workflows
.claude/skills/a5c-ai-cucumber-bdd-testing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 44% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 84% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 17% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -10% | 0% |
This skill provides expert-level capabilities for Cucumber-based BDD testing, enabling feature file creation, step definition generation, and living documentation.
e2e-test-suite.js - BDD-style E2E teststest-strategy.js - BDD strategy implementationshift-left-testing.js - Early BDD adoption@cucumber/cucumber - Cucumber.jsjavascript{ kind: 'skill', skill: { name: 'cucumber-bdd', context: { action: 'execute-features', featurePath: 'features/**/*.feature', tags: '@smoke and not @wip', format: ['html:reports/cucumber.html', 'json:reports/cucumber.json'] } } }
The skill uses cucumber.js configuration file and supports multiple output formats.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 10,958 | 14,818 | +35% | 1 | 1 | 0% | 2,324 | 3,354 | +44% | 0 | 0 | — |
case-02 | pass→pass | 6,230 | 10,964 | +76% | 1 | 1 | 0% | 1,155 | 2,120 | +84% | 0 | 0 | — |
case-03 | pass→pass | 6,214 | 5,747 | -8% | 1 | 1 | 0% | 1,131 | 1,327 | +17% | 0 | 0 | — |
case-04 | pass→pass | 9,497 | 7,535 | -21% | 1 | 1 | 0% | 1,854 | 1,673 | -10% | 0 | 0 | — |
case-05 | pass→pass | 9,283 | 9,000 | -3% | 1 | 1 | 0% | 1,631 | 1,798 | +10% | 0 | 0 | — |
case-06 | pass→pass | 8,457 | 3,828 | -55% | 1 | 1 | 0% | 1,525 | 1,091 | -28% | 0 | 0 | — |
case-07 | pass→pass | 7,630 | 7,219 | -5% | 1 | 1 | 0% | 1,431 | 1,826 | +28% | 0 | 0 | — |
case-08 | pass→pass | 11,600 | 14,093 | +21% | 1 | 1 | 0% | 2,129 | 2,277 | +7% | 0 | 0 | — |
case-09 | pass→pass | 5,570 | 5,710 | +3% | 1 | 1 | 0% | 1,130 | 1,348 | +19% | 0 | 0 | — |
case-10 | pass→pass | 6,221 | 1,906 | -69% | 1 | 1 | 0% | 1,280 | 716 | -44% | 0 | 0 | — |
case-11 | pass→pass | 13,942 | 13,932 | -0% | 1 | 1 | 0% | 2,487 | 3,070 | +23% | 0 | 0 | — |
case-12 | pass→pass | 11,425 | 14,594 | +28% | 1 | 1 | 0% | 2,155 | 2,720 | +26% | 0 | 0 | — |
case-13 | pass→pass | 15,918 | 11,249 | -29% | 1 | 1 | 0% | 2,654 | 2,916 | +10% | 0 | 0 | — |
case-14 | pass→pass | 15,423 | 12,831 | -17% | 1 | 1 | 0% | 2,912 | 2,935 | +1% | 0 | 0 | — |
case-15 | pass→pass | 18,348 | 12,204 | -33% | 1 | 1 | 0% | 2,720 | 2,550 | -6% | 0 | 0 | — |
case-16 | pass→pass | 16,953 | 18,587 | +10% | 1 | 1 | 0% | 2,723 | 3,664 | +35% | 0 | 0 | — |
case-17 | fail→pass | 15,079 | 14,258 | -5% | 1 | 1 | 0% | 2,160 | 2,484 | +15% | 0 | 0 | — |
case-18 | pass→pass | 9,196 | 9,089 | -1% | 1 | 1 | 0% | 1,870 | 2,264 | +21% | 0 | 0 | — |
case-19 | pass→pass | 4,980 | 3,968 | -20% | 1 | 1 | 0% | 938 | 1,148 | +22% | 0 | 0 | — |
case-20 | pass→pass | 3,269 | 3,177 | -3% | 1 | 1 | 0% | 657 | 959 | +46% | 0 | 0 | — |
case-21 | pass→pass | 7,095 | 8,994 | +27% | 1 | 1 | 0% | 1,227 | 1,818 | +48% | 0 | 0 | — |
case-22 | pass→pass | 8,298 | 5,605 | -32% | 1 | 1 | 0% | 1,570 | 1,440 | -8% | 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 +5 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.