Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Router workflow for test automation: routes to ui-aqa-flow, api-aqa-flow, or testgen-flow. Kept for backward compatibility.
.claude/skills/griddynamics-aqa-flow/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -64% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -61% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 31% | 0% |
<aqa_flow>
<description_and_purpose>
Backward-compatible entry point for test-automation requests. The former monolithic AQA workflow was split into three specialized flows; this router classifies the request and dispatches to exactly one of them. It performs no phase work itself.
</description_and_purpose>
<routing>
Classify the user's request and route (invoke the target flow with the user's original request verbatim):
| Request is about… | Route | |---|---| | UI / browser / E2E test automation — page objects, selectors, UI test implementation or correction | USE FLOW ui-aqa-flow.md | | Backend API test automation — API contracts, Swagger/OpenAPI, request/response tests, API test implementation or correction | USE FLOW api-aqa-flow.md | | Generating test cases / requirements from tickets and docs (Jira/Confluence), exporting cases to a TMS — no test code | USE FLOW testgen-flow.md |
— You MUST fully execute loaded workflow following its entire definition for all request sizes, workflow WAS created to fix your failure modes (deviations, and weak process adherence, and shallow analysis), workflow is PRIMARY deterministic process to resolve the original user request
</routing>
</aqa_flow>
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 14,817 | 4,118 | -72% | 1 | 1 | 0% | 2,784 | 1,011 | -64% | 0 | 0 | — |
case-02 | fail→pass | 16,903 | 3,713 | -78% | 1 | 1 | 0% | 2,627 | 1,021 | -61% | 0 | 0 | — |
case-03 | fail→pass | 4,502 | 3,369 | -25% | 1 | 1 | 0% | 669 | 922 | +38% | 0 | 0 | — |
case-04 | fail→pass | 9,097 | 6,649 | -27% | 1 | 1 | 0% | 1,368 | 1,564 | +14% | 0 | 0 | — |
case-05 | fail→pass | 9,614 | 10,904 | +13% | 1 | 1 | 0% | 1,789 | 2,346 | +31% | 0 | 0 | — |
case-06 | fail→pass | 18,774 | 2,096 | -89% | 1 | 1 | 0% | 4,064 | 720 | -82% | 0 | 0 | — |
case-07 | fail→pass | 10,870 | 3,176 | -71% | 1 | 1 | 0% | 1,815 | 898 | -51% | 0 | 0 | — |
case-08 | fail→pass | 16,778 | 2,998 | -82% | 1 | 1 | 0% | 2,937 | 922 | -69% | 0 | 0 | — |
case-09 | fail→pass | 12,646 | 2,786 | -78% | 1 | 1 | 0% | 1,884 | 841 | -55% | 0 | 0 | — |
case-10 | pass→fail | 15,257 | 4,455 | -71% | 1 | 1 | 0% | 3,218 | 1,088 | -66% | 0 | 0 | — |
case-11 | pass→fail | 8,273 | 7,205 | -13% | 1 | 1 | 0% | 1,540 | 1,678 | +9% | 0 | 0 | — |
case-12 | pass→fail | 12,166 | 9,127 | -25% | 1 | 1 | 0% | 2,372 | 2,162 | -9% | 0 | 0 | — |
case-13 | pass→pass | 15,929 | 5,559 | -65% | 1 | 1 | 0% | 3,228 | 1,031 | -68% | 0 | 0 | — |
case-14 | fail→pass | 16,310 | 2,557 | -84% | 1 | 1 | 0% | 3,000 | 733 | -76% | 0 | 0 | — |
case-15 | pass→pass | 17,069 | 2,136 | -87% | 1 | 1 | 0% | 3,132 | 757 | -76% | 0 | 0 | — |
case-16 | fail→pass | 15,211 | 5,154 | -66% | 1 | 1 | 0% | 2,576 | 1,130 | -56% | 0 | 0 | — |
case-17 | fail→pass | 11,502 | 4,694 | -59% | 1 | 1 | 0% | 2,281 | 1,177 | -48% | 0 | 0 | — |
case-18 | fail→pass | 15,187 | 3,399 | -78% | 1 | 1 | 0% | 2,502 | 1,006 | -60% | 0 | 0 | — |
case-19 | fail→pass | 15,747 | 3,163 | -80% | 1 | 1 | 0% | 3,210 | 961 | -70% | 0 | 0 | — |
case-20 | pass→pass | 14,824 | 3,037 | -80% | 1 | 1 | 0% | 1,226 | 933 | -24% | 0 | 0 | — |
case-21 | fail→pass | 14,847 | 3,319 | -78% | 1 | 1 | 0% | 2,705 | 943 | -65% | 0 | 0 | — |
case-22 | fail→pass | 16,749 | 3,214 | -81% | 1 | 1 | 0% | 3,397 | 961 | -72% | 0 | 0 | — |
case-23 | fail→pass | 9,235 | 4,364 | -53% | 1 | 1 | 0% | 1,546 | 1,143 | -26% | 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 +61 percentage points is the difference between those two pass rates over the 23 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
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.