Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →QA cycling workflow — repeat build/lint/test/fix cycles up to 5 times until all checks pass. Stops early if the same error repeats 3 times (signals a fundamental issue, not a fixable one).
.claude/skills/evolution-foundation-dev-ultraqa/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-23 | ✗→✓ | ▲ Improved | -47% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 12% | 0% |
Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.
QA cycling: keep running build/lint/test/fix loops until all checks pass — but with a circuit breaker. If the same error persists across 3 cycles, stop and report: it's a fundamental issue, not a fixable one.
dev-autopilot or @bolt-executor)dev-verify once@apex-architect, don't loopnpm run build, cargo build, go build)npm run lint, cargo clippy, golangci-lint)npm test, cargo test, pytest)@oath-verifier for final verification@bolt-executor or @hawk-debugger to fix the failureIf the same error fingerprint (file:line + error message) appears in 3 consecutive cycles:
@apex-architect with full context dumpSaved to workspace/development/verifications/[C]ultraqa-{component}-{date}.md:
markdown## UltraQA Report ### Cycles 1. Build ❌ → fixed by @bolt-executor 2. Build ✅ Lint ❌ → fixed by @bolt-executor 3. Build ✅ Lint ✅ Tests ❌ → fixed by @hawk-debugger 4. ✅ All green ### Final State - Build: ✅ - Lint: ✅ - Tests: ✅ N passed, 0 failed ### Handoff - → @oath-verifier for evidence-based verification
@bolt-executor (fixes during cycles)@hawk-debugger (fixes when bugs are non-trivial)@oath-verifier (final verification post-cycle)@apex-architect (escalation when 3-failure breaker trips)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | fail→pass | 10,275 | 7,777 | -24% | 1 | 1 | 0% | 1,771 | 1,924 | +9% | 0 | 0 | — |
case-23 | fail→pass | 14,763 | 3,906 | -74% | 1 | 1 | 0% | 2,238 | 1,181 | -47% | 0 | 0 | — |
case-01 | fail→fail | 16,655 | 6,075 | -64% | 1 | 1 | 0% | 2,499 | 1,015 | -59% | 0 | 0 | — |
case-02 | fail→fail | 16,183 | 4,686 | -71% | 1 | 1 | 0% | 3,045 | 911 | -70% | 0 | 0 | — |
case-03 | fail→fail | 4,725 | 5,548 | +17% | 1 | 1 | 0% | 196 | 910 | +364% | 0 | 0 | — |
case-04 | fail→pass | 8,833 | 3,113 | -65% | 1 | 1 | 0% | 1,353 | 1,086 | -20% | 0 | 0 | — |
case-05 | fail→pass | 8,954 | 4,649 | -48% | 1 | 1 | 0% | 1,366 | 1,382 | +1% | 0 | 0 | — |
case-06 | fail→pass | 6,013 | 2,463 | -59% | 1 | 1 | 0% | 897 | 1,003 | +12% | 0 | 0 | — |
case-07 | fail→pass | 7,981 | 3,614 | -55% | 1 | 1 | 0% | 1,273 | 1,256 | -1% | 0 | 0 | — |
case-08 | fail→pass | 8,558 | 4,767 | -44% | 1 | 1 | 0% | 1,480 | 1,458 | -1% | 0 | 0 | — |
case-10 | fail→pass | 5,041 | 2,121 | -58% | 1 | 1 | 0% | 790 | 899 | +14% | 0 | 0 | — |
case-11 | fail→pass | 7,496 | 3,405 | -55% | 1 | 1 | 0% | 1,026 | 1,159 | +13% | 0 | 0 | — |
case-12 | fail→pass | 30,703 | 3,523 | -89% | 1 | 1 | 0% | 1,106 | 1,163 | +5% | 0 | 0 | — |
case-13 | fail→pass | 8,417 | 2,018 | -76% | 1 | 1 | 0% | 1,426 | 912 | -36% | 0 | 0 | — |
case-14 | fail→fail | 7,792 | 2,167 | -72% | 1 | 1 | 0% | 1,343 | 983 | -27% | 0 | 0 | — |
case-15 | fail→pass | 7,992 | 2,290 | -71% | 1 | 1 | 0% | 1,200 | 951 | -21% | 0 | 0 | — |
case-16 | pass→pass | 9,878 | 6,495 | -34% | 1 | 1 | 0% | 1,593 | 1,704 | +7% | 0 | 0 | — |
case-17 | pass→pass | 5,803 | 3,735 | -36% | 1 | 1 | 0% | 966 | 1,210 | +25% | 0 | 0 | — |
case-18 | pass→pass | 11,052 | 7,718 | -30% | 1 | 1 | 0% | 1,807 | 1,853 | +3% | 0 | 0 | — |
case-19 | pass→pass | 10,841 | 2,403 | -78% | 1 | 1 | 0% | 1,945 | 1,042 | -46% | 0 | 0 | — |
case-20 | fail→pass | 12,663 | 6,503 | -49% | 1 | 1 | 0% | 2,025 | 1,805 | -11% | 0 | 0 | — |
case-21 | fail→pass | 5,380 | 4,272 | -21% | 1 | 1 | 0% | 798 | 1,355 | +70% | 0 | 0 | — |
case-22 | fail→fail | 11,166 | 3,382 | -70% | 1 | 1 | 0% | 1,660 | 1,147 | -31% | 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, and 20 counted toward the lift figure. The other 3 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +61 percentage points is the difference between those two pass rates over the 20 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.