Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Master TDD orchestrator specializing in red-green-refactor discipline, multi-agent workflow coordination, and comprehensive test-driven development practices. Enforces TDD best practices across teams with AI-assisted testing and modern frameworks. Use PROACTIVELY for TDD implementation and governance.
.claude/skills/dokhacgiakhoa-tdd-orchestrator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 162% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-09 | ✓→✗ | ▼ Worse | 7% | 0% |
| case-13 | ✓→✗ | ▼ Worse | -1% | 0% |
| case-21 | ✓→✗ | ▼ Worse | 29% | 0% |
resources/implementation-playbook.md.You are an expert TDD orchestrator specializing in comprehensive test-driven development coordination, modern TDD practices, and multi-agent workflow management.
Elite TDD orchestrator focused on enforcing disciplined test-driven development practices across complex software projects. Masters the complete red-green-refactor cycle, coordinates multi-agent TDD workflows, and ensures comprehensive test coverage while maintaining development velocity. Combines deep TDD expertise with modern AI-assisted testing tools to deliver robust, maintainable, and thoroughly tested software systems.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 28,111 | 25,271 | -10% | 1 | 1 | 0% | 4,019 | 4,034 | +0% | 0 | 0 | — |
case-02 | fail→fail | 29,377 | 18,039 | -39% | 1 | 1 | 0% | 4,157 | 2,759 | -34% | 0 | 0 | — |
case-03 | fail→fail | 20,465 | 20,143 | -2% | 1 | 1 | 0% | 3,482 | 3,963 | +14% | 0 | 0 | — |
case-04 | fail→fail | 23,936 | 14,471 | -40% | 1 | 1 | 0% | 3,907 | 3,828 | -2% | 0 | 0 | — |
case-05 | fail→fail | 23,080 | 26,388 | +14% | 1 | 1 | 0% | 3,255 | 4,276 | +31% | 0 | 0 | — |
case-06 | fail→fail | 21,534 | 20,877 | -3% | 1 | 1 | 0% | 2,721 | 2,987 | +10% | 0 | 0 | — |
case-07 | fail→pass | 43,258 | 25,652 | -41% | 1 | 1 | 0% | 1,575 | 4,129 | +162% | 0 | 0 | — |
case-08 | pass→pass | 26,611 | 26,683 | +0% | 1 | 1 | 0% | 4,067 | 4,851 | +19% | 0 | 0 | — |
case-09 | pass→fail | 20,786 | 19,077 | -8% | 1 | 1 | 0% | 2,678 | 2,868 | +7% | 0 | 0 | — |
case-10 | pass→pass | 23,606 | 20,693 | -12% | 1 | 1 | 0% | 3,439 | 3,729 | +8% | 0 | 0 | — |
case-11 | fail→fail | 24,890 | 21,052 | -15% | 1 | 1 | 0% | 3,449 | 3,202 | -7% | 0 | 0 | — |
case-12 | pass→pass | 23,413 | 25,317 | +8% | 1 | 1 | 0% | 3,068 | 4,246 | +38% | 0 | 0 | — |
case-13 | pass→fail | 24,022 | 20,562 | -14% | 1 | 1 | 0% | 3,424 | 3,387 | -1% | 0 | 0 | — |
case-14 | pass→pass | 19,248 | 26,690 | +39% | 1 | 1 | 0% | 3,034 | 4,657 | +53% | 0 | 0 | — |
case-15 | pass→pass | 31,339 | 35,977 | +15% | 1 | 1 | 0% | 4,612 | 6,245 | +35% | 0 | 0 | — |
case-16 | pass→pass | 21,979 | 23,112 | +5% | 1 | 1 | 0% | 2,738 | 3,955 | +44% | 0 | 0 | — |
case-17 | fail→fail | 28,789 | 27,105 | -6% | 1 | 1 | 0% | 5,936 | 4,818 | -19% | 0 | 0 | — |
case-18 | fail→fail | 25,546 | 20,582 | -19% | 1 | 1 | 0% | 3,538 | 3,986 | +13% | 0 | 0 | — |
case-19 | fail→fail | 24,409 | 26,784 | +10% | 1 | 1 | 0% | 3,887 | 6,252 | +61% | 0 | 0 | — |
case-20 | pass→pass | 18,491 | 21,792 | +18% | 1 | 1 | 0% | 3,158 | 3,538 | +12% | 0 | 0 | — |
case-21 | pass→fail | 23,164 | 20,421 | -12% | 1 | 1 | 0% | 3,162 | 4,089 | +29% | 0 | 0 | — |
case-22 | fail→pass | 21,007 | 26,206 | +25% | 1 | 1 | 0% | 4,451 | 5,237 | +18% | 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, and 21 counted toward the lift figure. The other 1 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 -20 percentage points is the difference between those two pass rates over the 21 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.