Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Expert Terraform/OpenTofu specialist mastering advanced IaC automation, state management, and enterprise infrastructure patterns. Handles complex module design, multi-cloud deployments, GitOps workflows, policy as code, and CI/CD integration. Covers migration strategies, security best practices, and modern IaC ecosystems. Use PROACTIVELY for advanced IaC, state management, or infrastructure automation.
.claude/skills/dokhacgiakhoa-terraform-specialist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-16 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 58% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 16% | 0% |
| case-09 | ✓→✓ | ＝ Same ✓ | 26% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 37% | 0% |
You are a Terraform/OpenTofu specialist focused on advanced infrastructure automation, state management, and modern IaC practices.
Expert Infrastructure as Code specialist with comprehensive knowledge of Terraform, OpenTofu, and modern IaC ecosystems. Masters advanced module design, state management, provider development, and enterprise-scale infrastructure automation. Specializes in GitOps workflows, policy as code, and complex multi-cloud deployments.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 35,722 | 35,040 | -2% | 1 | 1 | 0% | 4,492 | 5,586 | +24% | 0 | 0 | — |
case-02 | fail→fail | 37,407 | 36,798 | -2% | 1 | 1 | 0% | 5,751 | 5,608 | -2% | 0 | 0 | — |
case-03 | pass→pass | 6,210 | 23,345 | +276% | 1 | 1 | 0% | 1,072 | 1,695 | +58% | 0 | 0 | — |
case-04 | pass→pass | 15,571 | 12,101 | -22% | 1 | 1 | 0% | 1,576 | 1,822 | +16% | 0 | 0 | — |
case-09 | pass→pass | 24,840 | 25,626 | +3% | 1 | 1 | 0% | 2,792 | 3,519 | +26% | 0 | 0 | — |
case-05 | pass→pass | 11,978 | 11,535 | -4% | 1 | 1 | 0% | 1,157 | 1,583 | +37% | 0 | 0 | — |
case-06 | pass→pass | 18,647 | 15,782 | -15% | 1 | 1 | 0% | 2,107 | 2,443 | +16% | 0 | 0 | — |
case-07 | pass→pass | 17,620 | 34,693 | +97% | 1 | 1 | 0% | 2,245 | 2,712 | +21% | 0 | 0 | — |
case-08 | pass→pass | 20,246 | 23,651 | +17% | 1 | 1 | 0% | 2,337 | 3,725 | +59% | 0 | 0 | — |
case-10 | pass→pass | 25,103 | 28,492 | +14% | 1 | 1 | 0% | 2,228 | 3,905 | +75% | 0 | 0 | — |
case-11 | pass→pass | 20,227 | 21,706 | +7% | 1 | 1 | 0% | 2,817 | 3,010 | +7% | 0 | 0 | — |
case-12 | pass→pass | 20,799 | 15,453 | -26% | 1 | 1 | 0% | 2,534 | 2,582 | +2% | 0 | 0 | — |
case-13 | pass→pass | 17,790 | 16,336 | -8% | 1 | 1 | 0% | 2,142 | 2,872 | +34% | 0 | 0 | — |
case-14 | pass→pass | 13,027 | 7,737 | -41% | 1 | 1 | 0% | 1,551 | 2,017 | +30% | 0 | 0 | — |
case-15 | pass→pass | 18,184 | 10,981 | -40% | 1 | 1 | 0% | 2,119 | 2,319 | +9% | 0 | 0 | — |
case-16 | fail→pass | 19,671 | 18,677 | -5% | 1 | 1 | 0% | 2,355 | 2,720 | +15% | 0 | 0 | — |
case-17 | pass→pass | 13,571 | 12,092 | -11% | 1 | 1 | 0% | 1,644 | 1,837 | +12% | 0 | 0 | — |
case-18 | pass→pass | 15,906 | 8,945 | -44% | 1 | 1 | 0% | 1,709 | 1,933 | +13% | 0 | 0 | — |
case-19 | pass→pass | 19,246 | 20,760 | +8% | 1 | 1 | 0% | 2,397 | 2,777 | +16% | 0 | 0 | — |
case-20 | pass→pass | 16,601 | 11,864 | -29% | 1 | 1 | 0% | 1,993 | 2,444 | +23% | 0 | 0 | — |
case-21 | pass→pass | 17,729 | 23,007 | +30% | 1 | 1 | 0% | 2,857 | 3,601 | +26% | 0 | 0 | — |
case-22 | pass→pass | 12,713 | 14,110 | +11% | 1 | 1 | 0% | 2,198 | 2,880 | +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. 22 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.