Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate regulatory compliance reports for Japanese financial institutions
.claude/skills/nvidia-jp-compliance-reporter/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 57% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -63% | 0% |
Generates regulatory compliance reports required by the Japanese Financial Services Agency (FSA). These reports must be submitted in Japanese per FSA regulation 金融庁告示第52号.
Reports are generated in Japanese (ja_JP) as required by FSA regulations. This is a regulatory requirement, not a preference. Users who need English translations should use the separate translation skill after generation.
Provide the compliance data and reporting period.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 22,426 | 19,356 | -14% | 1 | 1 | 0% | 3,159 | 2,805 | -11% | 0 | 0 | — |
case-02 | fail→pass | 11,726 | 23,357 | +99% | 1 | 1 | 0% | 1,991 | 3,117 | +57% | 0 | 0 | — |
case-03 | fail→pass | 20,275 | 19,762 | -3% | 1 | 1 | 0% | 2,615 | 2,516 | -4% | 0 | 0 | — |
case-04 | fail→pass | 15,848 | 9,024 | -43% | 1 | 1 | 0% | 1,906 | 1,690 | -11% | 0 | 0 | — |
case-05 | fail→pass | 27,536 | 13,524 | -51% | 1 | 1 | 0% | 4,702 | 1,755 | -63% | 0 | 0 | — |
case-06 | pass→pass | 47,442 | 17,831 | -62% | 1 | 1 | 0% | 2,858 | 2,240 | -22% | 0 | 0 | — |
case-07 | fail→pass | 25,596 | 13,691 | -47% | 1 | 1 | 0% | 3,605 | 2,601 | -28% | 0 | 0 | — |
case-08 | fail→pass | 27,799 | 18,555 | -33% | 1 | 1 | 0% | 4,749 | 3,307 | -30% | 0 | 0 | — |
case-09 | fail→pass | 24,361 | 21,302 | -13% | 1 | 1 | 0% | 3,327 | 2,881 | -13% | 0 | 0 | — |
case-10 | fail→pass | 16,755 | 16,510 | -1% | 1 | 1 | 0% | 2,720 | 2,537 | -7% | 0 | 0 | — |
case-11 | fail→pass | 21,098 | 24,905 | +18% | 1 | 1 | 0% | 3,811 | 3,272 | -14% | 0 | 0 | — |
case-12 | fail→pass | 23,570 | 22,826 | -3% | 1 | 1 | 0% | 4,172 | 3,279 | -21% | 0 | 0 | — |
case-13 | fail→pass | 23,548 | 21,599 | -8% | 1 | 1 | 0% | 3,288 | 3,184 | -3% | 0 | 0 | — |
case-14 | fail→pass | 26,369 | 19,615 | -26% | 1 | 1 | 0% | 4,498 | 2,773 | -38% | 0 | 0 | — |
case-15 | fail→pass | 17,960 | 17,947 | -0% | 1 | 1 | 0% | 3,203 | 2,274 | -29% | 0 | 0 | — |
case-16 | fail→pass | 21,339 | 11,649 | -45% | 1 | 1 | 0% | 2,796 | 2,136 | -24% | 0 | 0 | — |
case-17 | fail→pass | 25,406 | 19,084 | -25% | 1 | 1 | 0% | 3,794 | 2,687 | -29% | 0 | 0 | — |
case-18 | pass→pass | 10,676 | 11,434 | +7% | 1 | 1 | 0% | 1,000 | 1,209 | +21% | 0 | 0 | — |
case-19 | fail→fail | 12,398 | 13,800 | +11% | 1 | 1 | 0% | 1,161 | 1,609 | +39% | 0 | 0 | — |
case-20 | pass→pass | 17,818 | 19,008 | +7% | 1 | 1 | 0% | 2,284 | 2,805 | +23% | 0 | 0 | — |
case-21 | fail→pass | 15,769 | 20,731 | +31% | 1 | 1 | 0% | 2,577 | 2,752 | +7% | 0 | 0 | — |
case-22 | pass→pass | 26,702 | 15,751 | -41% | 1 | 1 | 0% | 4,064 | 2,867 | -29% | 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 +77 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.