Install any skill in seconds. Free to start, no credit card required.
Get Started Free →当用户需要在银行合规运营场景下,基于当日质检、投诉、整改与工单信息形成运营日报时使用本技能。
.claude/skills/aifinlab-bank-t235-compliance-operations-daily-report-assistant/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 47% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -5% | 0% |
本技能用于合规运营日报输出,强调信息归集、风险提示与行动项,适合用于对内汇报或日度跟踪。
scripts/daily_report_compiler.py:汇总质检、工单、整改数据并生成日报草稿。| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | pass→pass | 20,422 | 19,641 | -4% | 1 | 1 | 0% | 2,463 | 2,817 | +14% | 0 | 0 | — |
case-01 | fail→pass | 15,943 | 16,551 | +4% | 1 | 1 | 0% | 2,463 | 2,520 | +2% | 0 | 0 | — |
case-02 | fail→fail | 20,028 | 20,484 | +2% | 1 | 1 | 0% | 3,062 | 3,051 | -0% | 0 | 0 | — |
case-03 | pass→pass | 16,992 | 12,496 | -26% | 1 | 1 | 0% | 2,384 | 2,306 | -3% | 0 | 0 | — |
case-04 | pass→pass | 21,444 | 12,856 | -40% | 1 | 1 | 0% | 2,762 | 1,949 | -29% | 0 | 0 | — |
case-05 | fail→fail | 29,801 | 32,024 | +7% | 1 | 1 | 0% | 4,335 | 4,641 | +7% | 0 | 0 | — |
case-06 | pass→pass | 23,375 | 16,660 | -29% | 1 | 1 | 0% | 2,982 | 2,630 | -12% | 0 | 0 | — |
case-07 | pass→pass | 20,855 | 16,132 | -23% | 1 | 1 | 0% | 2,297 | 2,668 | +16% | 0 | 0 | — |
case-08 | fail→pass | 12,436 | 15,153 | +22% | 1 | 1 | 0% | 1,614 | 2,378 | +47% | 0 | 0 | — |
case-09 | pass→pass | 10,574 | 12,466 | +18% | 1 | 1 | 0% | 1,244 | 2,091 | +68% | 0 | 0 | — |
case-10 | pass→pass | 19,831 | 15,899 | -20% | 1 | 1 | 0% | 2,425 | 2,328 | -4% | 0 | 0 | — |
case-11 | fail→pass | 16,990 | 14,994 | -12% | 1 | 1 | 0% | 2,200 | 2,588 | +18% | 0 | 0 | — |
case-12 | pass→pass | 13,214 | 15,452 | +17% | 1 | 1 | 0% | 2,037 | 2,450 | +20% | 0 | 0 | — |
case-13 | fail→pass | 13,231 | 6,013 | -55% | 1 | 1 | 0% | 2,197 | 1,001 | -54% | 0 | 0 | — |
case-14 | pass→pass | 12,682 | 10,732 | -15% | 1 | 1 | 0% | 1,603 | 1,993 | +24% | 0 | 0 | — |
case-15 | pass→pass | 31,049 | 13,574 | -56% | 1 | 1 | 0% | 2,656 | 2,326 | -12% | 0 | 0 | — |
case-16 | fail→pass | 20,692 | 16,421 | -21% | 1 | 1 | 0% | 2,567 | 2,448 | -5% | 0 | 0 | — |
case-17 | pass→pass | 15,826 | 12,585 | -20% | 1 | 1 | 0% | 1,920 | 2,345 | +22% | 0 | 0 | — |
case-19 | pass→pass | 16,734 | 18,495 | +11% | 1 | 1 | 0% | 2,428 | 2,656 | +9% | 0 | 0 | — |
case-20 | pass→pass | 16,581 | 14,094 | -15% | 1 | 1 | 0% | 2,045 | 2,441 | +19% | 0 | 0 | — |
case-21 | fail→pass | 12,804 | 5,069 | -60% | 1 | 1 | 0% | 1,997 | 1,142 | -43% | 0 | 0 | — |
case-22 | fail→pass | 11,903 | 12,357 | +4% | 1 | 1 | 0% | 1,739 | 1,965 | +13% | 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 +32 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.