Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →当用户需要进行保险欺诈风险筛查时使用此 skill。适用于投保欺诈识别、理赔欺诈预警、可疑行为分析等场景。
.claude/skills/aifinlab-anti-fraud-screening/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 0% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 0% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 32% | 0% |
你是一名经验丰富的保险反欺诈专家，目标是帮助用户识别和评估保险欺诈风险，提供科学的风险筛查和预警。
围绕投保/理赔场景，产出结构化、可操作的风险筛查报告。优先帮助用户解决以下任务：
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 30,010 | 28,487 | -5% | 1 | 1 | 0% | 4,021 | 4,027 | +0% | 0 | 0 | — |
case-02 | fail→pass | 24,697 | 17,340 | -30% | 1 | 1 | 0% | 2,928 | 2,918 | -0% | 0 | 0 | — |
case-03 | fail→fail | 28,749 | 25,128 | -13% | 1 | 1 | 0% | 5,002 | 4,257 | -15% | 0 | 0 | — |
case-04 | fail→fail | 20,870 | 24,231 | +16% | 1 | 1 | 0% | 3,026 | 4,154 | +37% | 0 | 0 | — |
case-05 | fail→fail | 17,898 | 21,256 | +19% | 1 | 1 | 0% | 2,641 | 3,611 | +37% | 0 | 0 | — |
case-06 | fail→pass | 18,772 | 14,993 | -20% | 1 | 1 | 0% | 2,380 | 2,846 | +20% | 0 | 0 | — |
case-07 | pass→pass | 20,638 | 21,578 | +5% | 1 | 1 | 0% | 2,930 | 3,898 | +33% | 0 | 0 | — |
case-08 | pass→pass | 36,965 | 28,403 | -23% | 1 | 1 | 0% | 4,533 | 4,800 | +6% | 0 | 0 | — |
case-09 | pass→pass | 14,110 | 19,804 | +40% | 1 | 1 | 0% | 1,833 | 2,185 | +19% | 0 | 0 | — |
case-10 | fail→pass | 16,216 | 15,221 | -6% | 1 | 1 | 0% | 2,301 | 2,800 | +22% | 0 | 0 | — |
case-11 | fail→pass | 16,313 | 16,592 | +2% | 1 | 1 | 0% | 2,391 | 3,163 | +32% | 0 | 0 | — |
case-12 | pass→pass | 17,153 | 16,423 | -4% | 1 | 1 | 0% | 2,581 | 2,982 | +16% | 0 | 0 | — |
case-13 | pass→pass | 13,461 | 17,685 | +31% | 1 | 1 | 0% | 1,953 | 2,794 | +43% | 0 | 0 | — |
case-14 | pass→pass | 22,492 | 21,381 | -5% | 1 | 1 | 0% | 3,113 | 3,406 | +9% | 0 | 0 | — |
case-15 | pass→pass | 20,215 | 17,769 | -12% | 1 | 1 | 0% | 2,654 | 3,189 | +20% | 0 | 0 | — |
case-16 | pass→pass | 18,413 | 14,772 | -20% | 1 | 1 | 0% | 2,364 | 2,889 | +22% | 0 | 0 | — |
case-17 | pass→pass | 19,520 | 19,224 | -2% | 1 | 1 | 0% | 2,564 | 3,527 | +38% | 0 | 0 | — |
case-18 | fail→pass | 15,569 | 23,034 | +48% | 1 | 1 | 0% | 2,032 | 3,704 | +82% | 0 | 0 | — |
case-19 | pass→pass | 22,321 | 22,934 | +3% | 1 | 1 | 0% | 2,696 | 3,432 | +27% | 0 | 0 | — |
case-20 | pass→pass | 25,694 | 23,604 | -8% | 1 | 1 | 0% | 3,160 | 3,332 | +5% | 0 | 0 | — |
case-21 | pass→pass | 16,119 | 21,422 | +33% | 1 | 1 | 0% | 2,366 | 3,322 | +40% | 0 | 0 | — |
case-22 | pass→pass | 24,608 | 17,610 | -28% | 1 | 1 | 0% | 2,971 | 3,208 | +8% | 0 | 0 | — |
case-23 | pass→pass | 25,059 | 22,598 | -10% | 1 | 1 | 0% | 3,266 | 3,912 | +20% | 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. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 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.