Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →当用户需要检查理赔材料是否齐全时使用此 skill。适用于理赔申请前材料预审、缺失材料提醒、材料规范性检查等场景。
.claude/skills/aifinlab-claims-material-check/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 68% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 7% | 0% |
| case-05 | ✓→✗ | ▼ Worse | 48% | 0% |
你是一名经验丰富的保险理赔专员，目标是帮助用户在提交理赔申请前全面检查材料完整性，提高理赔效率，减少因材料问题导致的延误。
围绕用户的理赔申请，产出结构化、可执行的材料检查清单，确保材料齐全、规范。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 | 17,182 | 20,010 | +16% | 1 | 1 | 0% | 2,633 | 3,263 | +24% | 0 | 0 | — |
case-02 | fail→fail | 26,205 | 24,054 | -8% | 1 | 1 | 0% | 3,900 | 4,560 | +17% | 0 | 0 | — |
case-03 | pass→fail | 16,721 | 13,659 | -18% | 1 | 1 | 0% | 3,024 | 3,235 | +7% | 0 | 0 | — |
case-04 | pass→pass | 23,186 | 23,437 | +1% | 1 | 1 | 0% | 3,679 | 4,442 | +21% | 0 | 0 | — |
case-05 | pass→fail | 18,263 | 22,217 | +22% | 1 | 1 | 0% | 2,759 | 4,085 | +48% | 0 | 0 | — |
case-06 | pass→pass | 12,633 | 12,535 | -1% | 1 | 1 | 0% | 1,835 | 2,541 | +38% | 0 | 0 | — |
case-07 | pass→pass | 17,591 | 15,856 | -10% | 1 | 1 | 0% | 2,755 | 3,273 | +19% | 0 | 0 | — |
case-08 | pass→pass | 17,087 | 16,099 | -6% | 1 | 1 | 0% | 2,675 | 3,141 | +17% | 0 | 0 | — |
case-09 | fail→pass | 13,303 | 17,500 | +32% | 1 | 1 | 0% | 2,089 | 3,509 | +68% | 0 | 0 | — |
case-10 | pass→pass | 16,276 | 15,949 | -2% | 1 | 1 | 0% | 2,391 | 3,079 | +29% | 0 | 0 | — |
case-11 | pass→pass | 24,802 | 23,153 | -7% | 1 | 1 | 0% | 3,719 | 4,361 | +17% | 0 | 0 | — |
case-12 | pass→pass | 15,334 | 14,402 | -6% | 1 | 1 | 0% | 2,359 | 3,047 | +29% | 0 | 0 | — |
case-13 | pass→pass | 18,643 | 16,529 | -11% | 1 | 1 | 0% | 2,700 | 3,310 | +23% | 0 | 0 | — |
case-14 | fail→fail | 12,969 | 14,228 | +10% | 1 | 1 | 0% | 1,950 | 2,805 | +44% | 0 | 0 | — |
case-15 | pass→pass | 16,401 | 19,847 | +21% | 1 | 1 | 0% | 2,467 | 3,817 | +55% | 0 | 0 | — |
case-16 | pass→pass | 14,952 | 16,939 | +13% | 1 | 1 | 0% | 2,164 | 3,185 | +47% | 0 | 0 | — |
case-17 | pass→pass | 17,078 | 17,536 | +3% | 1 | 1 | 0% | 2,445 | 3,324 | +36% | 0 | 0 | — |
case-18 | pass→pass | 14,769 | 37,649 | +155% | 1 | 1 | 0% | 2,399 | 3,447 | +44% | 0 | 0 | — |
case-19 | pass→pass | 15,898 | 18,214 | +15% | 1 | 1 | 0% | 2,551 | 3,587 | +41% | 0 | 0 | — |
case-20 | fail→fail | 16,561 | 14,628 | -12% | 1 | 1 | 0% | 2,513 | 3,133 | +25% | 0 | 0 | — |
case-21 | fail→pass | 12,883 | 15,648 | +21% | 1 | 1 | 0% | 2,050 | 2,908 | +42% | 0 | 0 | — |
case-22 | pass→pass | 23,783 | 22,596 | -5% | 1 | 1 | 0% | 3,413 | 4,180 | +22% | 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 +5 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 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.