Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →当用户需要评估保额与保险责任是否匹配时使用此 skill。适用于保额充足性分析、责任覆盖范围检查、家庭保障需求测算等场景。
.claude/skills/aifinlab-coverage-liability-matching/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 78% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 28% | 0% |
你是一名经验丰富的保险规划师，目标是帮助用户科学测算保障需求，评估现有保额是否充足，并提供合理的保额配置建议。
围绕用户的家庭情况和保障需求，产出结构化、可量化、可执行的保额评估与配置方案。优先帮助用户解决以下任务：
当用户请求保额评估时，按以下顺序推进：
先从用户问题中识别以下信息：
如果用户没有给全，基于已有信息先做"默认假设"，并在开头明确写出假设。
默认优先使用以下框架：
输出时尽量覆盖：
所有结论尽量遵循：
适用于用户说"帮我看看保额够不够"。
按以下结构输出：
适用于用户说"帮我做家庭保障规划"。
按以下结构输出：
适用于用户说"帮我做全面的家庭保障方案"。
按以下结构输出：
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 18,032 | 19,285 | +7% | 1 | 1 | 0% | 2,984 | 3,945 | +32% | 0 | 0 | — |
case-02 | fail→fail | 23,948 | 25,362 | +6% | 1 | 1 | 0% | 4,198 | 5,324 | +27% | 0 | 0 | — |
case-03 | pass→pass | 16,580 | 19,445 | +17% | 1 | 1 | 0% | 2,767 | 4,144 | +50% | 0 | 0 | — |
case-04 | pass→pass | 17,065 | 16,326 | -4% | 1 | 1 | 0% | 2,823 | 3,584 | +27% | 0 | 0 | — |
case-05 | pass→pass | 14,664 | 17,347 | +18% | 1 | 1 | 0% | 2,316 | 3,919 | +69% | 0 | 0 | — |
case-06 | pass→pass | 16,345 | 19,475 | +19% | 1 | 1 | 0% | 2,625 | 4,160 | +58% | 0 | 0 | — |
case-07 | pass→pass | 18,069 | 17,349 | -4% | 1 | 1 | 0% | 2,798 | 3,762 | +34% | 0 | 0 | — |
case-08 | fail→fail | 13,282 | 15,999 | +20% | 1 | 1 | 0% | 2,064 | 3,716 | +80% | 0 | 0 | — |
case-09 | fail→pass | 20,963 | 21,178 | +1% | 1 | 1 | 0% | 3,395 | 4,764 | +40% | 0 | 0 | — |
case-10 | fail→pass | 27,298 | 30,202 | +11% | 1 | 1 | 0% | 4,284 | 6,045 | +41% | 0 | 0 | — |
case-11 | fail→fail | 17,929 | 17,861 | -0% | 1 | 1 | 0% | 2,734 | 3,776 | +38% | 0 | 0 | — |
case-12 | pass→pass | 17,377 | 19,850 | +14% | 1 | 1 | 0% | 2,664 | 3,944 | +48% | 0 | 0 | — |
case-13 | fail→pass | 11,289 | 13,712 | +21% | 1 | 1 | 0% | 1,828 | 3,250 | +78% | 0 | 0 | — |
case-14 | pass→pass | 14,818 | 15,167 | +2% | 1 | 1 | 0% | 2,296 | 3,484 | +52% | 0 | 0 | — |
case-15 | pass→pass | 5,051 | 10,242 | +103% | 1 | 1 | 0% | 938 | 2,718 | +190% | 0 | 0 | — |
case-16 | pass→pass | 17,110 | 22,603 | +32% | 1 | 1 | 0% | 2,829 | 4,907 | +73% | 0 | 0 | — |
case-17 | pass→pass | 18,357 | 19,731 | +7% | 1 | 1 | 0% | 3,150 | 4,287 | +36% | 0 | 0 | — |
case-18 | pass→pass | 10,684 | 16,604 | +55% | 1 | 1 | 0% | 1,812 | 3,588 | +98% | 0 | 0 | — |
case-19 | fail→pass | 13,299 | 10,294 | -23% | 1 | 1 | 0% | 2,228 | 2,849 | +28% | 0 | 0 | — |
case-20 | pass→pass | 11,092 | 16,874 | +52% | 1 | 1 | 0% | 2,051 | 3,768 | +84% | 0 | 0 | — |
case-21 | pass→pass | 26,849 | 28,624 | +7% | 1 | 1 | 0% | 3,918 | 5,409 | +38% | 0 | 0 | — |
case-22 | pass→pass | 20,652 | 19,377 | -6% | 1 | 1 | 0% | 3,784 | 4,428 | +17% | 0 | 0 | — |
case-23 | pass→pass | 18,758 | 19,301 | +3% | 1 | 1 | 0% | 3,176 | 3,977 | +25% | 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 +22 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.