Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →整合核心系统、信贷系统、电子渠道等多源数据，生成客户360度全景画像，为客户经营决策提供基础信息底座。
.claude/skills/aifinlab-bank-customer-360/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 46% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 24% | 0% |
银行客户信息分散在核心系统、信贷系统、理财系统、电子渠道等多个业务系统中，客户经理难以快速获取客户全貌。需要将多系统数据整合为结构化的客户360度画像，支撑客户经营、营销触达和风险识别等工作。
| 输入项 | 说明 | 是否必填 | |--------|------|----------| | 客户编号/客户列表 | 单个或批量客户标识 | 必填 | | 基本信息 | 姓名、年龄、职业、地区、客户等级 | 必填 | | 资产信息 | AUM、存款余额、理财持仓、基金持仓 | 必填 | | 负债信息 | 贷款余额、信用卡额度及使用情况 | 选填 | | 交易信息 | 近期交易频次、交易金额、主要交易类型 | 选填 | | 渠道行为 | 手机银行登录频次、网银使用情况 | 选填 | | 产品持有 | 当前持有产品清单及到期日 | 选填 | | 服务记录 | 近期服务工单、投诉记录、满意度评分 | 选填 |
无（数据入口技能，直接对接银行各业务系统数据）
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | fail→pass | 12,798 | 17,164 | +34% | 1 | 1 | 0% | 2,066 | 3,446 | +67% | 0 | 0 | — |
case-09 | pass→pass | 17,587 | 14,550 | -17% | 1 | 1 | 0% | 2,566 | 3,337 | +30% | 0 | 0 | — |
case-01 | fail→pass | 19,569 | 33,510 | +71% | 1 | 1 | 0% | 2,868 | 4,184 | +46% | 0 | 0 | — |
case-02 | fail→pass | 16,477 | 16,856 | +2% | 1 | 1 | 0% | 2,551 | 3,663 | +44% | 0 | 0 | — |
case-03 | fail→pass | 15,878 | 14,172 | -11% | 1 | 1 | 0% | 2,103 | 2,905 | +38% | 0 | 0 | — |
case-04 | pass→pass | 16,990 | 13,832 | -19% | 1 | 1 | 0% | 2,217 | 3,087 | +39% | 0 | 0 | — |
case-05 | fail→pass | 19,821 | 16,378 | -17% | 1 | 1 | 0% | 2,692 | 3,325 | +24% | 0 | 0 | — |
case-06 | pass→pass | 9,269 | 14,717 | +59% | 1 | 1 | 0% | 1,652 | 3,514 | +113% | 0 | 0 | — |
case-07 | fail→pass | 16,310 | 14,878 | -9% | 1 | 1 | 0% | 2,375 | 3,451 | +45% | 0 | 0 | — |
case-15 | fail→pass | 14,661 | 14,021 | -4% | 1 | 1 | 0% | 2,415 | 3,571 | +48% | 0 | 0 | — |
case-10 | pass→pass | 17,692 | 19,516 | +10% | 1 | 1 | 0% | 2,569 | 3,637 | +42% | 0 | 0 | — |
case-11 | fail→pass | 18,419 | 20,172 | +10% | 1 | 1 | 0% | 2,671 | 4,099 | +53% | 0 | 0 | — |
case-12 | pass→pass | 18,794 | 21,982 | +17% | 1 | 1 | 0% | 2,660 | 3,483 | +31% | 0 | 0 | — |
case-13 | fail→pass | 13,256 | 17,497 | +32% | 1 | 1 | 0% | 2,005 | 3,684 | +84% | 0 | 0 | — |
case-14 | pass→pass | 13,490 | 15,179 | +13% | 1 | 1 | 0% | 2,157 | 3,475 | +61% | 0 | 0 | — |
case-16 | pass→pass | 14,651 | 12,857 | -12% | 1 | 1 | 0% | 2,202 | 3,014 | +37% | 0 | 0 | — |
case-17 | fail→pass | 19,480 | 16,746 | -14% | 1 | 1 | 0% | 2,726 | 3,815 | +40% | 0 | 0 | — |
case-18 | pass→pass | 16,508 | 14,183 | -14% | 1 | 1 | 0% | 2,148 | 3,145 | +46% | 0 | 0 | — |
case-19 | fail→fail | 14,097 | 14,253 | +1% | 1 | 1 | 0% | 1,904 | 2,928 | +54% | 0 | 0 | — |
case-20 | pass→pass | 14,807 | 14,985 | +1% | 1 | 1 | 0% | 1,915 | 3,290 | +72% | 0 | 0 | — |
case-21 | pass→pass | 12,157 | 16,746 | +38% | 1 | 1 | 0% | 1,721 | 3,344 | +94% | 0 | 0 | — |
case-22 | fail→pass | 14,648 | 22,439 | +53% | 1 | 1 | 0% | 2,194 | 3,768 | +72% | 0 | 0 | — |
case-23 | fail→fail | 17,308 | 24,849 | +44% | 1 | 1 | 0% | 2,353 | 4,489 | +91% | 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 +48 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.