Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →银行网点/支行/分行经营业绩横向对比排名。当用户要求对比各支行存款规模、分行业务排名、网点经营指标对标、机构绩效差距分析时使用。输出排名表、差距分析和分类标签。
.claude/skills/aifinlab-branch-performance-benchmarking/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 74% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 24% | 0% |
银行管理层需要定期对各分支机构的经营表现进行横向对比，识别哪些网点表现优秀、哪些网点落后，以及差距的主要成因。传统方式依赖人工整理排名表，难以多维度交叉分析和快速下钻。本技能对多网点经营数据进行标准化对比，输出排名、差距分析和分类标签，为差异化管理提供数据支撑。
| 输入项 | 说明 | 是否必须 | |--------|------|----------| | 机构清单 | 需要对比的网点/支行/分行列表 | 必须 | | 经营数据 | 各机构的核心经营指标数据（存款、贷款、客户数、AUM等） | 必须 | | 对比维度 | 需要横向对比的指标项 | 可选，默认全量核心指标 | | 对比基准 | 平均值/中位数/标杆机构/目标值 | 可选，默认平均值 | | 机构分类 | 网点类型标签（如社区型、商圈型） | 可选 | | 分析期间 | 数据所属时间范围 | 必须 |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 56,740 | 66,016 | +16% | 1 | 1 | 0% | 4,775 | 8,298 | +74% | 0 | 0 | — |
case-02 | fail→fail | 24,119 | 27,372 | +13% | 1 | 1 | 0% | 4,141 | 5,920 | +43% | 0 | 0 | — |
case-03 | fail→fail | 22,601 | 43,041 | +90% | 1 | 1 | 0% | 4,074 | 6,647 | +63% | 0 | 0 | — |
case-04 | fail→pass | 44,575 | 11,217 | -75% | 1 | 1 | 0% | 2,292 | 2,393 | +4% | 0 | 0 | — |
case-05 | fail→pass | 25,100 | 12,302 | -51% | 1 | 1 | 0% | 3,573 | 2,822 | -21% | 0 | 0 | — |
case-06 | fail→pass | 18,041 | 8,588 | -52% | 1 | 1 | 0% | 2,584 | 2,206 | -15% | 0 | 0 | — |
case-07 | fail→pass | 16,801 | 22,650 | +35% | 1 | 1 | 0% | 3,002 | 5,238 | +74% | 0 | 0 | — |
case-08 | fail→pass | 12,866 | 12,118 | -6% | 1 | 1 | 0% | 2,586 | 3,207 | +24% | 0 | 0 | — |
case-09 | pass→pass | 13,994 | 12,724 | -9% | 1 | 1 | 0% | 2,405 | 3,443 | +43% | 0 | 0 | — |
case-10 | fail→fail | 16,568 | 18,214 | +10% | 1 | 1 | 0% | 2,532 | 4,095 | +62% | 0 | 0 | — |
case-11 | pass→pass | 18,452 | 17,619 | -5% | 1 | 1 | 0% | 2,940 | 3,787 | +29% | 0 | 0 | — |
case-12 | pass→pass | 14,718 | 15,758 | +7% | 1 | 1 | 0% | 2,519 | 3,877 | +54% | 0 | 0 | — |
case-13 | pass→pass | 13,644 | 16,536 | +21% | 1 | 1 | 0% | 2,475 | 4,022 | +63% | 0 | 0 | — |
case-14 | pass→pass | 11,809 | 14,517 | +23% | 1 | 1 | 0% | 2,100 | 3,580 | +70% | 0 | 0 | — |
case-15 | pass→pass | 14,452 | 14,831 | +3% | 1 | 1 | 0% | 2,415 | 3,609 | +49% | 0 | 0 | — |
case-16 | pass→pass | 12,714 | 12,731 | +0% | 1 | 1 | 0% | 2,267 | 3,209 | +42% | 0 | 0 | — |
case-17 | pass→pass | 16,873 | 11,995 | -29% | 1 | 1 | 0% | 2,964 | 3,184 | +7% | 0 | 0 | — |
case-18 | pass→pass | 13,184 | 15,461 | +17% | 1 | 1 | 0% | 2,118 | 3,645 | +72% | 0 | 0 | — |
case-19 | pass→pass | 19,467 | 20,957 | +8% | 1 | 1 | 0% | 2,729 | 4,288 | +57% | 0 | 0 | — |
case-20 | pass→pass | 19,788 | 14,691 | -26% | 1 | 1 | 0% | 2,960 | 3,402 | +15% | 0 | 0 | — |
case-21 | fail→pass | 9,258 | 15,912 | +72% | 1 | 1 | 0% | 1,690 | 3,775 | +123% | 0 | 0 | — |
case-22 | pass→pass | 15,079 | 17,778 | +18% | 1 | 1 | 0% | 2,326 | 4,351 | +87% | 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 +27 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.