Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →当用户需要在银行零售金融场景下，对手机银行活跃度进行提升、分层排序、触达动作设计或效果指标设定时使用本技能。适合输出优先级名单、触达节奏与复盘指标。
.claude/skills/aifinlab-bank-t167-retail-finance-activity-lift-assistant/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✗→✓ | ▲ Improved | -67% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 145% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -51% | 0% |
本技能用于把“活跃提升目标”转成可执行的客群分层、触达计划与复盘指标。强调目标清晰、分层可解释、动作可执行、合规触达。
batch_id、time_window、customersreferences/input-schema.mdbashpython scripts/run_skill.py --input assets/example-input.json --format markdown python scripts/run_skill.py --input assets/example-input.json --format json
脚本入口：scripts/run_skill.py（调用 shared/retail_service_ops_skill_engine.py 的 t167 场景）。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | fail→pass | 17,772 | 2,743 | -85% | 1 | 1 | 0% | 2,870 | 933 | -67% | 0 | 0 | — |
case-01 | fail→fail | 24,352 | 34,565 | +42% | 1 | 1 | 0% | 3,767 | 5,063 | +34% | 0 | 0 | — |
case-02 | fail→pass | 12,037 | 15,773 | +31% | 1 | 1 | 0% | 1,159 | 2,840 | +145% | 0 | 0 | — |
case-03 | fail→pass | 42,479 | 22,347 | -47% | 1 | 1 | 0% | 5,244 | 3,677 | -30% | 0 | 0 | — |
case-04 | fail→fail | 22,947 | 25,655 | +12% | 1 | 1 | 0% | 3,017 | 3,675 | +22% | 0 | 0 | — |
case-05 | fail→fail | 20,090 | 19,034 | -5% | 1 | 1 | 0% | 2,629 | 3,524 | +34% | 0 | 0 | — |
case-06 | pass→pass | 18,256 | 21,441 | +17% | 1 | 1 | 0% | 2,678 | 3,392 | +27% | 0 | 0 | — |
case-07 | pass→pass | 16,294 | 18,993 | +17% | 1 | 1 | 0% | 2,212 | 3,149 | +42% | 0 | 0 | — |
case-08 | pass→pass | 16,897 | 19,116 | +13% | 1 | 1 | 0% | 2,519 | 3,553 | +41% | 0 | 0 | — |
case-09 | fail→fail | 23,086 | 17,190 | -26% | 1 | 1 | 0% | 2,976 | 3,433 | +15% | 0 | 0 | — |
case-10 | pass→pass | 21,079 | 33,014 | +57% | 1 | 1 | 0% | 2,975 | 3,542 | +19% | 0 | 0 | — |
case-11 | fail→pass | 5,941 | 6,393 | +8% | 1 | 1 | 0% | 1,120 | 1,665 | +49% | 0 | 0 | — |
case-12 | fail→pass | 11,078 | 2,378 | -79% | 1 | 1 | 0% | 1,910 | 944 | -51% | 0 | 0 | — |
case-13 | fail→pass | 21,574 | 22,385 | +4% | 1 | 1 | 0% | 2,678 | 3,644 | +36% | 0 | 0 | — |
case-14 | fail→pass | 17,036 | 15,024 | -12% | 1 | 1 | 0% | 2,513 | 2,982 | +19% | 0 | 0 | — |
case-15 | fail→pass | 18,053 | 14,459 | -20% | 1 | 1 | 0% | 2,584 | 2,646 | +2% | 0 | 0 | — |
case-16 | pass→pass | 10,929 | 17,788 | +63% | 1 | 1 | 0% | 1,739 | 3,218 | +85% | 0 | 0 | — |
case-17 | pass→pass | 19,680 | 20,255 | +3% | 1 | 1 | 0% | 2,417 | 3,130 | +29% | 0 | 0 | — |
case-18 | pass→pass | 21,654 | 20,741 | -4% | 1 | 1 | 0% | 2,623 | 2,950 | +12% | 0 | 0 | — |
case-19 | fail→pass | 17,689 | 11,012 | -38% | 1 | 1 | 0% | 2,566 | 2,269 | -12% | 0 | 0 | — |
case-20 | pass→pass | 19,491 | 16,628 | -15% | 1 | 1 | 0% | 2,700 | 2,689 | -0% | 0 | 0 | — |
case-22 | fail→pass | 16,496 | 4,520 | -73% | 1 | 1 | 0% | 1,987 | 1,260 | -37% | 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 +45 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.