Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when a wealth team needs to build a balanced client allocation plan, compare current vs target weights, and deliver rebalancing actions, product shortlist, and compliant communication notes.
.claude/skills/aifinlab-bank-t179-wealth-management-asset-allocation-assistant/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-14 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 36% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 43% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 22% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 53% | 0% |
这个 skill 面向平衡型客户，强调在收益与回撤之间取得可解释的平衡。输出包括目标配置、现有持仓差异、再平衡路径与沟通话术。
scripts/run_skill.py：生成配置与再平衡建议shared/wealth_management_skill_engine.py示例命令：
bashpython scripts/run_skill.py --input assets/example-input.json --format markdown
references/input-schema.mdreferences/output-schema.mdreferences/allocation-framework.mdreferences/rebalancing-guidelines.mdreferences/communication-points.md| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 34,698 | 41,840 | +21% | 1 | 1 | 0% | 4,525 | 5,761 | +27% | 0 | 0 | — |
case-02 | pass→pass | 21,146 | 24,363 | +15% | 1 | 1 | 0% | 3,245 | 4,415 | +36% | 0 | 0 | — |
case-03 | pass→pass | 16,035 | 22,952 | +43% | 1 | 1 | 0% | 2,481 | 3,539 | +43% | 0 | 0 | — |
case-04 | pass→pass | 22,966 | 24,051 | +5% | 1 | 1 | 0% | 3,126 | 3,829 | +22% | 0 | 0 | — |
case-05 | pass→pass | 14,759 | 18,123 | +23% | 1 | 1 | 0% | 2,251 | 3,451 | +53% | 0 | 0 | — |
case-06 | pass→pass | 17,253 | 18,036 | +5% | 1 | 1 | 0% | 2,622 | 3,419 | +30% | 0 | 0 | — |
case-07 | pass→pass | 24,161 | 17,157 | -29% | 1 | 1 | 0% | 2,242 | 3,046 | +36% | 0 | 0 | — |
case-08 | pass→pass | 24,685 | 26,013 | +5% | 1 | 1 | 0% | 3,456 | 4,274 | +24% | 0 | 0 | — |
case-09 | pass→pass | 11,658 | 18,064 | +55% | 1 | 1 | 0% | 1,751 | 3,254 | +86% | 0 | 0 | — |
case-10 | pass→pass | 22,686 | 24,520 | +8% | 1 | 1 | 0% | 2,938 | 4,471 | +52% | 0 | 0 | — |
case-11 | pass→pass | 20,552 | 15,799 | -23% | 1 | 1 | 0% | 2,478 | 2,962 | +20% | 0 | 0 | — |
case-12 | pass→pass | 30,975 | 21,624 | -30% | 1 | 1 | 0% | 2,606 | 3,220 | +24% | 0 | 0 | — |
case-13 | pass→pass | 25,791 | 28,314 | +10% | 1 | 1 | 0% | 3,835 | 4,560 | +19% | 0 | 0 | — |
case-14 | fail→pass | 17,978 | 17,077 | -5% | 1 | 1 | 0% | 2,247 | 2,840 | +26% | 0 | 0 | — |
case-15 | pass→pass | 19,223 | 20,460 | +6% | 1 | 1 | 0% | 3,004 | 3,673 | +22% | 0 | 0 | — |
case-16 | pass→pass | 13,610 | 14,919 | +10% | 1 | 1 | 0% | 2,322 | 3,069 | +32% | 0 | 0 | — |
case-17 | pass→pass | 22,994 | 22,515 | -2% | 1 | 1 | 0% | 2,946 | 3,976 | +35% | 0 | 0 | — |
case-18 | pass→pass | 15,289 | 16,187 | +6% | 1 | 1 | 0% | 1,953 | 2,497 | +28% | 0 | 0 | — |
case-19 | pass→pass | 20,506 | 25,190 | +23% | 1 | 1 | 0% | 3,038 | 4,272 | +41% | 0 | 0 | — |
case-20 | pass→pass | 17,051 | 23,438 | +37% | 1 | 1 | 0% | 2,419 | 3,608 | +49% | 0 | 0 | — |
case-21 | pass→pass | 6,730 | 14,437 | +115% | 1 | 1 | 0% | 842 | 2,515 | +199% | 0 | 0 | — |
case-22 | pass→pass | 12,720 | 13,770 | +8% | 1 | 1 | 0% | 1,790 | 2,305 | +29% | 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.
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.