Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股指数增强策略/超额收益分析。当用户说"指数增强"、"index enhance"、"超额收益"、"跑赢指数"、"增强策略"、"alpha"、"怎么跑赢沪深300"时触发。基于 cn-stock-data 获取指数成分股数据，量化构建指数增强组合。支持研报风格（formal）和快速分析风格（brief）。
.claude/skills/aifinlab-a-share-index-enhance/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -24% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" python "$SCRIPTS/cn_stock_data.py" kline --code [INDEX_CODE] --freq daily --start [日期] python "$SCRIPTS/cn_stock_data.py" finance --code [CODE1],[CODE2],... python "$SCRIPTS/cn_stock_data.py" quote --code [CODE1],[CODE2],...
沪深300(SH000300) / 中证500(SH000905) / 中证1000(SH000852)
获取指数成分股列表、权重、K线数据、财务指标。
| 维度 | formal | brief | |------|--------|-------| | 因子构成 | 多因子权重 + IC/IR | 主要alpha来源 | | 组合构建 | 完整超配/低配名单 | Top 10 超配 | | 跟踪误差 | TE 目标 + 信息比率 | 预期超额 |
默认风格：brief。
python# 调用 skill result = run_skill({ "param1": "value1", "param2": "value2" })
bashpython scripts/run_skill.py --input data.json
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | fail→pass | 27,247 | 27,719 | +2% | 1 | 1 | 0% | 4,175 | 4,241 | +2% | 0 | 0 | — |
case-01 | fail→fail | 27,496 | 21,624 | -21% | 1 | 1 | 0% | 3,747 | 3,845 | +3% | 0 | 0 | — |
case-02 | fail→fail | 35,471 | 10,132 | -71% | 1 | 1 | 0% | 5,931 | 1,139 | -81% | 0 | 0 | — |
case-03 | fail→fail | 30,360 | 11,973 | -61% | 1 | 1 | 0% | 4,015 | 1,156 | -71% | 0 | 0 | — |
case-05 | fail→pass | 20,131 | 17,789 | -12% | 1 | 1 | 0% | 3,190 | 2,657 | -17% | 0 | 0 | — |
case-06 | fail→fail | 22,614 | 23,009 | +2% | 1 | 1 | 0% | 2,983 | 3,667 | +23% | 0 | 0 | — |
case-07 | fail→pass | 20,891 | 12,236 | -41% | 1 | 1 | 0% | 3,263 | 2,372 | -27% | 0 | 0 | — |
case-08 | fail→fail | 21,864 | 33,016 | +51% | 1 | 1 | 0% | 3,267 | 3,059 | -6% | 0 | 0 | — |
case-09 | fail→pass | 38,542 | 11,947 | -69% | 1 | 1 | 0% | 2,639 | 2,434 | -8% | 0 | 0 | — |
case-10 | fail→fail | 23,404 | 43,196 | +85% | 1 | 1 | 0% | 3,219 | 3,812 | +18% | 0 | 0 | — |
case-11 | pass→pass | 14,779 | 10,567 | -28% | 1 | 1 | 0% | 2,359 | 2,034 | -14% | 0 | 0 | — |
case-12 | fail→fail | 15,474 | 7,579 | -51% | 1 | 1 | 0% | 2,256 | 1,894 | -16% | 0 | 0 | — |
case-13 | fail→fail | 30,930 | 44,161 | +43% | 1 | 1 | 0% | 4,042 | 6,076 | +50% | 0 | 0 | — |
case-14 | fail→pass | 13,437 | 6,653 | -50% | 1 | 1 | 0% | 1,978 | 1,504 | -24% | 0 | 0 | — |
case-15 | fail→fail | 14,682 | 4,862 | -67% | 1 | 1 | 0% | 1,884 | 1,321 | -30% | 0 | 0 | — |
case-16 | pass→pass | 5,938 | 2,211 | -63% | 1 | 1 | 0% | 1,028 | 966 | -6% | 0 | 0 | — |
case-17 | fail→pass | 21,998 | 7,384 | -66% | 1 | 1 | 0% | 3,380 | 1,705 | -50% | 0 | 0 | — |
case-18 | fail→pass | 12,925 | 5,402 | -58% | 1 | 1 | 0% | 2,295 | 1,201 | -48% | 0 | 0 | — |
case-19 | fail→pass | 17,817 | 4,350 | -76% | 1 | 1 | 0% | 3,262 | 1,330 | -59% | 0 | 0 | — |
case-20 | pass→pass | 13,064 | 5,878 | -55% | 1 | 1 | 0% | 2,094 | 1,531 | -27% | 0 | 0 | — |
case-21 | fail→pass | 23,849 | 11,827 | -50% | 1 | 1 | 0% | 3,296 | 2,565 | -22% | 0 | 0 | — |
case-22 | fail→fail | 20,380 | 16,022 | -21% | 1 | 1 | 0% | 3,108 | 2,785 | -10% | 0 | 0 | — |
case-23 | pass→pass | 31,221 | 37,486 | +20% | 1 | 1 | 0% | 4,412 | 4,589 | +4% | 0 | 0 | — |
case-24 | pass→pass | 29,515 | 53,081 | +80% | 1 | 1 | 0% | 4,219 | 4,826 | +14% | 0 | 0 | — |
case-25 | pass→pass | 26,348 | 52,778 | +100% | 1 | 1 | 0% | 4,319 | 5,287 | +22% | 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. 25 cases were attempted, and 23 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +36 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.