Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股板块动量/行业轮动动量策略。当用户说"板块动量"、"sector momentum"、"哪个板块强"、"行业动量"、"追强势板块"时触发。量化分析板块动量排序。支持formal和brief风格。
.claude/skills/aifinlab-a-share-sector-momentum/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -62% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" python "$SCRIPTS/cn_stock_data.py" kline --code [CODE] --freq daily --start [日期] python "$SCRIPTS/cn_stock_data.py" quote --code [CODE] python "$SCRIPTS/cn_stock_data.py" finance --code [CODE]
选Top K强势板块持有M日的历史表现
| 维度 | formal | brief | |------|--------|-------| | 板块排名 | 多周期完整排名 | Top/Bottom 5 | | 动量分析 | 动量持续性统计 | 当前强弱 | | 策略回测 | 完整回测结果 | 推荐板块 | 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-01 | fail→fail | 19,058 | 18,841 | -1% | 1 | 1 | 0% | 2,455 | 1,115 | -55% | 0 | 0 | — |
case-02 | fail→pass | 36,409 | 22,906 | -37% | 1 | 1 | 0% | 4,012 | 3,652 | -9% | 0 | 0 | — |
case-03 | pass→pass | 120,101 | 95,477 | -21% | 1 | 1 | 0% | 7,475 | 6,928 | -7% | 0 | 0 | — |
case-04 | pass→pass | 23,156 | 19,366 | -16% | 1 | 1 | 0% | 2,970 | 3,301 | +11% | 0 | 0 | — |
case-05 | pass→pass | 22,927 | 16,431 | -28% | 1 | 1 | 0% | 3,244 | 2,808 | -13% | 0 | 0 | — |
case-06 | pass→pass | 21,013 | 36,671 | +75% | 1 | 1 | 0% | 2,722 | 3,082 | +13% | 0 | 0 | — |
case-07 | pass→pass | 17,048 | 15,852 | -7% | 1 | 1 | 0% | 2,259 | 2,484 | +10% | 0 | 0 | — |
case-08 | fail→pass | 20,927 | 18,835 | -10% | 1 | 1 | 0% | 3,225 | 3,507 | +9% | 0 | 0 | — |
case-09 | fail→fail | 21,628 | 67,108 | +210% | 1 | 1 | 0% | 2,975 | 871 | -71% | 0 | 0 | — |
case-10 | pass→pass | 27,127 | 24,031 | -11% | 1 | 1 | 0% | 3,344 | 3,586 | +7% | 0 | 0 | — |
case-11 | fail→pass | 19,990 | 7,672 | -62% | 1 | 1 | 0% | 3,684 | 1,713 | -54% | 0 | 0 | — |
case-12 | pass→pass | 22,573 | 21,076 | -7% | 1 | 1 | 0% | 3,450 | 3,716 | +8% | 0 | 0 | — |
case-13 | fail→fail | 28,167 | 30,653 | +9% | 1 | 1 | 0% | 4,408 | 5,660 | +28% | 0 | 0 | — |
case-14 | fail→fail | 19,211 | 13,309 | -31% | 1 | 1 | 0% | 2,973 | 2,745 | -8% | 0 | 0 | — |
case-15 | pass→pass | 25,200 | 17,431 | -31% | 1 | 1 | 0% | 3,437 | 3,430 | -0% | 0 | 0 | — |
case-16 | fail→fail | 10,896 | 6,516 | -40% | 1 | 1 | 0% | 1,586 | 1,468 | -7% | 0 | 0 | — |
case-17 | fail→pass | 10,297 | 2,783 | -73% | 1 | 1 | 0% | 1,680 | 936 | -44% | 0 | 0 | — |
case-18 | fail→pass | 13,783 | 2,502 | -82% | 1 | 1 | 0% | 2,487 | 953 | -62% | 0 | 0 | — |
case-19 | pass→pass | 14,837 | 12,030 | -19% | 1 | 1 | 0% | 2,315 | 2,222 | -4% | 0 | 0 | — |
case-20 | pass→pass | 27,063 | 32,634 | +21% | 1 | 1 | 0% | 5,384 | 6,453 | +20% | 0 | 0 | — |
case-21 | pass→pass | 33,570 | 32,568 | -3% | 1 | 1 | 0% | 5,069 | 6,139 | +21% | 0 | 0 | — |
case-22 | pass→pass | 30,727 | 46,401 | +51% | 1 | 1 | 0% | 4,259 | 5,092 | +20% | 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, and 20 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 +23 percentage points is the difference between those two pass rates over the 20 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.