Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股资金轮动/板块资金流量化分析。当用户说"资金轮动"、"money rotation"、"板块资金流"、"资金去哪了"、"哪个板块在吸金"、"资金迁移"、"轮动节奏"时触发。基于 cn-stock-data 获取多板块资金流向数据，量化分析资金在不同板块间的轮动规律。支持研报风格（formal）和快速分析风格（brief）。
.claude/skills/aifinlab-a-share-money-rotation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-15 | ✗→✓ | ▲ Improved | -41% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -39% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 8% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 各板块代表股的资金流向 python "$SCRIPTS/cn_stock_data.py" fund_flow --code [CODE] --days 30 # 板块指数K线 python "$SCRIPTS/cn_stock_data.py" kline --code [INDEX_CODE] --freq daily --start [日期] # 实时行情 python "$SCRIPTS/cn_stock_data.py" quote --code [CODE1],[CODE2],...
量化计算：
bashQSCRIPTS="$SKILLS_ROOT/a-share-money-rotation/scripts" python "$QSCRIPTS/rotation_tracker.py" --data sectors_flow.json --window 20
| 维度 | formal | brief | |------|--------|-------| | 板块排名 | 全板块资金流向表 | Top 3 流入/流出 | | 轮动方向 | 详细流向矩阵 | 一句话总结 | | 历史对比 | 与过去轮动周期对比 | 无 |
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-15 | fail→pass | 13,717 | 2,629 | -81% | 1 | 1 | 0% | 1,709 | 1,009 | -41% | 0 | 0 | — |
case-01 | fail→fail | 19,445 | 17,144 | -12% | 1 | 1 | 0% | 2,314 | 1,545 | -33% | 0 | 0 | — |
case-02 | fail→fail | 32,026 | 9,631 | -70% | 1 | 1 | 0% | 4,885 | 965 | -80% | 0 | 0 | — |
case-03 | fail→fail | 21,133 | 13,530 | -36% | 1 | 1 | 0% | 2,758 | 1,296 | -53% | 0 | 0 | — |
case-04 | pass→pass | 27,959 | 32,128 | +15% | 1 | 1 | 0% | 5,724 | 5,539 | -3% | 0 | 0 | — |
case-05 | pass→fail | 20,673 | 7,058 | -66% | 1 | 1 | 0% | 2,824 | 877 | -69% | 0 | 0 | — |
case-06 | pass→fail | 26,154 | 9,531 | -64% | 1 | 1 | 0% | 4,044 | 1,049 | -74% | 0 | 0 | — |
case-16 | fail→pass | 13,276 | 2,294 | -83% | 1 | 1 | 0% | 1,555 | 945 | -39% | 0 | 0 | — |
case-17 | fail→fail | 9,607 | 3,492 | -64% | 1 | 1 | 0% | 1,365 | 1,052 | -23% | 0 | 0 | — |
case-07 | fail→pass | 13,854 | 8,665 | -37% | 1 | 1 | 0% | 2,018 | 1,979 | -2% | 0 | 0 | — |
case-08 | pass→pass | 21,970 | 18,620 | -15% | 1 | 1 | 0% | 2,905 | 3,340 | +15% | 0 | 0 | — |
case-09 | pass→pass | 23,855 | 16,687 | -30% | 1 | 1 | 0% | 2,900 | 3,233 | +11% | 0 | 0 | — |
case-10 | pass→pass | 18,455 | 7,794 | -58% | 1 | 1 | 0% | 2,351 | 1,596 | -32% | 0 | 0 | — |
case-11 | fail→pass | 15,681 | 13,052 | -17% | 1 | 1 | 0% | 2,288 | 2,467 | +8% | 0 | 0 | — |
case-12 | pass→pass | 13,047 | 14,684 | +13% | 1 | 1 | 0% | 1,962 | 2,798 | +43% | 0 | 0 | — |
case-13 | fail→pass | 20,652 | 20,119 | -3% | 1 | 1 | 0% | 2,940 | 3,172 | +8% | 0 | 0 | — |
case-14 | fail→fail | 13,842 | 4,952 | -64% | 1 | 1 | 0% | 2,257 | 1,418 | -37% | 0 | 0 | — |
case-18 | pass→pass | 12,883 | 7,426 | -42% | 1 | 1 | 0% | 1,768 | 1,342 | -24% | 0 | 0 | — |
case-19 | fail→pass | 14,349 | 10,784 | -25% | 1 | 1 | 0% | 1,939 | 1,951 | +1% | 0 | 0 | — |
case-20 | pass→pass | 20,477 | 10,082 | -51% | 1 | 1 | 0% | 3,092 | 1,692 | -45% | 0 | 0 | — |
case-21 | pass→pass | 14,034 | 21,758 | +55% | 1 | 1 | 0% | 2,228 | 2,141 | -4% | 0 | 0 | — |
case-22 | fail→pass | 20,699 | 17,065 | -18% | 1 | 1 | 0% | 2,594 | 2,992 | +15% | 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 17 counted toward the lift figure. The other 5 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 17 comparable cases. 5 cases got worse with the skill loaded, and they are 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.