Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股季节性规律/日历效应。当用户说"季节性"、"日历效应"、"春季躁动"、"红十月"、"五穷六绝七翻身"、"开门红"、"年末效应"、"seasonal"、"XX月一般怎么样"、"节前效应"、"两会行情"时触发。分析A股市场的历史月度/节假日/事件季节性规律，评估当前时间节点的统计概率优势，辅助择时决策。支持研报风格（formal）和快速参考风格（brief）。
.claude/skills/aifinlab-a-share-seasonal-pattern/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-16 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 70% | 0% |
| case-23 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-24 | ✗→✓ | ▲ Improved | 46% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 拉取上证指数近10年以上历史日K线（将 START 替换为10年前日期 YYYYMMDD） python "$SCRIPTS/cn_stock_data.py" kline --code SH000001 --freq daily --start START # 沪深300/创业板指同期数据 python "$SCRIPTS/cn_stock_data.py" kline --code SH000300,SZ399006 --freq daily --start START # 月K线（用于快速统计月度收益率） python "$SCRIPTS/cn_stock_data.py" kline --code SH000001,SH000300,SZ399006 --freq monthly --start START
补充数据通过 web 搜索获取：历史日历效应统计研报、节假日前后涨跌概率、两会/中央经济工作会议时间表、当前宏观政策环境。
默认 brief；用户说"详细分析/完整报告/formal"时用 formal。
| 维度 | formal | brief | |------|--------|-------| | 篇幅 | 2-3 页 | 半页 | | 月度统计 | 全部12个月详细数据表 | 仅当前及相邻月份 | | 效应检验 | 多个效应完整回测 | 当前最相关的1-2个效应 | | 行业季节性 | 分行业季节性排名 | 仅提当季强势行业 | | 历史对比 | 类似宏观环境年份对比 | 省略 | | 免责声明 | 需要 | 不需要 |
references/seasonal-pattern-guide.md> 以上季节性分析基于历史统计数据，仅供参考，不构成投资建议。历史规律不代表未来必然重现，市场有风险，投资需谨慎。
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-03 | fail→fail | 28,323 | 12,496 | -56% | 1 | 1 | 0% | 3,925 | 1,533 | -61% | 0 | 0 | — |
case-04 | pass→fail | 19,752 | 8,506 | -57% | 1 | 1 | 0% | 2,837 | 1,487 | -48% | 0 | 0 | — |
case-05 | pass→fail | 29,000 | 13,187 | -55% | 1 | 1 | 0% | 5,698 | 1,671 | -71% | 0 | 0 | — |
case-01 | fail→fail | 18,502 | 42,647 | +130% | 1 | 1 | 0% | 3,127 | 5,009 | +60% | 0 | 0 | — |
case-02 | fail→fail | 39,030 | 9,218 | -76% | 1 | 1 | 0% | 5,173 | 1,498 | -71% | 0 | 0 | — |
case-06 | pass→pass | 24,498 | 28,556 | +17% | 1 | 1 | 0% | 4,335 | 5,496 | +27% | 0 | 0 | — |
case-07 | fail→fail | 14,988 | 21,129 | +41% | 1 | 1 | 0% | 2,136 | 2,235 | +5% | 0 | 0 | — |
case-08 | fail→fail | 27,864 | 32,372 | +16% | 1 | 1 | 0% | 4,057 | 5,362 | +32% | 0 | 0 | — |
case-09 | pass→fail | 15,376 | 53,280 | +247% | 1 | 1 | 0% | 1,976 | 2,373 | +20% | 0 | 0 | — |
case-10 | pass→fail | 21,817 | 19,174 | -12% | 1 | 1 | 0% | 3,011 | 3,702 | +23% | 0 | 0 | — |
case-11 | fail→fail | 22,706 | 18,336 | -19% | 1 | 1 | 0% | 2,986 | 2,079 | -30% | 0 | 0 | — |
case-12 | pass→pass | 18,915 | 37,958 | +101% | 1 | 1 | 0% | 2,750 | 4,735 | +72% | 0 | 0 | — |
case-13 | pass→pass | 19,235 | 19,260 | +0% | 1 | 1 | 0% | 2,732 | 3,784 | +39% | 0 | 0 | — |
case-14 | fail→fail | 12,269 | 27,662 | +125% | 1 | 1 | 0% | 1,929 | 2,336 | +21% | 0 | 0 | — |
case-15 | fail→fail | 25,717 | 9,126 | -65% | 1 | 1 | 0% | 3,443 | 1,496 | -57% | 0 | 0 | — |
case-16 | fail→pass | 22,853 | 6,582 | -71% | 1 | 1 | 0% | 3,521 | 2,212 | -37% | 0 | 0 | — |
case-17 | fail→pass | 26,223 | 10,542 | -60% | 1 | 1 | 0% | 3,715 | 3,043 | -18% | 0 | 0 | — |
case-18 | fail→pass | 30,209 | 35,459 | +17% | 1 | 1 | 0% | 4,022 | 6,848 | +70% | 0 | 0 | — |
case-19 | pass→fail | 13,292 | 14,032 | +6% | 1 | 1 | 0% | 2,156 | 1,482 | -31% | 0 | 0 | — |
case-20 | pass→pass | 21,120 | 19,098 | -10% | 1 | 1 | 0% | 2,575 | 3,372 | +31% | 0 | 0 | — |
case-21 | pass→fail | 15,802 | 10,334 | -35% | 1 | 1 | 0% | 2,624 | 1,409 | -46% | 0 | 0 | — |
case-22 | pass→pass | 13,375 | 11,518 | -14% | 1 | 1 | 0% | 1,666 | 2,580 | +55% | 0 | 0 | — |
case-23 | fail→pass | 21,988 | 19,484 | -11% | 1 | 1 | 0% | 3,357 | 4,044 | +20% | 0 | 0 | — |
case-24 | fail→pass | 19,852 | 23,294 | +17% | 1 | 1 | 0% | 2,818 | 4,108 | +46% | 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. 24 cases were attempted, and 13 counted toward the lift figure. The other 11 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 -4 percentage points is the difference between those two pass rates over the 13 comparable cases. 8 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.