Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股晨会纪要/盘前要点生成。当用户说"晨会纪要"、"今日看点"、"盘前要点"、"隔夜发生了什么"、"morning note"、"今天市场怎么样"时触发。基于实时行情、北向资金、资金流向和隔夜新闻生成结构化晨会纪要。支持券商研报风格（formal）和个人笔记风格（brief）。不适用于个股深度分析（用 a-share-earnings-analysis）或选股（用 a-share-stock-screen）。
.claude/skills/aifinlab-a-share-morning-note/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-16 | ✓→✗ | ▼ Worse | -39% | 0% |
| case-08 | ✓→✗ | ▼ Worse | -39% | 0% |
| case-09 | ✓→✗ | ▼ Worse | -63% | 0% |
| case-11 | ✓→✗ | ▼ Worse | -30% | 0% |
实时数据（必需，通过 cn-stock-data 统一层获取）：
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 1. 大盘指数行情（上证、深证、创业板） python "$SCRIPTS/cn_stock_data.py" quote --code SH000001,SZ399001,SZ399006 # 2. 北向资金（注意：自2024年8月起北向明细不再实时披露，数据可能为0） python "$SCRIPTS/cn_stock_data.py" north_flow # 如 north_flow 返回全0，改用 web 搜索"北向资金 今日"获取 # 3. 热门个股行情（可选，用户指定或默认关注池） python "$SCRIPTS/cn_stock_data.py" quote --code SH600519,SZ000001,SZ300750
新闻资讯（必需，通过 web 搜索）：
数据时效性要求：行情数据必须是最新交易日的，新闻必须是24小时内的。
运行上述 cn-stock-data 命令获取：
如果 cn-stock-data 返回 "ok": false，在输出中标注"数据暂不可用"而非虚构数据。
通过 web 搜索获取：
根据用户要求的风格输出。如果用户未指定风格，默认使用 brief（个人笔记）。 如果用户说"正式"、"券商格式"、"给领导看"，使用 formal。
参见 references/report-template.md 获取两种风格的具体模板。
| 维度 | formal（券商晨会） | brief（个人笔记） | |------|-------------------|------------------| | 篇幅 | 1-2 页 A4 | 半页以内 | | 语言 | 书面语，第三人称 | 口语化，可用简写 | | 结构 | 完整标题+摘要+正文 | 要点列表 | | 数据表格 | 必须有指数+北向资金表格 | 关键数字内嵌 | | 观点 | 客观陈述，不给结论 | 可加个人判断 | | 免责声明 | 需要 | 不需要 |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,316 | 12,777 | -22% | 1 | 1 | 0% | 2,490 | 1,261 | -49% | 0 | 0 | — |
case-02 | fail→fail | 24,769 | 10,069 | -59% | 1 | 1 | 0% | 3,443 | 1,493 | -57% | 0 | 0 | — |
case-03 | fail→fail | 20,994 | 10,131 | -52% | 1 | 1 | 0% | 3,085 | 1,528 | -50% | 0 | 0 | — |
case-04 | fail→fail | 7,521 | 8,287 | +10% | 1 | 1 | 0% | 1,286 | 1,212 | -6% | 0 | 0 | — |
case-05 | fail→pass | 13,941 | 6,546 | -53% | 1 | 1 | 0% | 2,357 | 1,798 | -24% | 0 | 0 | — |
case-06 | fail→fail | 20,967 | 14,430 | -31% | 1 | 1 | 0% | 3,493 | 2,695 | -23% | 0 | 0 | — |
case-16 | pass→fail | 16,214 | 32,407 | +100% | 1 | 1 | 0% | 2,314 | 1,400 | -39% | 0 | 0 | — |
case-07 | fail→fail | 15,473 | 5,107 | -67% | 1 | 1 | 0% | 2,668 | 1,629 | -39% | 0 | 0 | — |
case-08 | pass→fail | 14,378 | 16,774 | +17% | 1 | 1 | 0% | 2,085 | 1,266 | -39% | 0 | 0 | — |
case-09 | pass→fail | 23,031 | 8,366 | -64% | 1 | 1 | 0% | 3,488 | 1,289 | -63% | 0 | 0 | — |
case-10 | fail→fail | 22,314 | 10,281 | -54% | 1 | 1 | 0% | 2,839 | 1,337 | -53% | 0 | 0 | — |
case-11 | pass→fail | 16,709 | 13,081 | -22% | 1 | 1 | 0% | 2,362 | 1,656 | -30% | 0 | 0 | — |
case-12 | fail→fail | 41,688 | 10,693 | -74% | 1 | 1 | 0% | 2,406 | 1,471 | -39% | 0 | 0 | — |
case-13 | pass→pass | 12,439 | 7,310 | -41% | 1 | 1 | 0% | 2,305 | 2,013 | -13% | 0 | 0 | — |
case-14 | pass→pass | 8,118 | 11,369 | +40% | 1 | 1 | 0% | 1,339 | 2,894 | +116% | 0 | 0 | — |
case-15 | pass→fail | 18,108 | 11,001 | -39% | 1 | 1 | 0% | 2,681 | 1,368 | -49% | 0 | 0 | — |
case-17 | pass→pass | 15,921 | 13,747 | -14% | 1 | 1 | 0% | 2,209 | 2,613 | +18% | 0 | 0 | — |
case-18 | fail→fail | 11,427 | 4,649 | -59% | 1 | 1 | 0% | 1,535 | 1,384 | -10% | 0 | 0 | — |
case-19 | pass→pass | 25,982 | 18,212 | -30% | 1 | 1 | 0% | 3,948 | 3,318 | -16% | 0 | 0 | — |
case-20 | pass→pass | 19,165 | 19,154 | -0% | 1 | 1 | 0% | 2,899 | 3,540 | +22% | 0 | 0 | — |
case-21 | pass→pass | 21,965 | 15,473 | -30% | 1 | 1 | 0% | 3,151 | 3,112 | -1% | 0 | 0 | — |
case-22 | pass→pass | 21,517 | 19,032 | -12% | 1 | 1 | 0% | 4,442 | 4,746 | +7% | 0 | 0 | — |
case-23 | fail→fail | 10,334 | 10,084 | -2% | 1 | 1 | 0% | 1,744 | 2,313 | +33% | 0 | 0 | — |
case-24 | pass→pass | 40,934 | 33,011 | -19% | 1 | 1 | 0% | 5,912 | 6,294 | +6% | 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 -17 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.