Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股个股财报分析/业绩点评。当用户说"财报分析"、"季报解读"、"年报分析"、"业绩点评"、"XX的财报怎么样"、"earnings analysis"时触发。从 cn-stock-data 获取财务指标和行情数据,结合 web 搜索获取财报公告原文和市场预期,生成结构化财报分析报告。支持券商业绩点评风格(formal, 8-12页 .docx)和个人分析笔记风格(brief, 1-2页 Markdown)。不适用于晨会纪要(用 a-share-morning-note)或选股筛选(用 a-share-stock-screen)。
.claude/skills/aifinlab-a-share-earnings-analysis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -17% | 0% |
训练数据中的财务数据已过期。 每次执行时必须:
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 财务指标(adata 43 字段) python "$SCRIPTS/cn_stock_data.py" finance --code [CODE] # 近 6 个月日线行情(分析股价走势 vs 业绩) python "$SCRIPTS/cn_stock_data.py" kline --code [CODE] --freq daily --start [6个月前日期] # 近 30 日资金流向 python "$SCRIPTS/cn_stock_data.py" fund_flow --code [CODE] --days 30 # 实时行情(当前估值水平) python "$SCRIPTS/cn_stock_data.py" quote --code [CODE]
详见 references/workflow.md 的 5 阶段分步指引。
参见 references/financial-metrics.md 的指标解读规则。
根据用户风格要求选择模板,参见 references/report-structure.md。
| 维度 | formal(券商业绩点评) | brief(个人分析笔记) | |------|---------------------|---------------------| | 篇幅 | 8-12 页 .docx | 1-2 页 Markdown | | 标题 | "公司名YYYY年QX]业绩点评" | "XX QX] 财报笔记" | | 结构 | 投资要点→财务分析→估值→风险→附表 | 关键数字→亮点→风险→个人判断 | | 图表 | 4-6 个(营收趋势、利润趋势、ROE、PE Band) | 关键数据表 1 个 | | 引用 | 必须标注数据来源 | 可省略 | | 免责声明 | 需要 | 不需要 | | 投资评级 | 不给(避免合规风险) | 可给个人判断 |
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 | 18,334 | 17,155 | -6% | 1 | 1 | 0% | 2,587 | 1,979 | -24% | 0 | 0 | — |
case-02 | fail→fail | 27,818 | 11,305 | -59% | 1 | 1 | 0% | 4,316 | 1,708 | -60% | 0 | 0 | — |
case-03 | fail→fail | 22,330 | 10,767 | -52% | 1 | 1 | 0% | 3,044 | 1,564 | -49% | 0 | 0 | — |
case-04 | fail→pass | 12,697 | 4,428 | -65% | 1 | 1 | 0% | 1,902 | 1,778 | -7% | 0 | 0 | — |
case-05 | fail→pass | 21,832 | 13,597 | -38% | 1 | 1 | 0% | 3,392 | 3,336 | -2% | 0 | 0 | — |
case-06 | fail→fail | 17,070 | 11,161 | -35% | 1 | 1 | 0% | 2,383 | 1,671 | -30% | 0 | 0 | — |
case-07 | pass→fail | 28,805 | 15,891 | -45% | 1 | 1 | 0% | 3,806 | 1,788 | -53% | 0 | 0 | — |
case-08 | fail→fail | 13,547 | 12,740 | -6% | 1 | 1 | 0% | 2,010 | 1,807 | -10% | 0 | 0 | — |
case-09 | pass→pass | 14,947 | 13,105 | -12% | 1 | 1 | 0% | 2,422 | 3,125 | +29% | 0 | 0 | — |
case-10 | fail→pass | 25,785 | 25,183 | -2% | 1 | 1 | 0% | 3,776 | 4,805 | +27% | 0 | 0 | — |
case-11 | pass→pass | 8,954 | 9,113 | +2% | 1 | 1 | 0% | 1,225 | 2,241 | +83% | 0 | 0 | — |
case-12 | pass→pass | 21,499 | 14,875 | -31% | 1 | 1 | 0% | 2,926 | 3,294 | +13% | 0 | 0 | — |
case-13 | fail→pass | 19,782 | 12,065 | -39% | 1 | 1 | 0% | 3,093 | 2,934 | -5% | 0 | 0 | — |
case-14 | fail→pass | 13,282 | 3,896 | -71% | 1 | 1 | 0% | 1,991 | 1,661 | -17% | 0 | 0 | — |
case-15 | fail→pass | 18,594 | 11,033 | -41% | 1 | 1 | 0% | 2,825 | 2,862 | +1% | 0 | 0 | — |
case-16 | pass→pass | 19,319 | 11,789 | -39% | 1 | 1 | 0% | 2,778 | 3,009 | +8% | 0 | 0 | — |
case-17 | fail→pass | 20,691 | 3,148 | -85% | 1 | 1 | 0% | 2,851 | 1,544 | -46% | 0 | 0 | — |
case-18 | pass→pass | 13,933 | 7,378 | -47% | 1 | 1 | 0% | 1,880 | 1,974 | +5% | 0 | 0 | — |
case-19 | pass→fail | 22,041 | 8,063 | -63% | 1 | 1 | 0% | 4,004 | 1,348 | -66% | 0 | 0 | — |
case-20 | pass→fail | 26,369 | 9,114 | -65% | 1 | 1 | 0% | 3,859 | 1,546 | -60% | 0 | 0 | — |
case-21 | pass→fail | 21,269 | 9,851 | -54% | 1 | 1 | 0% | 3,283 | 1,532 | -53% | 0 | 0 | — |
case-22 | fail→pass | 15,529 | 4,425 | -72% | 1 | 1 | 0% | 2,254 | 1,655 | -27% | 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 13 counted toward the lift figure. The other 9 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 +18 percentage points is the difference between those two pass rates over the 13 comparable cases. 7 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/28/2026 | +36% |
Other measured skills in the registry, with their headline benchmark lift.