Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股现金流分析/自由现金流质量评估。当用户说"现金流"、"自由现金流"、"FCF"、"cash flow"、"经营现金流"、"XX的现金流怎么样"、"收现比"、"净现比"、"资本开支"、"现金流量表"、"现金流分析"、"现金流质量"时触发。MUST USE when user asks about cash flow analysis, free cash flow (FCF), operating cash flow quality, or cash conversion ratio. 深度分析企业三大现金流（经营/投资/筹资）质量，计算自由现金流，评估盈利含金量和资本配置效率。支持研报风格（formal）和快速分析风格（brief）。
.claude/skills/aifinlab-a-share-cash-flow/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-06 | ✓→✗ | ▼ Worse | -47% | 0% |
训练数据中的财务数据已过期。 每次执行时必须：
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 财务指标（含现金流量表核心字段） python "$SCRIPTS/cn_stock_data.py" finance --code [CODE] # 近 2 年日线行情（分析股价走势 vs 现金流趋势） python "$SCRIPTS/cn_stock_data.py" kline --code [CODE] --freq daily --start [2年前日期] # 实时行情（当前市值，用于计算 FCF Yield） python "$SCRIPTS/cn_stock_data.py" quote --code [CODE]
详见 references/cash-flow-guide.md 的分析框架和判断标准。
参见 references/cash-flow-guide.md 的指标解读规则。
根据用户风格要求选择输出格式。
| 维度 | formal（研报风格） | brief（快速分析） | |------|-------------------|------------------| | 篇幅 | 6-10 页 .docx | 1-2 页 Markdown | | 标题 | "公司名现金流质量深度分析" | "XX 现金流速评" | | 结构 | 核心结论→经营CF分析→投资CF分析→FCF评估→资本配置→风险提示→附表 | 关键指标→现金流画像→FCF→风险点 | | 图表 | 4-6 个（三大CF趋势、收现比/净现比、FCF、capex占比） | 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 |
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 | 18,396 | 16,096 | -13% | 1 | 1 | 0% | 2,540 | 1,920 | -24% | 0 | 0 | — |
case-02 | fail→fail | 31,546 | 8,228 | -74% | 1 | 1 | 0% | 5,316 | 1,952 | -63% | 0 | 0 | — |
case-03 | fail→fail | 16,837 | 9,758 | -42% | 1 | 1 | 0% | 2,969 | 1,866 | -37% | 0 | 0 | — |
case-04 | fail→fail | 17,841 | 9,304 | -48% | 1 | 1 | 0% | 2,874 | 1,732 | -40% | 0 | 0 | — |
case-05 | fail→fail | 32,585 | 11,643 | -64% | 1 | 1 | 0% | 4,821 | 1,933 | -60% | 0 | 0 | — |
case-06 | pass→fail | 20,839 | 10,196 | -51% | 1 | 1 | 0% | 3,264 | 1,723 | -47% | 0 | 0 | — |
case-22 | pass→fail | 17,179 | 9,960 | -42% | 1 | 1 | 0% | 2,750 | 1,718 | -38% | 0 | 0 | — |
case-07 | pass→pass | 17,825 | 16,259 | -9% | 1 | 1 | 0% | 2,452 | 3,718 | +52% | 0 | 0 | — |
case-08 | pass→fail | 24,137 | 9,609 | -60% | 1 | 1 | 0% | 3,396 | 1,641 | -52% | 0 | 0 | — |
case-09 | fail→fail | 20,855 | 27,279 | +31% | 1 | 1 | 0% | 2,847 | 3,831 | +35% | 0 | 0 | — |
case-10 | pass→pass | 15,279 | 10,359 | -32% | 1 | 1 | 0% | 2,439 | 2,882 | +18% | 0 | 0 | — |
case-11 | pass→pass | 12,053 | 8,129 | -33% | 1 | 1 | 0% | 1,947 | 2,505 | +29% | 0 | 0 | — |
case-12 | pass→fail | 19,805 | 11,106 | -44% | 1 | 1 | 0% | 2,799 | 1,636 | -42% | 0 | 0 | — |
case-13 | pass→fail | 19,383 | 9,611 | -50% | 1 | 1 | 0% | 3,139 | 1,593 | -49% | 0 | 0 | — |
case-14 | pass→fail | 18,887 | 16,934 | -10% | 1 | 1 | 0% | 3,071 | 2,148 | -30% | 0 | 0 | — |
case-15 | pass→pass | 15,366 | 11,247 | -27% | 1 | 1 | 0% | 2,239 | 2,888 | +29% | 0 | 0 | — |
case-16 | pass→pass | 19,190 | 10,045 | -48% | 1 | 1 | 0% | 2,551 | 2,780 | +9% | 0 | 0 | — |
case-17 | fail→pass | 30,727 | 20,724 | -33% | 1 | 1 | 0% | 3,789 | 4,145 | +9% | 0 | 0 | — |
case-18 | pass→fail | 21,566 | 11,264 | -48% | 1 | 1 | 0% | 3,130 | 1,808 | -42% | 0 | 0 | — |
case-19 | fail→pass | 9,744 | 6,085 | -38% | 1 | 1 | 0% | 1,483 | 2,119 | +43% | 0 | 0 | — |
case-20 | fail→pass | 19,762 | 3,890 | -80% | 1 | 1 | 0% | 2,851 | 1,925 | -32% | 0 | 0 | — |
case-21 | fail→pass | 21,936 | 24,197 | +10% | 1 | 1 | 0% | 3,235 | 5,004 | +55% | 0 | 0 | — |
case-23 | pass→pass | 17,381 | 16,402 | -6% | 1 | 1 | 0% | 2,638 | 3,637 | +38% | 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. 23 cases were attempted, and 11 counted toward the lift figure. The other 12 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 -13 percentage points is the difference between those two pass rates over the 11 comparable cases. 10 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 | -5% |
Other measured skills in the registry, with their headline benchmark lift.