Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股分红/股息分析。当用户说"分红"、"股息"、"高股息"、"dividend"、"分红率"、"XX分红怎么样"、"高分红股推荐"、"红利策略"、"现金分红"、"股息率排名"时触发。分析个股分红历史和可持续性,或筛选高股息标的,评估红利投资策略。通过 cn-stock-data 获取财务数据和行情,结合 web 搜索获取分红方案和历史记录。支持机构红利研究风格(formal)和个人分红笔记风格(brief)。
.claude/skills/aifinlab-a-share-dividend/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✓→✗ | ▼ Worse | -45% | 0% |
| case-04 | ✓→✗ | ▼ Worse | -49% | 0% |
| case-05 | ✓→✗ | ▼ Worse | -45% | 0% |
| case-06 | ✓→✗ | ▼ Worse | -60% | 0% |
| case-18 | ✓→✗ | ▼ Worse | -53% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 个股财务指标(盈利/现金流 — 分红可持续性基础) python "$SCRIPTS/cn_stock_data.py" finance --code [CODE] # 个股行情(计算当前股息率) python "$SCRIPTS/cn_stock_data.py" quote --code [CODE] # K线(股价走势 vs 除权日) python "$SCRIPTS/cn_stock_data.py" kline --code [CODE] --freq daily --start [2年前日期]
补充 web 搜索(分红数据 cn-stock-data 不直接提供):
Step 1: 确定分析模式
Step 2: 数据获取
个股模式:
构建分红历史表: | 年度 | 每股派息(元) | 分红总额(亿) | 分红率(%) | 股息率(%) | 除权日 | |------|------------|------------|---------|---------|--------|
筛选模式:
Step 3: 分红质量分析
可持续性评估(核心):
分红意愿评估:
Step 4: 估值与收益分析
Step 5: 输出
| 维度 | formal(机构红利研究) | brief(个人分红笔记) | |------|---------------------|---------------------| | 篇幅 | 3-5 页 | 1 页 | | 分红历史 | 近 5-10 年完整表格 | 近 3 年关键数据 | | 可持续性 | 多维度详细评估 | 核心 2-3 项判断 | | 行业对比 | 同行分红对比表 | 一句话行业对比 | | 策略评估 | 红利因子历史表现分析 | 当前适合性判断 | | 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 | 32,478 | 9,618 | -70% | 1 | 1 | 0% | 4,576 | 1,657 | -64% | 0 | 0 | — |
case-02 | fail→fail | 25,528 | 14,913 | -42% | 1 | 1 | 0% | 3,911 | 1,858 | -52% | 0 | 0 | — |
case-03 | pass→fail | 19,545 | 7,922 | -59% | 1 | 1 | 0% | 2,862 | 1,587 | -45% | 0 | 0 | — |
case-04 | pass→fail | 27,019 | 13,032 | -52% | 1 | 1 | 0% | 3,535 | 1,794 | -49% | 0 | 0 | — |
case-05 | pass→fail | 28,608 | 10,504 | -63% | 1 | 1 | 0% | 3,654 | 1,994 | -45% | 0 | 0 | — |
case-06 | pass→fail | 26,224 | 42,310 | +61% | 1 | 1 | 0% | 3,864 | 1,546 | -60% | 0 | 0 | — |
case-07 | fail→fail | 18,740 | 11,991 | -36% | 1 | 1 | 0% | 2,737 | 1,548 | -43% | 0 | 0 | — |
case-08 | fail→fail | 24,542 | 9,951 | -59% | 1 | 1 | 0% | 3,148 | 1,826 | -42% | 0 | 0 | — |
case-09 | fail→fail | 27,151 | 13,133 | -52% | 1 | 1 | 0% | 4,077 | 1,797 | -56% | 0 | 0 | — |
case-17 | fail→fail | 19,082 | 14,374 | -25% | 1 | 1 | 0% | 2,830 | 1,646 | -42% | 0 | 0 | — |
case-18 | pass→fail | 33,894 | 14,836 | -56% | 1 | 1 | 0% | 4,598 | 2,173 | -53% | 0 | 0 | — |
case-10 | pass→fail | 14,542 | 55,960 | +285% | 1 | 1 | 0% | 2,578 | 1,747 | -32% | 0 | 0 | — |
case-11 | pass→pass | 16,243 | 14,911 | -8% | 1 | 1 | 0% | 2,242 | 3,426 | +53% | 0 | 0 | — |
case-12 | pass→fail | 23,062 | 35,001 | +52% | 1 | 1 | 0% | 2,952 | 1,717 | -42% | 0 | 0 | — |
case-13 | pass→fail | 23,551 | 14,331 | -39% | 1 | 1 | 0% | 3,299 | 1,851 | -44% | 0 | 0 | — |
case-14 | pass→fail | 23,215 | 10,796 | -53% | 1 | 1 | 0% | 3,203 | 1,618 | -49% | 0 | 0 | — |
case-15 | pass→fail | 19,085 | 13,931 | -27% | 1 | 1 | 0% | 2,653 | 1,929 | -27% | 0 | 0 | — |
case-16 | pass→pass | 8,643 | 11,268 | +30% | 1 | 1 | 0% | 1,433 | 2,716 | +90% | 0 | 0 | — |
case-19 | pass→pass | 16,509 | 18,446 | +12% | 1 | 1 | 0% | 2,347 | 4,009 | +71% | 0 | 0 | — |
case-20 | pass→pass | 23,243 | 77,074 | +232% | 1 | 1 | 0% | 2,866 | 3,774 | +32% | 0 | 0 | — |
case-21 | pass→pass | 20,114 | 17,088 | -15% | 1 | 1 | 0% | 2,616 | 3,672 | +40% | 0 | 0 | — |
case-22 | pass→pass | 21,025 | 19,584 | -7% | 1 | 1 | 0% | 2,466 | 3,957 | +60% | 0 | 0 | — |
case-23 | pass→pass | 18,054 | 23,977 | +33% | 1 | 1 | 0% | 2,596 | 3,131 | +21% | 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 7 counted toward the lift figure. The other 16 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. A headline lift is not published for this run.
Other measured skills in the registry, with their headline benchmark lift.