Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股可转债分析/转债策略/可转债估值与筛选。当用户提到"可转债"或"转债"就触发。关键词："可转债"、"转债"、"convertible bond"、"转股"、"转股价值"、"双低策略"、"强赎"、"XX转债怎么样"、"可转债打新"、"转债溢价率"、"转债分析"、"可转债筛选"、"转债到期收益率"、"YTM"、"下修"、"转债回售"。MUST USE when user mentions 可转债 or 转债 in any context — this is the ONLY skill for convertible bond analysis. 基于 efinance 可转债数据和 cn-stock-data 正股数据，分析转债估值（溢价率/到期收益率）、正股联动、强赎风险、转债策略。支持研报风格（formal）和快速解读风格（brief）。
.claude/skills/aifinlab-a-share-convertible-bond/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-19 | ✗→✓ | ▲ Improved | -46% | 0% |
| case-12 | ✓→✗ | ▼ Worse | 54% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 6% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 45% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 31% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 可转债实时行情（通过 akshare，cn-stock-data 不支持此命令） # 方式1: akshare 集思录可转债数据（推荐，含溢价率/YTM/双低值等完整指标） python -c "import akshare as ak; df=ak.bond_cb_jsl(); print(df.to_json(orient='records', force_ascii=False))" # 方式2: efinance 可转债行情（备选，数据源可能不稳定） python -c "import efinance as ef; df=ef.bond.get_realtime_quotes(); print(df.to_json(orient='records', force_ascii=False))" # 正股行情（通过 cn-stock-data） python "$SCRIPTS/cn_stock_data.py" quote --code [正股代码] # 正股K线 python "$SCRIPTS/cn_stock_data.py" kline --code [正股代码] --freq daily --start [日期] # 正股财务 python "$SCRIPTS/cn_stock_data.py" finance --code [正股代码]
补充：通过 web 搜索获取转股价调整公告、强赎公告等最新信息。
根据用户意图选择模式：
convertible_bond 获取全市场转债行情（价格/溢价率/转股价值/到期收益率等）quote / kline / finance 获取正股数据核心指标计算（详见 references/convertible-bond-guide.md）：
根据风格要求输出：
| 维度 | formal（研报风格） | brief（快速解读） | |------|-------------------|-------------------| | 输出 | 完整转债研究报告 | 快速要点总结 | | 估值分析 | 全面（纯债+期权+Greeks） | 核心指标（溢价率+YTM+双低值） | | 正股分析 | 详细基本面分析 | 仅关键财务指标 | | 策略建议 | 情景分析+多空因素 | 直接结论 |
默认风格：brief。用户要求"详细分析"/"研报"时切换为 formal。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 29,502 | 10,143 | -66% | 1 | 1 | 0% | 3,814 | 1,685 | -56% | 0 | 0 | — |
case-02 | fail→fail | 21,314 | 10,958 | -49% | 1 | 1 | 0% | 3,267 | 1,577 | -52% | 0 | 0 | — |
case-03 | fail→fail | 24,840 | 10,883 | -56% | 1 | 1 | 0% | 3,327 | 1,469 | -56% | 0 | 0 | — |
case-04 | pass→pass | 24,671 | 24,916 | +1% | 1 | 1 | 0% | 4,269 | 4,542 | +6% | 0 | 0 | — |
case-05 | pass→pass | 24,430 | 24,390 | -0% | 1 | 1 | 0% | 3,142 | 4,567 | +45% | 0 | 0 | — |
case-11 | fail→fail | 49,171 | 37,522 | -24% | 1 | 1 | 0% | 3,492 | 5,153 | +48% | 0 | 0 | — |
case-22 | fail→fail | 28,503 | 9,918 | -65% | 1 | 1 | 0% | 3,898 | 1,532 | -61% | 0 | 0 | — |
case-06 | pass→pass | 29,749 | 28,584 | -4% | 1 | 1 | 0% | 3,890 | 5,093 | +31% | 0 | 0 | — |
case-07 | pass→pass | 27,816 | 31,148 | +12% | 1 | 1 | 0% | 3,880 | 5,228 | +35% | 0 | 0 | — |
case-08 | pass→pass | 14,069 | 9,322 | -34% | 1 | 1 | 0% | 1,905 | 2,259 | +19% | 0 | 0 | — |
case-09 | fail→fail | 27,559 | 12,612 | -54% | 1 | 1 | 0% | 4,746 | 1,531 | -68% | 0 | 0 | — |
case-10 | fail→fail | 12,750 | 9,096 | -29% | 1 | 1 | 0% | 2,118 | 2,574 | +22% | 0 | 0 | — |
case-12 | pass→fail | 20,862 | 23,358 | +12% | 1 | 1 | 0% | 2,691 | 4,145 | +54% | 0 | 0 | — |
case-13 | fail→fail | 23,835 | 22,426 | -6% | 1 | 1 | 0% | 3,046 | 4,113 | +35% | 0 | 0 | — |
case-14 | pass→pass | 13,758 | 16,253 | +18% | 1 | 1 | 0% | 2,086 | 3,107 | +49% | 0 | 0 | — |
case-15 | fail→fail | 35,270 | 52,875 | +50% | 1 | 1 | 0% | 5,251 | 9,567 | +82% | 0 | 0 | — |
case-16 | pass→pass | 9,867 | 6,551 | -34% | 1 | 1 | 0% | 1,503 | 2,016 | +34% | 0 | 0 | — |
case-17 | pass→pass | 39,436 | 31,578 | -20% | 1 | 1 | 0% | 3,428 | 4,244 | +24% | 0 | 0 | — |
case-18 | pass→pass | 20,841 | 19,182 | -8% | 1 | 1 | 0% | 2,601 | 3,753 | +44% | 0 | 0 | — |
case-19 | fail→pass | 16,417 | 4,464 | -73% | 1 | 1 | 0% | 2,889 | 1,567 | -46% | 0 | 0 | — |
case-20 | pass→pass | 19,580 | 10,232 | -48% | 1 | 1 | 0% | 2,805 | 2,350 | -16% | 0 | 0 | — |
case-21 | fail→fail | 28,888 | 10,060 | -65% | 1 | 1 | 0% | 4,281 | 1,396 | -67% | 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 14 counted toward the lift figure. The other 8 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 0 percentage points is the difference between those two pass rates over the 14 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.