Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股期权策略分析。当用户说"期权"、"option"、"认购"、"认沽"、"50ETF期权"、"300ETF期权"、"隐含波动率"、"IV"、"Greeks"、"期权策略"、"备兑"、"保险策略"、"跨式"、"宽跨式"时触发。分析A股期权(上证50ETF/沪深300ETF/中证500ETF/个股期权)的隐含波动率、Greeks指标、期权策略构建,辅助风险管理和增强收益决策。支持研报风格(formal)和快速分析风格(brief)。
.claude/skills/aifinlab-a-share-option-strategy/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-16 | ✗→✓ | ▲ Improved | 206% | 0% |
| case-04 | ✓→✗ | ▼ Worse | 2% | 0% |
| case-05 | ✓→✗ | ▼ Worse | -8% | 0% |
| case-06 | ✓→✗ | ▼ Worse | 32% | 0% |
| case-13 | ✓→✗ | ▼ Worse | 29% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 标的ETF实时行情 python "$SCRIPTS/cn_stock_data.py" quote --code SH510050,SH510300,SH510500 # SH510050=50ETF, SH510300=300ETF, SH510500=500ETF # 标的ETF日线K线(近60个交易日,用于历史波动率计算) python "$SCRIPTS/cn_stock_data.py" kline --code SH510050 --freq daily --start [60日前日期]
python# 期权行情数据(akshare) import akshare as ak # 50ETF期权 — 标的现货行情+期权链 df_spot = ak.option_sse_underlying_spot_em(symbol='510050') # 50ETF期权标的 df_spot_300 = ak.option_sse_underlying_spot_em(symbol='510300') # 300ETF期权标的 # 期权合约列表(按到期月份) df_list = ak.option_sse_list_sina(symbol='510050', exchange='null') # 期权实时行情(东方财富) df_call = ak.option_sse_spot_price_sina(symbol='510050') # 认购+认沽全部合约 # Greeks 数据 df_greeks = ak.option_sse_greeks_em(symbol='510050') # Delta/Gamma/Theta/Vega/IV # 300ETF期权 df_300_greeks = ak.option_sse_greeks_em(symbol='510300')
web 搜索补充(数据源无法覆盖的关键信息):
Step 1: 期权链数据获取 获取标的ETF现价及期权链数据: | 标的 | ETF代码 | 交易所 | 合约单位 | 行权方式 | |------|--------|--------|---------|---------| | 上证50ETF | 510050 | 上交所 | 10000份 | 欧式 | | 沪深300ETF(沪) | 510300 | 上交所 | 10000份 | 欧式 | | 沪深300ETF(深) | 159919 | 深交所 | 10000份 | 欧式 | | 中证500ETF | 510500 | 上交所 | 10000份 | 欧式 |
获取内容:
Step 2: 隐含波动率(IV)分析
Step 3: 策略推荐 根据用户市场观点推荐合适策略:
| 市场观点 | 推荐策略 | IV环境适配 | |---------|---------|-----------| | 强烈看涨 | 买入认购 / 牛市价差 | 低IV买购,高IV用价差 | | 温和看涨 | 卖出认沽 / 备兑策略 | 高IV卖沽收益更好 | | 强烈看跌 | 买入认沽 / 熊市价差 | 低IV买沽,高IV用价差 | | 温和看跌 | 卖出认购 / 熊市价差 | 高IV卖购收益更好 | | 震荡/中性 | 卖出跨式/宽跨式 / 铁鹰 | 高IV最适合 | | 突破预期 | 买入跨式/宽跨式 | 低IV时买入成本低 | | 持仓保护 | 保险策略(持股+买沽) | 低IV时保险成本低 | | 增强收益 | 备兑策略(持股+卖购) | 高IV时权利金更厚 |
对每个推荐策略给出:
Step 4: 盈亏分析描述 用文字描述到期盈亏结构:
Step 5: 输出
| 维度 | formal(策略报告) | brief(快速分析) | |------|------------------|-----------------| | 标的覆盖 | 50ETF/300ETF/500ETF 全分析 | 仅用户指定标的 | | IV 分析 | IV水平+偏度+期限结构+HV对比+PCR | IV百分位+高/低判断 | | 策略推荐 | 2-3个策略详细对比+Greeks暴露 | 1个最优策略+关键参数 | | 盈亏分析 | 完整盈亏描述+关键价位表 | 盈亏平衡点+最大亏损 | | Greeks | Delta/Gamma/Theta/Vega 完整解读 | Delta+Theta 方向 | | 风险提示 | 详细风险分析+保证金要求 | 一句话风险提示 | | 免责声明 | 需要(衍生品高风险) | 不需要 |
默认风格:brief
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-03 | fail→fail | 13,874 | 63,655 | +359% | 1 | 1 | 0% | 2,317 | 4,688 | +102% | 0 | 0 | — |
case-01 | fail→fail | 13,411 | 35,256 | +163% | 1 | 1 | 0% | 2,375 | 2,488 | +5% | 0 | 0 | — |
case-02 | fail→fail | 44,549 | 17,141 | -62% | 1 | 1 | 0% | 6,842 | 3,696 | -46% | 0 | 0 | — |
case-04 | pass→fail | 15,692 | 11,676 | -26% | 1 | 1 | 0% | 2,591 | 2,643 | +2% | 0 | 0 | — |
case-05 | pass→fail | 22,395 | 8,557 | -62% | 1 | 1 | 0% | 2,831 | 2,602 | -8% | 0 | 0 | — |
case-06 | pass→fail | 16,409 | 12,403 | -24% | 1 | 1 | 0% | 2,337 | 3,081 | +32% | 0 | 0 | — |
case-07 | pass→pass | 18,717 | 33,786 | +81% | 1 | 1 | 0% | 2,494 | 6,276 | +152% | 0 | 0 | — |
case-13 | pass→fail | 19,889 | 9,183 | -54% | 1 | 1 | 0% | 2,017 | 2,610 | +29% | 0 | 0 | — |
case-08 | pass→pass | 16,696 | 10,019 | -40% | 1 | 1 | 0% | 2,692 | 3,571 | +33% | 0 | 0 | — |
case-09 | pass→fail | 12,396 | 12,633 | +2% | 1 | 1 | 0% | 1,681 | 3,114 | +85% | 0 | 0 | — |
case-10 | pass→pass | 9,111 | 24,822 | +172% | 1 | 1 | 0% | 1,578 | 4,867 | +208% | 0 | 0 | — |
case-11 | pass→pass | 12,452 | 15,489 | +24% | 1 | 1 | 0% | 1,956 | 3,783 | +93% | 0 | 0 | — |
case-12 | pass→pass | 16,602 | 26,551 | +60% | 1 | 1 | 0% | 2,528 | 4,856 | +92% | 0 | 0 | — |
case-14 | pass→fail | 8,475 | 9,912 | +17% | 1 | 1 | 0% | 1,362 | 2,628 | +93% | 0 | 0 | — |
case-15 | pass→fail | 17,717 | 25,621 | +45% | 1 | 1 | 0% | 2,210 | 5,956 | +170% | 0 | 0 | — |
case-16 | fail→pass | 7,077 | 4,763 | -33% | 1 | 1 | 0% | 921 | 2,820 | +206% | 0 | 0 | — |
case-17 | pass→pass | 6,211 | 5,253 | -15% | 1 | 1 | 0% | 849 | 3,116 | +267% | 0 | 0 | — |
case-18 | pass→fail | 20,893 | 49,325 | +136% | 1 | 1 | 0% | 3,728 | 11,255 | +202% | 0 | 0 | — |
case-19 | pass→fail | 20,465 | 8,525 | -58% | 1 | 1 | 0% | 2,681 | 2,579 | -4% | 0 | 0 | — |
case-20 | pass→fail | 20,343 | 11,274 | -45% | 1 | 1 | 0% | 3,088 | 2,916 | -6% | 0 | 0 | — |
case-21 | pass→fail | 15,803 | 14,700 | -7% | 1 | 1 | 0% | 2,238 | 2,438 | +9% | 0 | 0 | — |
case-22 | fail→fail | 14,734 | 9,427 | -36% | 1 | 1 | 0% | 2,465 | 2,818 | +14% | 0 | 0 | — |
case-23 | pass→pass | 16,347 | 12,170 | -26% | 1 | 1 | 0% | 2,328 | 3,638 | +56% | 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 8 counted toward the lift figure. The other 15 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 -43 percentage points is the difference between those two pass rates over the 8 comparable cases. 11 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.