Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股期权定价/BSM模型分析。当用户说"期权定价"、"BSM"、"Black-Scholes"、"期权估值"、"理论价格"、"定价模型"、"二叉树定价"时触发。基于 cn-stock-data 获取数据，进行期权定价与估值分析。支持 formal/brief 两种输出风格。
.claude/skills/aifinlab-a-share-option-pricing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-11 | ✓→✓ | ＝ Same ✓ | 63% | 0% |
| case-10 | ✓→✓ | ＝ Same ✓ | 67% | 0% |
通过 cn-stock-data skill 获取数据：
# [标的] 期权定价分析报告
## 一、BSM定价
| 合约 | 市场价 | 理论价 | 偏差 |
|------|--------|--------|------|
## 二、定价偏差
[偏差分析、套利机会]
## 三、高级模型
[Heston/SABR定价对比]
## 四、交易建议## [标的] 定价速览
- BSM理论价 0.285，市场价 0.295
- 溢价 +3.5%，流动性溢价为主
- Put-Call Parity偏差 0.2%，无套利
- Heston定价 0.290，更接近市场参考 references/option-pricing-guide.md 获取详细方法论与 A股实证研究。
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-11 | pass→pass | 5,358 | 5,564 | +4% | 1 | 1 | 0% | 1,041 | 1,693 | +63% | 0 | 0 | — |
case-10 | pass→pass | 12,997 | 12,408 | -5% | 1 | 1 | 0% | 1,798 | 2,996 | +67% | 0 | 0 | — |
case-01 | fail→pass | 20,970 | 37,161 | +77% | 1 | 1 | 0% | 3,206 | 3,023 | -6% | 0 | 0 | — |
case-02 | fail→fail | 40,587 | 41,435 | +2% | 1 | 1 | 0% | 5,989 | 8,912 | +49% | 0 | 0 | — |
case-03 | fail→pass | 26,168 | 29,040 | +11% | 1 | 1 | 0% | 4,757 | 5,989 | +26% | 0 | 0 | — |
case-04 | pass→pass | 12,165 | 12,926 | +6% | 1 | 1 | 0% | 1,990 | 2,714 | +36% | 0 | 0 | — |
case-05 | pass→pass | 14,213 | 20,350 | +43% | 1 | 1 | 0% | 2,175 | 3,243 | +49% | 0 | 0 | — |
case-06 | pass→pass | 17,796 | 17,806 | +0% | 1 | 1 | 0% | 2,838 | 3,226 | +14% | 0 | 0 | — |
case-07 | pass→pass | 30,429 | 13,135 | -57% | 1 | 1 | 0% | 1,809 | 2,596 | +44% | 0 | 0 | — |
case-08 | pass→pass | 11,510 | 12,203 | +6% | 1 | 1 | 0% | 1,892 | 2,698 | +43% | 0 | 0 | — |
case-09 | pass→pass | 9,937 | 10,822 | +9% | 1 | 1 | 0% | 1,653 | 2,544 | +54% | 0 | 0 | — |
case-12 | pass→pass | 44,130 | 22,088 | -50% | 1 | 1 | 0% | 2,934 | 3,672 | +25% | 0 | 0 | — |
case-13 | pass→pass | 23,317 | 24,499 | +5% | 1 | 1 | 0% | 3,360 | 3,984 | +19% | 0 | 0 | — |
case-14 | pass→pass | 15,332 | 17,734 | +16% | 1 | 1 | 0% | 2,097 | 3,135 | +49% | 0 | 0 | — |
case-15 | fail→fail | 48,348 | 68,970 | +43% | 1 | 1 | 0% | 8,230 | 8,822 | +7% | 0 | 0 | — |
case-16 | fail→pass | 18,094 | 17,089 | -6% | 1 | 1 | 0% | 3,019 | 3,769 | +25% | 0 | 0 | — |
case-17 | fail→fail | 22,158 | 17,444 | -21% | 1 | 1 | 0% | 3,456 | 4,099 | +19% | 0 | 0 | — |
case-18 | pass→pass | 26,421 | 28,084 | +6% | 1 | 1 | 0% | 4,985 | 6,475 | +30% | 0 | 0 | — |
case-19 | fail→fail | 31,014 | 38,958 | +26% | 1 | 1 | 0% | 5,699 | 6,464 | +13% | 0 | 0 | — |
case-20 | fail→fail | 26,521 | 41,688 | +57% | 1 | 1 | 0% | 5,209 | 8,884 | +71% | 0 | 0 | — |
case-21 | pass→pass | 19,147 | 19,985 | +4% | 1 | 1 | 0% | 2,895 | 3,349 | +16% | 0 | 0 | — |
case-22 | pass→pass | 17,947 | 21,695 | +21% | 1 | 1 | 0% | 2,287 | 3,997 | +75% | 0 | 0 | — |
case-23 | pass→pass | 19,924 | 21,766 | +9% | 1 | 1 | 0% | 3,103 | 3,505 | +13% | 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. The headline lift of +13 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is 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 | +9% |
Other measured skills in the registry, with their headline benchmark lift.