Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股商品-股票联动/通胀交易策略。当用户说"商品股票联动"、"通胀交易"、"commodity equity"、"油价影响"、"铜价联动"、"资源股"、"周期股联动"时触发。基于 cn-stock-data 获取数据，分析商品与股票的联动关系。支持 formal/brief 两种输出风格。
.claude/skills/aifinlab-a-share-commodity-equity/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✓→✓ | ＝ Same ✓ | 25% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 33% | 0% |
| case-07 | ✓→✓ | ＝ Same ✓ | 12% | 0% |
| case-08 | ✓→✓ | ＝ Same ✓ | 16% | 0% |
| case-09 | ✓→✓ | ＝ Same ✓ | 55% | 0% |
通过 cn-stock-data skill 获取数据：
# 商品-股票联动分析报告
## 一、联动关系
| 商品 | 相关股票 | 相关性 | 弹性 |
|------|---------|--------|------|
## 二、通胀传导
[当前传导阶段]
## 三、交易信号
[商品趋势→股票配置建议]
## 四、风险提示## 商品-股票联动速览
- 铜价近20日涨 +5%，弹性系数1.2
- 紫金矿业预期涨幅 +6%
- PPI回升，上游资源股受益
- 建议：超配有色/石化板块参考 references/commodity-equity-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-01 | fail→fail | 27,601 | 28,611 | +4% | 1 | 1 | 0% | 4,133 | 3,735 | -10% | 0 | 0 | — |
case-02 | fail→fail | 19,465 | 14,134 | -27% | 1 | 1 | 0% | 2,937 | 2,781 | -5% | 0 | 0 | — |
case-03 | fail→fail | 26,444 | 19,011 | -28% | 1 | 1 | 0% | 3,873 | 3,624 | -6% | 0 | 0 | — |
case-04 | pass→pass | 22,959 | 22,763 | -1% | 1 | 1 | 0% | 3,119 | 3,893 | +25% | 0 | 0 | — |
case-05 | fail→fail | 23,456 | 24,008 | +2% | 1 | 1 | 0% | 3,389 | 4,223 | +25% | 0 | 0 | — |
case-06 | pass→pass | 9,502 | 9,914 | +4% | 1 | 1 | 0% | 1,798 | 2,385 | +33% | 0 | 0 | — |
case-07 | pass→pass | 23,072 | 19,303 | -16% | 1 | 1 | 0% | 3,138 | 3,529 | +12% | 0 | 0 | — |
case-08 | pass→pass | 21,898 | 20,852 | -5% | 1 | 1 | 0% | 2,796 | 3,256 | +16% | 0 | 0 | — |
case-09 | pass→pass | 25,842 | 38,523 | +49% | 1 | 1 | 0% | 3,814 | 5,919 | +55% | 0 | 0 | — |
case-10 | pass→pass | 20,687 | 25,270 | +22% | 1 | 1 | 0% | 3,191 | 4,413 | +38% | 0 | 0 | — |
case-11 | pass→pass | 17,984 | 20,897 | +16% | 1 | 1 | 0% | 2,717 | 3,926 | +44% | 0 | 0 | — |
case-12 | pass→pass | 23,183 | 26,113 | +13% | 1 | 1 | 0% | 3,476 | 4,555 | +31% | 0 | 0 | — |
case-13 | pass→pass | 18,423 | 22,340 | +21% | 1 | 1 | 0% | 2,803 | 3,859 | +38% | 0 | 0 | — |
case-14 | pass→pass | 22,001 | 22,180 | +1% | 1 | 1 | 0% | 3,199 | 3,476 | +9% | 0 | 0 | — |
case-15 | pass→pass | 23,834 | 24,792 | +4% | 1 | 1 | 0% | 2,836 | 3,768 | +33% | 0 | 0 | — |
case-16 | fail→fail | 19,535 | 15,376 | -21% | 1 | 1 | 0% | 2,791 | 3,143 | +13% | 0 | 0 | — |
case-17 | fail→fail | 27,184 | 21,254 | -22% | 1 | 1 | 0% | 3,767 | 3,893 | +3% | 0 | 0 | — |
case-18 | fail→fail | 22,934 | 23,030 | +0% | 1 | 1 | 0% | 3,331 | 3,972 | +19% | 0 | 0 | — |
case-19 | fail→fail | 17,527 | 10,177 | -42% | 1 | 1 | 0% | 2,312 | 2,073 | -10% | 0 | 0 | — |
case-20 | pass→pass | 22,981 | 26,633 | +16% | 1 | 1 | 0% | 3,299 | 4,611 | +40% | 0 | 0 | — |
case-21 | pass→pass | 25,567 | 34,550 | +35% | 1 | 1 | 0% | 5,038 | 6,214 | +23% | 0 | 0 | — |
case-22 | pass→pass | 19,132 | 23,550 | +23% | 1 | 1 | 0% | 4,227 | 5,102 | +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. 22 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/28/2026 | +17% |
Other measured skills in the registry, with their headline benchmark lift.