Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股系统性风险/传染效应分析。当用户说"系统性风险"、"systematic risk"、"传染效应"、"风险传染"、"系统风险"、"金融风险传导"时触发。基于 cn-stock-data 获取数据，评估系统性风险水平。支持 formal/brief 两种输出风格。
.claude/skills/aifinlab-a-share-systematic-risk/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-23 | ✓→✗ | ▼ Worse | 55% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 40% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 12% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 22% | 0% |
| case-07 | ✓→✓ | ＝ Same ✓ | 12% | 0% |
通过 cn-stock-data skill 获取数据：
# 系统性风险评估报告
## 一、风险指标
| 指标 | 数值 | 分位数 | 状态 |
|------|------|--------|------|
## 二、传染效应
[风险传导路径]
## 三、风险状态
[综合评估与历史对比]
## 四、防御建议## 系统性风险速览
- 综合风险指数 45 (P55)，中等
- 信用利差 85bp，正常
- 市场相关性 0.45，未见异常
- 状态：正常，无需特别防御参考 references/systematic-risk-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 | 36,721 | 31,626 | -14% | 1 | 1 | 0% | 5,413 | 4,702 | -13% | 0 | 0 | — |
case-02 | fail→fail | 11,407 | 9,553 | -16% | 1 | 1 | 0% | 1,782 | 2,059 | +16% | 0 | 0 | — |
case-03 | fail→fail | 38,216 | 44,327 | +16% | 1 | 1 | 0% | 5,123 | 6,461 | +26% | 0 | 0 | — |
case-04 | pass→pass | 19,015 | 19,482 | +2% | 1 | 1 | 0% | 2,392 | 3,357 | +40% | 0 | 0 | — |
case-05 | pass→pass | 21,875 | 19,537 | -11% | 1 | 1 | 0% | 3,323 | 3,720 | +12% | 0 | 0 | — |
case-06 | pass→pass | 28,271 | 32,872 | +16% | 1 | 1 | 0% | 4,680 | 5,719 | +22% | 0 | 0 | — |
case-07 | pass→pass | 20,322 | 17,841 | -12% | 1 | 1 | 0% | 3,154 | 3,533 | +12% | 0 | 0 | — |
case-08 | pass→pass | 20,911 | 26,461 | +27% | 1 | 1 | 0% | 3,149 | 4,314 | +37% | 0 | 0 | — |
case-09 | pass→pass | 19,633 | 25,365 | +29% | 1 | 1 | 0% | 2,753 | 4,352 | +58% | 0 | 0 | — |
case-14 | pass→pass | 21,086 | 24,452 | +16% | 1 | 1 | 0% | 3,259 | 3,712 | +14% | 0 | 0 | — |
case-10 | pass→pass | 17,214 | 18,845 | +9% | 1 | 1 | 0% | 2,569 | 3,338 | +30% | 0 | 0 | — |
case-11 | pass→pass | 20,439 | 27,225 | +33% | 1 | 1 | 0% | 3,174 | 4,484 | +41% | 0 | 0 | — |
case-12 | pass→pass | 20,206 | 22,866 | +13% | 1 | 1 | 0% | 2,878 | 3,876 | +35% | 0 | 0 | — |
case-13 | pass→pass | 19,224 | 12,927 | -33% | 1 | 1 | 0% | 2,985 | 2,569 | -14% | 0 | 0 | — |
case-15 | pass→pass | 13,701 | 8,395 | -39% | 1 | 1 | 0% | 2,039 | 1,706 | -16% | 0 | 0 | — |
case-16 | pass→pass | 22,562 | 15,197 | -33% | 1 | 1 | 0% | 2,883 | 3,042 | +6% | 0 | 0 | — |
case-17 | fail→fail | 7,106 | 7,104 | -0% | 1 | 1 | 0% | 1,139 | 1,868 | +64% | 0 | 0 | — |
case-18 | fail→fail | 18,739 | 17,003 | -9% | 1 | 1 | 0% | 2,890 | 3,127 | +8% | 0 | 0 | — |
case-19 | fail→fail | 24,838 | 20,246 | -18% | 1 | 1 | 0% | 3,711 | 3,620 | -2% | 0 | 0 | — |
case-20 | pass→pass | 10,610 | 13,443 | +27% | 1 | 1 | 0% | 1,710 | 2,663 | +56% | 0 | 0 | — |
case-21 | pass→pass | 28,766 | 39,861 | +39% | 1 | 1 | 0% | 4,750 | 8,851 | +86% | 0 | 0 | — |
case-22 | pass→pass | 25,529 | 20,689 | -19% | 1 | 1 | 0% | 3,607 | 4,264 | +18% | 0 | 0 | — |
case-23 | pass→fail | 24,107 | 29,703 | +23% | 1 | 1 | 0% | 3,385 | 5,231 | +55% | 0 | 0 | — |
case-24 | pass→pass | 24,910 | 21,695 | -13% | 1 | 1 | 0% | 3,340 | 4,181 | +25% | 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. 24 cases were attempted. The headline lift of -4 percentage points is the difference between those two pass rates over the 24 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Other measured skills in the registry, with their headline benchmark lift.