Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股量价异常检测/异动监控。当用户说"异常检测"、"异动"、"anomaly"、"量价异常"、"异常波动"、"XX有异动"时触发。量化检测股价和成交量异常。支持formal和brief风格。
.claude/skills/aifinlab-a-share-anomaly-detection/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -36% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" python "$SCRIPTS/cn_stock_data.py" kline --code [CODE] --freq daily --start [日期] python "$SCRIPTS/cn_stock_data.py" quote --code [CODE] python "$SCRIPTS/cn_stock_data.py" finance --code [CODE]
关联近期公告/新闻/资金流，尝试解释异常原因
| 维度 | formal | brief | |------|--------|-------| | 异常信号 | 完整异常事件列表 | 最新异常 | | 统计分析 | Z-score+分布 | 异常等级 | | 归因 | 可能原因分析 | 一句话 | 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-01 | fail→fail | 12,473 | 13,093 | +5% | 1 | 1 | 0% | 1,705 | 1,070 | -37% | 0 | 0 | — |
case-02 | fail→fail | 29,563 | 7,589 | -74% | 1 | 1 | 0% | 4,111 | 981 | -76% | 0 | 0 | — |
case-03 | fail→fail | 19,522 | 8,430 | -57% | 1 | 1 | 0% | 2,985 | 993 | -67% | 0 | 0 | — |
case-04 | fail→pass | 23,153 | 24,767 | +7% | 1 | 1 | 0% | 3,771 | 4,284 | +14% | 0 | 0 | — |
case-05 | pass→pass | 16,597 | 29,144 | +76% | 1 | 1 | 0% | 3,028 | 3,945 | +30% | 0 | 0 | — |
case-06 | fail→pass | 24,574 | 50,860 | +107% | 1 | 1 | 0% | 3,490 | 3,551 | +2% | 0 | 0 | — |
case-07 | pass→pass | 22,297 | 22,775 | +2% | 1 | 1 | 0% | 3,378 | 3,817 | +13% | 0 | 0 | — |
case-08 | pass→fail | 21,094 | 10,374 | -51% | 1 | 1 | 0% | 2,873 | 1,145 | -60% | 0 | 0 | — |
case-09 | fail→pass | 21,184 | 21,681 | +2% | 1 | 1 | 0% | 3,550 | 3,798 | +7% | 0 | 0 | — |
case-10 | pass→pass | 26,208 | 21,010 | -20% | 1 | 1 | 0% | 3,847 | 3,934 | +2% | 0 | 0 | — |
case-11 | pass→pass | 24,337 | 21,780 | -11% | 1 | 1 | 0% | 3,425 | 3,541 | +3% | 0 | 0 | — |
case-12 | pass→fail | 18,278 | 9,189 | -50% | 1 | 1 | 0% | 2,349 | 918 | -61% | 0 | 0 | — |
case-13 | pass→fail | 29,441 | 9,725 | -67% | 1 | 1 | 0% | 4,390 | 1,203 | -73% | 0 | 0 | — |
case-14 | pass→pass | 20,629 | 19,549 | -5% | 1 | 1 | 0% | 2,739 | 3,324 | +21% | 0 | 0 | — |
case-15 | pass→fail | 21,210 | 8,115 | -62% | 1 | 1 | 0% | 2,711 | 1,615 | -40% | 0 | 0 | — |
case-16 | pass→pass | 22,279 | 26,438 | +19% | 1 | 1 | 0% | 3,125 | 3,385 | +8% | 0 | 0 | — |
case-17 | fail→pass | 13,418 | 3,477 | -74% | 1 | 1 | 0% | 2,607 | 1,174 | -55% | 0 | 0 | — |
case-18 | fail→pass | 11,915 | 4,699 | -61% | 1 | 1 | 0% | 1,860 | 1,192 | -36% | 0 | 0 | — |
case-19 | fail→pass | 12,705 | 4,070 | -68% | 1 | 1 | 0% | 2,129 | 1,304 | -39% | 0 | 0 | — |
case-20 | pass→pass | 20,138 | 23,461 | +17% | 1 | 1 | 0% | 3,107 | 3,484 | +12% | 0 | 0 | — |
case-21 | pass→pass | 32,249 | 31,873 | -1% | 1 | 1 | 0% | 6,389 | 7,191 | +13% | 0 | 0 | — |
case-22 | fail→pass | 24,814 | 25,339 | +2% | 1 | 1 | 0% | 3,952 | 4,607 | +17% | 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 16 counted toward the lift figure. The other 6 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 +14 percentage points is the difference between those two pass rates over the 16 comparable cases. 6 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/28/2026 | +5% |
Other measured skills in the registry, with their headline benchmark lift.