Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股日内模式/分时走势量化分析。当用户说"日内模式"、"intraday"、"分时"、"盘中走势"、"几点涨"、"尾盘规律"时触发。量化分析A股日内交易模式。支持formal和brief风格。
.claude/skills/aifinlab-a-share-intraday-pattern/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -58% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -47% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 2% | 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]
各时段（开盘/早盘/午盘/尾盘）的平均涨跌幅
各时段成交量占比和变化规律
| 维度 | formal | brief | |------|--------|-------| | 时段分析 | 各时段详细统计 | 关键时段 | | 量能分布 | 分时量能图 | U型特征 | | 规律总结 | 历史统计验证 | 今日模式 | 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 | pass→fail | 18,770 | 25,085 | +34% | 1 | 1 | 0% | 3,007 | 1,022 | -66% | 0 | 0 | — |
case-02 | fail→fail | 25,772 | 9,960 | -61% | 1 | 1 | 0% | 3,855 | 1,027 | -73% | 0 | 0 | — |
case-03 | fail→fail | 21,766 | 7,611 | -65% | 1 | 1 | 0% | 3,415 | 905 | -73% | 0 | 0 | — |
case-04 | fail→pass | 27,449 | 28,983 | +6% | 1 | 1 | 0% | 4,171 | 4,283 | +3% | 0 | 0 | — |
case-05 | fail→fail | 19,311 | 18,234 | -6% | 1 | 1 | 0% | 3,120 | 3,387 | +9% | 0 | 0 | — |
case-06 | fail→fail | 27,553 | 24,473 | -11% | 1 | 1 | 0% | 4,231 | 4,665 | +10% | 0 | 0 | — |
case-07 | fail→fail | 23,902 | 20,257 | -15% | 1 | 1 | 0% | 3,698 | 3,579 | -3% | 0 | 0 | — |
case-08 | fail→fail | 19,894 | 21,484 | +8% | 1 | 1 | 0% | 3,030 | 3,927 | +30% | 0 | 0 | — |
case-09 | fail→pass | 25,538 | 7,989 | -69% | 1 | 1 | 0% | 4,545 | 1,907 | -58% | 0 | 0 | — |
case-14 | fail→pass | 23,268 | 7,741 | -67% | 1 | 1 | 0% | 3,744 | 1,903 | -49% | 0 | 0 | — |
case-10 | fail→fail | 25,998 | 17,399 | -33% | 1 | 1 | 0% | 3,134 | 3,212 | +2% | 0 | 0 | — |
case-11 | fail→fail | 21,553 | 18,551 | -14% | 1 | 1 | 0% | 3,385 | 3,661 | +8% | 0 | 0 | — |
case-12 | pass→pass | 23,098 | 17,943 | -22% | 1 | 1 | 0% | 3,707 | 3,442 | -7% | 0 | 0 | — |
case-13 | fail→pass | 23,342 | 6,631 | -72% | 1 | 1 | 0% | 3,010 | 1,584 | -47% | 0 | 0 | — |
case-15 | fail→fail | 20,158 | 21,261 | +5% | 1 | 1 | 0% | 3,161 | 3,771 | +19% | 0 | 0 | — |
case-16 | fail→pass | 21,876 | 17,107 | -22% | 1 | 1 | 0% | 3,318 | 3,400 | +2% | 0 | 0 | — |
case-17 | fail→fail | 27,749 | 24,747 | -11% | 1 | 1 | 0% | 3,654 | 4,500 | +23% | 0 | 0 | — |
case-18 | pass→pass | 19,143 | 18,698 | -2% | 1 | 1 | 0% | 2,945 | 3,442 | +17% | 0 | 0 | — |
case-19 | pass→pass | 15,050 | 17,680 | +17% | 1 | 1 | 0% | 2,317 | 2,390 | +3% | 0 | 0 | — |
case-20 | pass→fail | 34,384 | 7,478 | -78% | 1 | 1 | 0% | 6,776 | 958 | -86% | 0 | 0 | — |
case-21 | pass→fail | 22,732 | 12,625 | -44% | 1 | 1 | 0% | 3,079 | 1,432 | -53% | 0 | 0 | — |
case-22 | pass→pass | 27,411 | 27,564 | +1% | 1 | 1 | 0% | 4,914 | 5,469 | +11% | 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 17 counted toward the lift figure. The other 5 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 +9 percentage points is the difference between those two pass rates over the 17 comparable cases. 4 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.