Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股盈利动量/业绩趋势量化分析。当用户说"盈利动量"、"earnings momentum"、"业绩趋势"、"盈利加速"、"业绩改善"时触发。量化分析盈利变化趋势。支持formal和brief风格。
.claude/skills/aifinlab-a-share-earnings-momentum/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-16 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -61% | 0% |
| case-08 | ✓→✗ | ▼ Worse | -80% | 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]
分析师预期的上调/下调趋势
| 维度 | formal | brief | |------|--------|-------| | 盈利趋势 | 多季度趋势图 | 加速/减速 | | 动量信号 | 各指标综合评分 | 盈利动量方向 | 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 | 16,287 | 10,510 | -35% | 1 | 1 | 0% | 2,348 | 1,037 | -56% | 0 | 0 | — |
case-02 | fail→fail | 16,840 | 7,622 | -55% | 1 | 1 | 0% | 2,217 | 802 | -64% | 0 | 0 | — |
case-03 | fail→fail | 15,886 | 8,407 | -47% | 1 | 1 | 0% | 2,219 | 711 | -68% | 0 | 0 | — |
case-04 | pass→pass | 23,580 | 17,172 | -27% | 1 | 1 | 0% | 3,107 | 3,167 | +2% | 0 | 0 | — |
case-05 | pass→pass | 26,698 | 25,149 | -6% | 1 | 1 | 0% | 3,257 | 3,775 | +16% | 0 | 0 | — |
case-16 | fail→pass | 14,541 | 4,165 | -71% | 1 | 1 | 0% | 1,867 | 964 | -48% | 0 | 0 | — |
case-06 | pass→pass | 20,052 | 23,249 | +16% | 1 | 1 | 0% | 2,846 | 4,025 | +41% | 0 | 0 | — |
case-07 | fail→pass | 23,837 | 20,880 | -12% | 1 | 1 | 0% | 2,777 | 3,335 | +20% | 0 | 0 | — |
case-08 | pass→fail | 25,913 | 10,455 | -60% | 1 | 1 | 0% | 4,180 | 831 | -80% | 0 | 0 | — |
case-09 | fail→pass | 11,004 | 3,073 | -72% | 1 | 1 | 0% | 2,039 | 944 | -54% | 0 | 0 | — |
case-10 | fail→fail | 22,483 | 22,580 | +0% | 1 | 1 | 0% | 3,034 | 2,420 | -20% | 0 | 0 | — |
case-11 | pass→pass | 17,112 | 17,007 | -1% | 1 | 1 | 0% | 2,456 | 2,916 | +19% | 0 | 0 | — |
case-12 | fail→fail | 10,804 | 13,120 | +21% | 1 | 1 | 0% | 1,414 | 2,125 | +50% | 0 | 0 | — |
case-13 | fail→pass | 18,635 | 5,818 | -69% | 1 | 1 | 0% | 3,154 | 1,242 | -61% | 0 | 0 | — |
case-14 | pass→pass | 10,488 | 5,131 | -51% | 1 | 1 | 0% | 1,321 | 1,117 | -15% | 0 | 0 | — |
case-15 | fail→fail | 12,582 | 8,882 | -29% | 1 | 1 | 0% | 1,649 | 875 | -47% | 0 | 0 | — |
case-17 | pass→pass | 21,929 | 19,661 | -10% | 1 | 1 | 0% | 2,732 | 3,133 | +15% | 0 | 0 | — |
case-18 | pass→pass | 14,278 | 13,614 | -5% | 1 | 1 | 0% | 2,021 | 2,374 | +17% | 0 | 0 | — |
case-19 | fail→fail | 13,119 | 6,300 | -52% | 1 | 1 | 0% | 1,875 | 782 | -58% | 0 | 0 | — |
case-20 | pass→fail | 23,985 | 8,273 | -66% | 1 | 1 | 0% | 3,122 | 947 | -70% | 0 | 0 | — |
case-21 | pass→fail | 23,162 | 8,096 | -65% | 1 | 1 | 0% | 4,452 | 738 | -83% | 0 | 0 | — |
case-22 | pass→pass | 24,582 | 25,008 | +2% | 1 | 1 | 0% | 3,099 | 3,792 | +22% | 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 14 counted toward the lift figure. The other 8 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 +5 percentage points is the difference between those two pass rates over the 14 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.