Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →提供获取用户自选分组(LongPort)、股票市场数据(LongPort + AkShare)、技术指标计算（ta-lib）的技能。分析数并与用户研究交易策略，形成买卖点建议，并可做回测、寻优与报告输出。
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -41% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -69% | 0% |
当用户提出下列任意请求时，加载本 Skill：
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 22,170 | 12,662 | -43% | 1 | 1 | 0% | 4,116 | 2,637 | -36% | 0 | 0 | — |
case-02 | fail→fail | 24,752 | 9,059 | -63% | 1 | 1 | 0% | 4,619 | 1,964 | -57% | 0 | 0 | — |
case-03 | fail→fail | 16,533 | 10,427 | -37% | 1 | 1 | 0% | 2,871 | 2,179 | -24% | 0 | 0 | — |
case-04 | pass→pass | 24,737 | 25,797 | +4% | 1 | 1 | 0% | 4,545 | 5,102 | +12% | 0 | 0 | — |
case-05 | pass→pass | 17,940 | 22,847 | +27% | 1 | 1 | 0% | 2,989 | 3,253 | +9% | 0 | 0 | — |
case-06 | pass→pass | 19,013 | 13,378 | -30% | 1 | 1 | 0% | 3,121 | 2,454 | -21% | 0 | 0 | — |
case-07 | fail→pass | 10,908 | 3,436 | -69% | 1 | 1 | 0% | 1,591 | 932 | -41% | 0 | 0 | — |
case-08 | fail→pass | 15,840 | 4,735 | -70% | 1 | 1 | 0% | 2,643 | 1,227 | -54% | 0 | 0 | — |
case-09 | fail→pass | 15,140 | 2,840 | -81% | 1 | 1 | 0% | 2,405 | 915 | -62% | 0 | 0 | — |
case-10 | pass→pass | 7,712 | 5,337 | -31% | 1 | 1 | 0% | 1,162 | 1,196 | +3% | 0 | 0 | — |
case-11 | fail→pass | 18,766 | 3,246 | -83% | 1 | 1 | 0% | 2,817 | 881 | -69% | 0 | 0 | — |
case-12 | pass→pass | 15,480 | 10,910 | -30% | 1 | 1 | 0% | 2,207 | 1,992 | -10% | 0 | 0 | — |
case-13 | pass→pass | 14,743 | 8,905 | -40% | 1 | 1 | 0% | 2,235 | 1,825 | -18% | 0 | 0 | — |
case-14 | pass→pass | 17,445 | 9,036 | -48% | 1 | 1 | 0% | 2,746 | 2,039 | -26% | 0 | 0 | — |
case-15 | fail→pass | 11,531 | 2,469 | -79% | 1 | 1 | 0% | 1,581 | 856 | -46% | 0 | 0 | — |
case-16 | fail→pass | 16,758 | 1,740 | -90% | 1 | 1 | 0% | 2,471 | 748 | -70% | 0 | 0 | — |
case-17 | fail→pass | 11,472 | 2,572 | -78% | 1 | 1 | 0% | 2,059 | 894 | -57% | 0 | 0 | — |
case-18 | pass→pass | 19,830 | 13,979 | -30% | 1 | 1 | 0% | 2,896 | 2,706 | -7% | 0 | 0 | — |
case-19 | pass→pass | 16,538 | 9,819 | -41% | 1 | 1 | 0% | 2,539 | 2,014 | -21% | 0 | 0 | — |
case-20 | pass→pass | 11,217 | 3,287 | -71% | 1 | 1 | 0% | 1,776 | 985 | -45% | 0 | 0 | — |
case-21 | pass→pass | 11,686 | 2,441 | -79% | 1 | 1 | 0% | 1,761 | 869 | -51% | 0 | 0 | — |
case-22 | pass→pass | 20,135 | 18,243 | -9% | 1 | 1 | 0% | 2,982 | 3,424 | +15% | 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 +36 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.
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.