Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →AkShare 融资融券数据，提供两融余额、杠杆资金监控、两融标的筛选。当用户需要通过 AkShare 获取融资融券数据时使用。
.claude/skills/aifinlab-akshare-margin/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -73% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -76% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -80% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -59% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -71% | 0% |
融资融券数据查询与分析 Skill。提供A股市场融资融券余额、杠杆资金监控、两融标的筛选等功能。
查询沪深两市融资融券余额：
bashpython scripts/margin_balance.py
查询个股融资融券数据：
bashpython scripts/margin_stock.py --code 000001
查询融资买入排行：
bashpython scripts/margin_rank.py --type buy --limit 20
查询融券卖出排行：
bashpython scripts/margin_rank.py --type sell --limit 20
bashpip install akshare pandas
MIT
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 15,229 | 9,765 | -36% | 1 | 1 | 0% | 2,415 | 814 | -66% | 0 | 0 | — |
case-02 | fail→fail | 11,229 | 12,372 | +10% | 1 | 1 | 0% | 1,813 | 1,054 | -42% | 0 | 0 | — |
case-03 | fail→fail | 15,843 | 8,308 | -48% | 1 | 1 | 0% | 2,241 | 829 | -63% | 0 | 0 | — |
case-04 | fail→pass | 21,492 | 3,562 | -83% | 1 | 1 | 0% | 3,500 | 947 | -73% | 0 | 0 | — |
case-05 | fail→pass | 19,840 | 2,720 | -86% | 1 | 1 | 0% | 3,274 | 796 | -76% | 0 | 0 | — |
case-06 | fail→pass | 17,633 | 2,261 | -87% | 1 | 1 | 0% | 3,233 | 644 | -80% | 0 | 0 | — |
case-07 | fail→pass | 11,363 | 3,111 | -73% | 1 | 1 | 0% | 2,141 | 876 | -59% | 0 | 0 | — |
case-08 | fail→pass | 21,162 | 4,411 | -79% | 1 | 1 | 0% | 3,864 | 1,115 | -71% | 0 | 0 | — |
case-09 | fail→pass | 13,446 | 2,305 | -83% | 1 | 1 | 0% | 2,430 | 713 | -71% | 0 | 0 | — |
case-10 | pass→pass | 14,033 | 10,942 | -22% | 1 | 1 | 0% | 2,304 | 2,389 | +4% | 0 | 0 | — |
case-11 | pass→pass | 17,756 | 11,949 | -33% | 1 | 1 | 0% | 2,411 | 2,332 | -3% | 0 | 0 | — |
case-12 | pass→pass | 17,813 | 17,515 | -2% | 1 | 1 | 0% | 2,258 | 2,658 | +18% | 0 | 0 | — |
case-13 | pass→pass | 16,210 | 13,535 | -17% | 1 | 1 | 0% | 1,982 | 2,449 | +24% | 0 | 0 | — |
case-14 | fail→pass | 13,012 | 3,916 | -70% | 1 | 1 | 0% | 2,132 | 775 | -64% | 0 | 0 | — |
case-15 | pass→pass | 10,223 | 1,755 | -83% | 1 | 1 | 0% | 1,659 | 546 | -67% | 0 | 0 | — |
case-16 | fail→pass | 17,724 | 4,221 | -76% | 1 | 1 | 0% | 3,369 | 968 | -71% | 0 | 0 | — |
case-17 | pass→pass | 13,353 | 2,684 | -80% | 1 | 1 | 0% | 1,803 | 742 | -59% | 0 | 0 | — |
case-18 | pass→pass | 9,206 | 2,551 | -72% | 1 | 1 | 0% | 1,437 | 686 | -52% | 0 | 0 | — |
case-19 | pass→pass | 8,580 | 1,788 | -79% | 1 | 1 | 0% | 1,481 | 629 | -58% | 0 | 0 | — |
case-20 | fail→fail | 16,632 | 17,195 | +3% | 1 | 1 | 0% | 2,195 | 3,568 | +63% | 0 | 0 | — |
case-21 | fail→fail | 15,079 | 22,399 | +49% | 1 | 1 | 0% | 2,208 | 3,615 | +64% | 0 | 0 | — |
case-22 | fail→fail | 16,961 | 28,903 | +70% | 1 | 1 | 0% | 2,773 | 3,808 | +37% | 0 | 0 | — |
case-23 | fail→pass | 11,207 | 4,790 | -57% | 1 | 1 | 0% | 1,819 | 997 | -45% | 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. 23 cases were attempted, and 20 counted toward the lift figure. The other 3 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 +39 percentage points is the difference between those two pass rates over the 20 comparable cases.
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.