Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →龙虎榜Skill - 提供营业部追踪、游资图谱、机构席位分析 via AkShare
.claude/skills/aifinlab-akshare-lhb-detail/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 96% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 52% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -8% | 0% |
| 属性 | 内容 | |:---|:---| | 名称 | akshare-lhb-detail | | 版本 | 1.1.0 | | 分类 | 龙虎榜 | | 状态 | ✅ 已上线 | | 维护者 | FinClaw Core Team | | 最后更新 | 2026-03-19 |
龙虎榜Skill，提供A股每日龙虎榜数据查询、营业部追踪（游资动向）、机构专用席位分析、游资图谱等功能。追踪市场主力资金动向。
| 用户输入 | 识别意图 | 调用函数 | |:---|:---|:---| | 今天龙虎榜 | lhb_daily | lhb_daily.py | | 茅台龙虎榜历史 | lhb_stock | lhb_stock.py --code 600519 | | 追踪游资营业部 | lhb_dealer | lhb_dealer.py --name "中信证券上海溧阳路" | | 机构席位动向 | lhb_institution | lhb_institution.py --days 30 |
| 数据类型 | 主要来源 | 备用来源 | 认证要求 | |:---|:---|:---|:---:| | 龙虎榜数据 | AkShare-交易所 | - | 无需 | | 营业部数据 | AkShare | - | 无需 |
| 条件 | 说明 | |:---|:---| | 日涨跌幅偏离值达±7% | 相对大盘指数偏离 | | 日振幅达15% | 当日最高价-最低价 | | 日换手率超20% | 成交量/流通股本 | | 连续3日涨跌幅偏离值累计达±20% | 累计偏离 | | ST股票日涨跌幅达±5% | ST股特殊条件 |
| 游资 | 常用席位 | 风格 | |:---|:---|:---| | 赵老哥 | 银河证券绍兴 | 超短线、打板 | | 章盟主 | 国泰君安上海江苏路 | 大单封板、趋势 | | 方新侠 | 兴业证券陕西分公司 | 趋势、接力 | | 作手新一 | 国泰君安南京太平南路 | 龙头战法 | | 炒股养家 | 华鑫证券上海宛平南路 | 隔日超短 | | 小鳄鱼 | 南京证券南京大钟亭 | 趋势、龙头 |
bash python scripts/lhb_daily.py python scripts/lhb_daily.py --date 20260318
bash python scripts/lhb_stock.py --code 600519
bash python scripts/lhb_dealer.py --name "中信证券上海溧阳路"
bash python scripts/lhb_institution.py --days 30
| 脚本名 | 功能 | 入口点 | |:---|:---|:---:| | lhb_daily.py | 每日龙虎榜 | ✅ | | lhb_stock.py | 个股龙虎榜 | ✅ | | lhb_dealer.py | 营业部追踪 | ✅ | | lhb_institution.py | 机构席位分析 | ✅ |
| 类型 | 说明 | 信号意义 | |:---|:---|:---| | 机构专用 | 公募基金、保险、QFII | 中长期资金，信号较稳 | | 游资营业部 | 知名游资席位 | 短线资金，波动大 | | 普通营业部 | 散户/小机构 | 参考意义有限 |
| 情形 | 解读 | |:---|:---| | 机构大买 | 看好中长期，可能上涨 | | 机构大卖 | 看空或调仓，可能下跌 | | 游资大买 | 短线炒作，波动加大 | | 游资大卖 | 获利了结，短期承压 | | 机构+游资同买 | 强烈看多信号 | | 机构买+游资卖 | 分歧，观望 |
markdown--- 📊 **数据来源**: AkShare-交易所龙虎榜 ⏱️ **数据时间**: 2026-03-19 📌 **上榜日期**: 2026-03-18 🔗 **原始来源**: 沪深交易所龙虎榜 🔧 **分析工具**: FinClaw v1.0 ⚠️ **风险提示**: 龙虎榜数据仅供参考，不构成投资建议
akshare>=1.10.0
pandas>=1.3.0
pyyaml>=5.4.0| 版本 | 日期 | 变更内容 | |:---|:---:|:---| | 1.1.0 | 2026-03-19 | 符合FinClaw数据规范v1.0，新增知名游资席位 | | 1.0.0 | 2026-03-13 | 初始版本 |
本Skill遵循 FinClaw 数据规范 v1.0 | 数据来源强制标注 | 禁止训练数据编造
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 23,975 | 27,607 | +15% | 1 | 1 | 0% | 4,096 | 5,847 | +43% | 0 | 0 | — |
case-02 | fail→fail | 20,829 | 21,814 | +5% | 1 | 1 | 0% | 2,819 | 4,926 | +75% | 0 | 0 | — |
case-03 | fail→pass | 22,132 | 31,341 | +42% | 1 | 1 | 0% | 3,256 | 6,374 | +96% | 0 | 0 | — |
case-04 | pass→pass | 18,565 | 14,767 | -20% | 1 | 1 | 0% | 2,825 | 3,734 | +32% | 0 | 0 | — |
case-05 | pass→pass | 23,600 | 15,298 | -35% | 1 | 1 | 0% | 3,333 | 3,822 | +15% | 0 | 0 | — |
case-06 | fail→pass | 17,802 | 14,658 | -18% | 1 | 1 | 0% | 2,467 | 3,749 | +52% | 0 | 0 | — |
case-07 | pass→pass | 15,123 | 5,835 | -61% | 1 | 1 | 0% | 2,112 | 2,418 | +14% | 0 | 0 | — |
case-08 | pass→pass | 17,005 | 11,739 | -31% | 1 | 1 | 0% | 2,583 | 3,513 | +36% | 0 | 0 | — |
case-09 | fail→pass | 19,043 | 9,443 | -50% | 1 | 1 | 0% | 2,709 | 3,018 | +11% | 0 | 0 | — |
case-10 | pass→pass | 15,292 | 12,619 | -17% | 1 | 1 | 0% | 2,711 | 3,611 | +33% | 0 | 0 | — |
case-11 | pass→pass | 14,473 | 6,747 | -53% | 1 | 1 | 0% | 1,901 | 2,796 | +47% | 0 | 0 | — |
case-12 | pass→pass | 14,574 | 10,057 | -31% | 1 | 1 | 0% | 1,988 | 3,359 | +69% | 0 | 0 | — |
case-13 | pass→pass | 10,585 | 7,981 | -25% | 1 | 1 | 0% | 1,872 | 2,996 | +60% | 0 | 0 | — |
case-14 | fail→pass | 22,007 | 8,236 | -63% | 1 | 1 | 0% | 4,486 | 2,864 | -36% | 0 | 0 | — |
case-15 | pass→pass | 21,855 | 17,316 | -21% | 1 | 1 | 0% | 2,574 | 4,210 | +64% | 0 | 0 | — |
case-16 | pass→pass | 20,578 | 19,872 | -3% | 1 | 1 | 0% | 2,816 | 4,408 | +57% | 0 | 0 | — |
case-17 | pass→pass | 11,569 | 5,673 | -51% | 1 | 1 | 0% | 1,530 | 2,545 | +66% | 0 | 0 | — |
case-18 | fail→pass | 15,036 | 5,000 | -67% | 1 | 1 | 0% | 2,585 | 2,385 | -8% | 0 | 0 | — |
case-19 | fail→pass | 25,949 | 5,047 | -81% | 1 | 1 | 0% | 4,901 | 2,289 | -53% | 0 | 0 | — |
case-20 | pass→pass | 24,708 | 27,365 | +11% | 1 | 1 | 0% | 3,388 | 5,669 | +67% | 0 | 0 | — |
case-21 | pass→pass | 17,524 | 14,566 | -17% | 1 | 1 | 0% | 2,553 | 4,404 | +73% | 0 | 0 | — |
case-22 | pass→pass | 25,325 | 44,933 | +77% | 1 | 1 | 0% | 4,104 | 8,496 | +107% | 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 +27 percentage points is the difference between those two pass rates over the 22 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.