Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →大宗交易Skill - 提供大宗交易明细、机构席位追踪、折溢价分析 via AkShare
.claude/skills/aifinlab-akshare-block-trade/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 221% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 80% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 86% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 27% | 0% |
| 属性 | 内容 | |:---|:---| | 名称 | akshare-block-trade | | 版本 | 1.1.0 | | 分类 | 大宗交易 | | 状态 | ✅ 已上线 | | 维护者 | FinClaw Core Team | | 最后更新 | 2026-03-19 |
大宗交易Skill，提供A股市场大宗交易明细查询、机构席位追踪、折溢价分析、营业部统计等功能。识别机构大宗交易动向和折价套利机会。
| 用户输入 | 识别意图 | 调用函数 | |:---|:---|:---| | 今天大宗交易 | block_daily | block_daily.py | | 茅台大宗交易 | block_stock | block_stock.py --code 600519 | | 机构大宗交易 | block_institution | block_institution.py --days 30 | | 折溢价分析 | block_premium | block_premium.py |
| 数据类型 | 主要来源 | 备用来源 | 认证要求 | |:---|:---|:---|:---:| | 大宗交易 | AkShare-交易所 | - | 无需 |
| 市场 | 单笔标准 | |:---|:---| | 沪市A股 | ≥30万股 或 ≥200万元 | | 深市A股 | ≥30万股 或 ≥300万元 | | 沪市B股 | ≥30万股 或 ≥20万美元 | | 深市B股 | ≥3万股 或 ≥20万港币 |
bash python scripts/block_daily.py
bash python scripts/block_stock.py --code 600519
bash python scripts/block_institution.py --days 30
bash python scripts/block_premium.py
| 脚本名 | 功能 | 入口点 | |:---|:---|:---:| | block_daily.py | 每日大宗交易 | ✅ | | block_stock.py | 个股大宗交易 | ✅ | | block_institution.py | 机构大宗交易 | ✅ | | block_premium.py | 折溢价分析 | ✅ |
| 折溢价 | 说明 | 市场含义 | |:---|:---:|:---| | 溢价 >5% | 成交价比市价高 | 强烈看好，急于买入 | | 溢价 0-5% | 略高于市价 | 正常交易 | | 平价 | 等于市价 | 正常大宗交易 | | 折价 0-5% | 略低于市价 | 正常大宗交易 | | 折价 5-10% | 明显低于市价 | 卖方急于出货 | | 折价 >10% | 大幅低于市价 | 重大利空或流动性危机 |
markdown--- 📊 **数据来源**: AkShare-交易所 ⏱️ **数据时间**: 2026-03-19 📌 **交易日期**: 2026-03-18 🔗 **原始来源**: 沪深交易所大宗交易数据 🔧 **分析工具**: FinClaw v1.0
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，新增折溢价解读 | | 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 | 14,586 | 12,941 | -11% | 1 | 1 | 0% | 2,121 | 3,249 | +53% | 0 | 0 | — |
case-02 | fail→fail | 19,861 | 18,731 | -6% | 1 | 1 | 0% | 2,725 | 5,018 | +84% | 0 | 0 | — |
case-03 | fail→fail | 20,578 | 8,204 | -60% | 1 | 1 | 0% | 3,038 | 1,717 | -43% | 0 | 0 | — |
case-04 | fail→fail | 13,242 | 20,931 | +58% | 1 | 1 | 0% | 1,968 | 4,363 | +122% | 0 | 0 | — |
case-05 | fail→pass | 14,560 | 34,254 | +135% | 1 | 1 | 0% | 2,407 | 7,730 | +221% | 0 | 0 | — |
case-06 | fail→fail | 24,060 | 28,150 | +17% | 1 | 1 | 0% | 3,561 | 5,651 | +59% | 0 | 0 | — |
case-07 | pass→pass | 13,922 | 7,116 | -49% | 1 | 1 | 0% | 2,446 | 2,599 | +6% | 0 | 0 | — |
case-08 | pass→pass | 6,545 | 4,384 | -33% | 1 | 1 | 0% | 913 | 1,991 | +118% | 0 | 0 | — |
case-09 | fail→pass | 8,484 | 6,231 | -27% | 1 | 1 | 0% | 1,421 | 2,557 | +80% | 0 | 0 | — |
case-10 | fail→pass | 9,902 | 6,212 | -37% | 1 | 1 | 0% | 1,455 | 2,436 | +67% | 0 | 0 | — |
case-11 | fail→pass | 6,361 | 3,482 | -45% | 1 | 1 | 0% | 1,051 | 1,960 | +86% | 0 | 0 | — |
case-12 | pass→pass | 4,596 | 4,104 | -11% | 1 | 1 | 0% | 780 | 2,068 | +165% | 0 | 0 | — |
case-13 | pass→pass | 16,382 | 16,336 | -0% | 1 | 1 | 0% | 2,297 | 3,740 | +63% | 0 | 0 | — |
case-14 | pass→pass | 19,274 | 16,925 | -12% | 1 | 1 | 0% | 2,532 | 3,754 | +48% | 0 | 0 | — |
case-15 | pass→pass | 17,500 | 8,423 | -52% | 1 | 1 | 0% | 2,487 | 2,411 | -3% | 0 | 0 | — |
case-16 | fail→pass | 20,735 | 12,666 | -39% | 1 | 1 | 0% | 2,588 | 3,288 | +27% | 0 | 0 | — |
case-17 | pass→pass | 24,558 | 18,230 | -26% | 1 | 1 | 0% | 3,237 | 4,267 | +32% | 0 | 0 | — |
case-18 | fail→pass | 10,960 | 3,829 | -65% | 1 | 1 | 0% | 1,934 | 1,861 | -4% | 0 | 0 | — |
case-19 | fail→pass | 17,187 | 3,365 | -80% | 1 | 1 | 0% | 2,124 | 1,956 | -8% | 0 | 0 | — |
case-20 | fail→pass | 14,390 | 27,818 | +93% | 1 | 1 | 0% | 2,313 | 6,177 | +167% | 0 | 0 | — |
case-21 | pass→pass | 17,216 | 25,057 | +46% | 1 | 1 | 0% | 2,079 | 4,920 | +137% | 0 | 0 | — |
case-22 | pass→pass | 8,767 | 10,737 | +22% | 1 | 1 | 0% | 1,480 | 3,195 | +116% | 0 | 0 | — |
case-23 | pass→pass | 12,170 | 13,070 | +7% | 1 | 1 | 0% | 1,788 | 4,023 | +125% | 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 22 counted toward the lift figure. The other 1 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 +35 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.