Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →期货数据Skill - 提供商品期货、股指期货、持仓龙虎榜、期现价差分析 via AkShare/新浪财经
.claude/skills/aifinlab-akshare-futures/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 107% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 116% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 129% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -37% | 0% |
| 属性 | 内容 | |:---|:---| | 名称 | akshare-futures | | 版本 | 1.1.0 | | 分类 | 期货数据 | | 状态 | ✅ 已上线 | | 维护者 | FinClaw Core Team | | 最后更新 | 2026-03-19 |
期货数据Skill，提供商品期货、股指期货的实时行情、历史数据、持仓量龙虎榜、期现价差分析等。覆盖上期所、大商所、郑商所、中金所、能源中心五大交易所。
| 用户输入 | 识别意图 | 调用函数 | |:---|:---|:---| | 螺纹钢期货行情 | futures_quote | futures_quote.py RB0 | | 查询豆粕持仓龙虎榜 | futures_hold | futures_hold.py M0 | | 原油期货多少钱？ | futures_quote | futures_quote.py SC0 | | 股指期货走势 | futures_index | futures_quote.py IF0 | | 期现价差分析 | futures_spread | futures_spread.py RB |
| 数据类型 | 主要来源 | 备用来源 | 认证要求 | |:---|:---|:---|:---:| | 期货实时行情 | 新浪财经 | AkShare | 无需 | | 期货历史数据 | AkShare-东方财富 | - | 无需 | | 持仓龙虎榜 | AkShare-交易所 | - | 无需 | | 外盘期货 | AkShare-新浪 | - | 无需 |
| 交易所 | 代码 | 主要品种 | |:---|:---|:---| | 上海期货交易所 | SHFE | 铜CU、铝AL、螺纹钢RB、黄金AU、原油SC | | 大连商品交易所 | DCE | 豆粕M、铁矿石I、棕榈油P、玉米C | | 郑州商品交易所 | ZCE | PTATA、甲醇MA、白糖SR、棉花CF | | 中国金融期货交易所 | CFFEX | 沪深300IF、上证50IH、中证500IC、国债T | | 上海国际能源交易中心 | INE | 原油期货SC、20号胶NR、低硫燃料油LU |
bash python scripts/futures_quote.py RB0 # 螺纹钢主力 python scripts/futures_quote.py SC0 # 原油主力 python scripts/futures_quote.py IF0 # 沪深300股指
bash python scripts/futures_hist.py RB2505 20250101 20260319
bash python scripts/futures_hold.py RB2505 # 螺纹钢具体合约 python scripts/futures_hold.py OI2501 # 菜籽油合约
bash python scripts/futures_main.py shfe # 上期所主力合约 python scripts/futures_main.py dce # 大商所主力合约
bash python scripts/futures_board.py
bash python scripts/futures_spread.py RB # 螺纹钢期现价差 python scripts/futures_spread.py M # 豆粕期现价差
bash python scripts/futures_global.py
| 脚本名 | 功能 | 入口点 | |:---|:---|:---:| | futures_quote.py | 期货实时行情 | ✅ | | futures_hist.py | 期货历史数据 | ✅ | | futures_hold.py | 持仓龙虎榜 | ✅ | | futures_main.py | 主力合约切换 | ✅ | | futures_board.py | 期货板块监控 | ✅ | | futures_spread.py | 期现价差分析 | ✅ | | futures_global.py | 外盘期货行情 | ✅ |
| 表示法 | 含义 | 示例 | |:---|:---|:---| | RB0 | 螺纹钢主力合约 | RB0 | | RB2505 | 螺纹钢2025年5月合约 | RB2505 | | M0 | 豆粕主力合约 | M0 | | SC0 | 原油主力合约 | SC0 |
| 品种 | 代码 | 交易所 | |:---|:---|:---| | 螺纹钢 | RB | SHFE | | 热轧卷板 | HC | SHFE | | 铁矿石 | I | DCE | | 焦煤 | JM | DCE | | 焦炭 | J | DCE | | 豆粕 | M | DCE | | 棕榈油 | P | DCE | | 原油 | SC | INE | | 黄金 | AU | SHFE | | 白银 | AG | SHFE | | 铜 | CU | SHFE | | 铝 | AL | SHFE | | 沪镍 | NI | SHFE | | 碳酸锂 | LC | GFEX | | 工业硅 | SI | GFEX |
| 价差状态 | 含义 | 市场判断 | |:---|:---|:---| | 正价差(Contango) | 期货 > 现货 | 远期供应充足，仓储成本 | | 逆价差(Backwardation) | 期货 < 现货 | 即期供应紧张，需求旺盛 | | 基差扩大 | 期现价差变大 | 供需失衡加剧 | | 基差收窄 | 期现价差变小 | 期现趋于一致 |
本Skill所有输出数据将按以下格式标注来源：
markdown--- 📊 **数据来源**: 新浪财经 / AkShare-东方财富 ⏱️ **数据时间**: 2026-03-19 10:30:15 📌 **交易所**: 上海期货交易所 🔗 **数据接口**: 新浪财经期货API 🔧 **分析工具**: FinClaw v1.0 ⚠️ **风险提示**: 期货交易风险高，入市需谨慎
akshare>=1.10.0
pandas>=1.3.0
requests>=2.25.0
pyyaml>=5.4.0| 交易所 | 日盘 | 夜盘 | |:---|:---|:---| | 上期所/大商所/郑商所 | 09:00-11:30, 13:30-15:00 | 21:00-次日02:30（部分品种） | | 中金所 | 09:30-11:30, 13:00-15:00 | 无 | | 能源中心 | 09:00-11:30, 13:30-15:00 | 21:00-次日02:30 |
| 指标 | 目标值 | 当前值 | |:---|:---:|:---:| | 实时行情延迟 | < 1s | ~500ms | | 数据可用性 | > 95% | 98% | | 数据准确率 | > 98% | 99% |
| 版本 | 日期 | 变更内容 | |:---|:---:|:---| | 1.1.0 | 2026-03-19 | 符合FinClaw数据规范v1.0，新增触发意图、期现价差解读 | | 1.0.0 | 2026-03-13 | 初始版本 |
finclaw/config/data_source_config.yaml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→pass | 11,591 | 8,911 | -23% | 1 | 1 | 0% | 1,923 | 3,971 | +107% | 0 | 0 | — |
case-22 | pass→pass | 21,595 | 3,711 | -83% | 1 | 1 | 0% | 2,556 | 3,085 | +21% | 0 | 0 | — |
case-02 | fail→fail | 18,544 | 18,741 | +1% | 1 | 1 | 0% | 3,167 | 6,455 | +104% | 0 | 0 | — |
case-03 | fail→pass | 25,110 | 29,476 | +17% | 1 | 1 | 0% | 3,406 | 7,363 | +116% | 0 | 0 | — |
case-04 | fail→fail | 16,590 | 25,729 | +55% | 1 | 1 | 0% | 2,924 | 6,486 | +122% | 0 | 0 | — |
case-05 | fail→pass | 13,353 | 15,473 | +16% | 1 | 1 | 0% | 2,130 | 4,877 | +129% | 0 | 0 | — |
case-06 | pass→pass | 15,607 | 16,757 | +7% | 1 | 1 | 0% | 3,378 | 6,378 | +89% | 0 | 0 | — |
case-07 | pass→pass | 12,456 | 5,768 | -54% | 1 | 1 | 0% | 1,918 | 3,453 | +80% | 0 | 0 | — |
case-08 | pass→pass | 19,130 | 7,565 | -60% | 1 | 1 | 0% | 2,250 | 3,860 | +72% | 0 | 0 | — |
case-09 | fail→pass | 15,154 | 3,487 | -77% | 1 | 1 | 0% | 2,726 | 3,351 | +23% | 0 | 0 | — |
case-10 | fail→pass | 28,653 | 2,534 | -91% | 1 | 1 | 0% | 4,819 | 3,023 | -37% | 0 | 0 | — |
case-11 | fail→pass | 21,462 | 4,994 | -77% | 1 | 1 | 0% | 4,111 | 3,334 | -19% | 0 | 0 | — |
case-12 | pass→pass | 27,152 | 10,740 | -60% | 1 | 1 | 0% | 4,418 | 4,599 | +4% | 0 | 0 | — |
case-13 | fail→fail | 13,347 | 6,104 | -54% | 1 | 1 | 0% | 2,509 | 3,430 | +37% | 0 | 0 | — |
case-14 | fail→pass | 20,214 | 4,410 | -78% | 1 | 1 | 0% | 3,129 | 3,398 | +9% | 0 | 0 | — |
case-15 | pass→pass | 2,565 | 3,555 | +39% | 1 | 1 | 0% | 449 | 3,249 | +624% | 0 | 0 | — |
case-16 | pass→pass | 10,199 | 13,242 | +30% | 1 | 1 | 0% | 1,476 | 4,434 | +200% | 0 | 0 | — |
case-17 | fail→pass | 27,348 | 3,337 | -88% | 1 | 1 | 0% | 2,766 | 3,134 | +13% | 0 | 0 | — |
case-18 | pass→pass | 8,489 | 3,996 | -53% | 1 | 1 | 0% | 1,361 | 3,205 | +135% | 0 | 0 | — |
case-19 | pass→pass | 9,922 | 10,791 | +9% | 1 | 1 | 0% | 1,901 | 4,145 | +118% | 0 | 0 | — |
case-20 | pass→pass | 4,901 | 4,777 | -3% | 1 | 1 | 0% | 818 | 3,470 | +324% | 0 | 0 | — |
case-21 | pass→pass | 11,508 | 11,073 | -4% | 1 | 1 | 0% | 2,212 | 4,696 | +112% | 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.
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.