Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →adata 免费A股量化数据源，专注A股市场，提供股票行情、资金流向、核心财务指标（43个字段）、北向资金、热门股票排行、分红数据、概念板块、指数成分股、基金及可转债数据。当用户需要A股行情、资金流、北向资金、热门排行、财务指标、分红信息时可使用此 skill。与 akshare-finance 和 efinance-data 互补——adata 的独特优势在于：(1)北向资金实时数据 (2)热门股票排行 (3)核心财务指标一次返回43个字段（EPS/ROE/毛利率/资产负债率/周转率等）(4)概念板块成分股。
.claude/skills/aifinlab-adata-source/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -43% | 0% |
adata 是专注 A 股的免费量化数据库（4.1k GitHub Stars），数据来源为东方财富、同花顺等公开平台。与 akshare/efinance 互补，adata 在以下方面有独特优势：
bashpip install adata pandas
| 脚本 | 用途 | 核心命令 | |------|------|---------| | stock_data.py | 股票行情/资金流/财务/分红 | list / market / capital / finance / dividend / board | | sentiment_data.py | 北向资金/热门排行 | north / hot | | fund_bond_data.py | 基金和可转债 | fund_info / bond_info |
bashpython stock_data.py list
返回：stock_code, short_name, exchange, list_date（全部A股）
bash# 日线行情 python stock_data.py market --code 000001 --start 2026-01-01 # 分钟级行情 python stock_data.py market_min --code 000001
返回：open, close, high, low, volume, amount, change_pct, turnover_ratio
bash# 历史资金流（主力/小单/中单/大单/超大单） python stock_data.py capital --code 000001 # 当日分钟级资金流 python stock_data.py capital_min --code 000001
bashpython stock_data.py finance --code 000001
这是 adata 最有价值的接口，单次返回：
bashpython stock_data.py dividend --code 000001
bash# 概念板块成分股（东财） python stock_data.py concept_members --concept BK0475 # 指数成分股 python stock_data.py index_members --index 000300
bash# 历史北向资金流入（沪股通/深股通/合计） python sentiment_data.py north # 当日实时北向资金 python sentiment_data.py north_current # 分钟级北向资金 python sentiment_data.py north_min
返回：net_hgt（沪股通净买入）, net_sgt（深股通净买入）, net_tgt（合计）
bash# 东方财富热门 TOP 100 python sentiment_data.py hot
返回：rank, stock_code, short_name, price, change_pct
bash# 基金信息 python fund_bond_data.py fund_info --code 005827 # 可转债信息 python fund_bond_data.py bond_info --code 123456
| 数据需求 | 优先用 | 原因 | |---------|-------|------| | 北向资金（实时/分钟级） | adata-source | akshare 有日级，adata 有分钟级 | | 热门股票排行 | adata-source | 其他源无此接口 | | 核心财务43字段 | adata-source | 单次调用最全，含周转/偿债指标 | | 概念板块成分股 | adata-source | 东财/同花顺双源 | | 财务三表完整数据 | akshare-finance | adata 只有核心指标 | | 资金流分层 | efinance-data | efinance 字段更细致 | | 宏观经济 | akshare-finance | adata 无宏观数据 | | 估值历史序列 | akshare-finance | adata 动态指标接口不稳定 |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 25,782 | 67,002 | +160% | 1 | 1 | 0% | 3,668 | 1,893 | -48% | 0 | 0 | — |
case-02 | fail→fail | 12,294 | 39,970 | +225% | 1 | 1 | 0% | 2,293 | 1,571 | -31% | 0 | 0 | — |
case-03 | fail→fail | 14,395 | 11,164 | -22% | 1 | 1 | 0% | 2,989 | 2,491 | -17% | 0 | 0 | — |
case-04 | fail→pass | 9,966 | 2,967 | -70% | 1 | 1 | 0% | 1,912 | 1,841 | -4% | 0 | 0 | — |
case-05 | fail→pass | 41,891 | 4,422 | -89% | 1 | 1 | 0% | 2,387 | 2,167 | -9% | 0 | 0 | — |
case-06 | fail→pass | 31,502 | 35,264 | +12% | 1 | 1 | 0% | 4,869 | 2,320 | -52% | 0 | 0 | — |
case-07 | fail→pass | 48,296 | 4,428 | -91% | 1 | 1 | 0% | 2,754 | 2,124 | -23% | 0 | 0 | — |
case-08 | fail→pass | 24,639 | 7,202 | -71% | 1 | 1 | 0% | 4,711 | 2,705 | -43% | 0 | 0 | — |
case-09 | fail→pass | 12,256 | 3,724 | -70% | 1 | 1 | 0% | 2,234 | 1,985 | -11% | 0 | 0 | — |
case-10 | fail→pass | 5,603 | 32,855 | +486% | 1 | 1 | 0% | 1,076 | 1,681 | +56% | 0 | 0 | — |
case-11 | fail→pass | 7,189 | 2,719 | -62% | 1 | 1 | 0% | 1,333 | 1,806 | +35% | 0 | 0 | — |
case-12 | fail→pass | 14,862 | 10,123 | -32% | 1 | 1 | 0% | 2,439 | 3,074 | +26% | 0 | 0 | — |
case-13 | fail→pass | 11,794 | 8,511 | -28% | 1 | 1 | 0% | 2,114 | 2,383 | +13% | 0 | 0 | — |
case-14 | fail→pass | 22,759 | 8,446 | -63% | 1 | 1 | 0% | 2,636 | 2,963 | +12% | 0 | 0 | — |
case-15 | fail→pass | 11,531 | 4,305 | -63% | 1 | 1 | 0% | 2,017 | 2,088 | +4% | 0 | 0 | — |
case-16 | fail→pass | 11,229 | 11,421 | +2% | 1 | 1 | 0% | 1,971 | 3,242 | +64% | 0 | 0 | — |
case-17 | pass→pass | 14,295 | 11,631 | -19% | 1 | 1 | 0% | 2,226 | 3,202 | +44% | 0 | 0 | — |
case-18 | fail→pass | 15,601 | 3,740 | -76% | 1 | 1 | 0% | 2,463 | 1,963 | -20% | 0 | 0 | — |
case-19 | pass→pass | 13,912 | 9,916 | -29% | 1 | 1 | 0% | 2,446 | 3,109 | +27% | 0 | 0 | — |
case-20 | fail→pass | 13,108 | 9,519 | -27% | 1 | 1 | 0% | 2,147 | 3,137 | +46% | 0 | 0 | — |
case-21 | fail→pass | 15,978 | 15,233 | -5% | 1 | 1 | 0% | 2,532 | 3,878 | +53% | 0 | 0 | — |
case-22 | pass→pass | 18,061 | 6,109 | -66% | 1 | 1 | 0% | 2,813 | 2,416 | -14% | 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, and 19 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 +73 percentage points is the difference between those two pass rates over the 19 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.