quant-data
灵台量化数据管道——股票/加密币行情采集、清洗、存储与质量检查。含 Yahoo Finance + Kraken (CCXT) 双数据源,支持日线/1h 自动增量更新。
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
灵台量化数据管道——股票/加密币行情采集、清洗、存储与质量检查。含 Yahoo Finance + Kraken (CCXT) 双数据源,支持日线/1h 自动增量更新。
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
Configure the Feishu (Lark) bot addon for this agent — read this when the human asks to set up a Feishu bot.
Configure the IMAP email addon for this agent — read this when the human asks to set up email.
Configure the Telegram bot addon for this agent — read this when the human asks to set up a Telegram bot.
Configure the WeChat addon for this agent — read this when the human asks to set up WeChat.
Reference manual for the `avatar` tool — debugging, advanced patterns, best practices, and worked examples. Read when you need depth beyond the tool description.
Reference manual for the `bash` tool — debugging, advanced patterns, best practices, and worked examples. Read when you need depth beyond the tool description.
SOC 직업 분류 기준
| name | quant-data |
| description | 灵台量化数据管道——股票/加密币行情采集、清洗、存储与质量检查。含 Yahoo Finance + Kraken (CCXT) 双数据源,支持日线/1h 自动增量更新。 |
| version | 1.0.0 |
| agent | data_engineer |
| created | "2026-04-25T00:00:00.000Z" |
股票(Yahoo Finance)与加密货币(Kraken/CCXT)行情数据的采集、清洗、存储工具集。
| 源 | 类型 | 标的 | 限制 |
|---|---|---|---|
| Yahoo Finance | 美股日线/1h | SPY, QQQ, IWM, GLD, TLT, AAPL, MSFT, GOOGL, AMZN, NVDA | 日内深度 ~90 天 |
| Kraken (CCXT) | 加密币日线/1h | BTC/USD, ETH/USD | 1h 回溯上限 ~720 条 |
| Binance | ❌ 不可用 | — | 美国地区 HTTP 451 |
data/raw/{TICKER}/ ← Parquet 文件,按标的分目录
data/processed/ ← 清洗后数据(待建)
data/features/ ← 特征工程输出(待建)
文件命名:{TICKER}_{freq}_{开始日期}_{结束日期}.parquet
cd /Users/huangzesen/work/lingtai-projects/quant_company
python3 scripts/ingest_all.py --status
# 全量日线 + 日内
python3 scripts/ingest_all.py --all
# 仅日线增量更新
python3 scripts/ingest_all.py --daily
# 仅日内 1h
python3 scripts/ingest_all.py --intraday
# 指定标的
python3 scripts/ingest_all.py --daily --tickers SPY QQQ AAPL
python3 scripts/ingest_all.py --quality
python3 scripts/ingest_all.py --quality daily
python3 scripts/ingest_all.py --quality intraday
shared_lib/data_fetcher.py — 数据获取层from shared_lib.data_fetcher import fetch_yahoo, fetch_ccxt, load_local, list_local_data
# 采集 Yahoo 数据
df = fetch_yahoo('SPY', interval='1d', start=..., end=...)
# 采集 Kraken 数据
df = fetch_ccxt('BTC/USD', exchange='kraken', timeframe='1d', limit=500)
# 加载本地数据(自动合并去重排序)
df = load_local('SPY')
# 列出本地文件
files = list_local_data('SPY')
shared_lib/features.py — 技术指标from shared_lib.features import add_all_indicators
df = add_all_indicators(df) # 返回含 SMA/MACD/RSI/布林带/ATR/OBV 的 DataFrame
ingest_all.py --daily 会:
设置每日收盘后运行:
# 每天 16:30 ET 运行(美股收盘后半小时)
30 20 * * 1-5 cd /path/to/quant_company && python3 scripts/ingest_all.py --daily
/opt/anaconda3/bin/python3