每日股票分析器

每日股票分析器:基于 Qlib + LLM Agent 的 A 股自选股每日 AI 信号生成器 含 0 条反模式约束。

✓ 0 人报告成功·v0.1.0·

晶体简介

基于 Qlib 的 A 股自选股智能分析系统,集成 LLM Agent ReAct 推理引擎和技术指标择时模块(MA 多头排列、乖离率阈值严进策略),自动生成每日 buy/hold/sell 指令并推送至微信。 本 skill 基于开源项目构建并集成 0 条 anti-pattern 约束。访问 doramagic.ai/r/daily-stock-analyzer 查看中英双语完整文档和触发场景。适用于 Doramagic 生态(Claude Code / Cursor / openclaw / ChatGPT / Gemini 等)。

Blueprint Source

finance-bp-004

tangweigang-jpg/doramagic-skills6360a631 source file

Constraints

38total
38fatal
38 must-not-violate

Evidence Quality

Confidence90%

Medium confidence — review before critical use

38 条不可违反的约束

FATALdomain_rulefinance-C-001

WHENWhen implementing a BaseFetcher subclass for data collection

ACTIONnormalize stock codes to exactly 6-digit format for A-share, 5-digit with HK prefix for HK, and uppercase alphanumeric for US stocks

CONSEQUENCEIncorrect stock code format causes routing failures and data fetcher returns empty DataFrame, breaking analysis pipeline for that stock

FATALdomain_rulefinance-C-002

WHENWhen calculating volume_ratio technical indicator

ACTIONuse shift(1) on the 5-day rolling average volume denominator to exclude current day from the calculation

CONSEQUENCEWithout shift(1), the volume_ratio includes current day volume in both numerator and denominator, creating look-ahead contamination that inflates the ratio

FATALdomain_rulefinance-C-003

WHENWhen returning fetched data from any fetcher implementation

ACTIONinclude exactly these columns in the output DataFrame: date, open, high, low, close, volume, amount, pct_chg, ma5, ma10, ma20, volume_ratio

CONSEQUENCEMissing required columns cause downstream analysis stages to fail with KeyError, breaking the entire trading signal generation pipeline

常见问题

讨论 (0)

暂无讨论,成为第一个发言的人吧!

更新历史

v0.1.02026-04-23·贡献者: tangweigang-jpg

v0.1.0: 首次发布到 Doramagic.ai。基于 tangweigang-jpg/doramagic-skills 的自动化 batch-v1 元数据 + 自动生成 FAQ。