AI Agent Hub
Back to skills
A-Share Weekly Review and Feishu Report icon

A-Share Weekly Review and Feishu Report

Professional Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md to install @user_be9bdc17/a-stock-weekly.

About this skill

Problem

Weekly A-share review often gets scattered across index K-lines, ETF trends, Tonghuashun hotspot attribution, Sina news, and representative stocks. Turning that into a consistent report and publishing it to Feishu can be manual, with extra risk of Chinese text corruption in Windows PowerShell pipes. A-Share Weekly Review and Feishu Report collapses those steps into a local pipeline.

How It Works

  • Data capture: pulls index and ETF forward-adjusted K-lines from Tencent, 5-day hotspot attribution from Tonghuashun, A/H/US news from Sina, and optional announcements from Juchao; all sources are zero-key and do not depend on East Money.
  • Analysis and reporting: calculates the previous complete trading week, summarizes index/ETF weekly and daily moves plus representative stock performance, then outputs intermediate JSON and a Markdown research report.
  • Feishu publishing: writes the report to Feishu Docs through lark-cli bot identity, preferring file-based upload to reduce GBK pipe encoding issues.
    Boundary: it is best used as a local skill inside Codex or Claude sessions, with a workspace, Feishu credentials, and network access configured. If a data source hits risk controls, an ETF is missing data, or a Feishu URL expires, usually only the affected step needs to be rerun.

Use Cases

  • After the weekend close, compile index, ETF, and representative stock moves into a Markdown review.
  • Merge Tonghuashun hotspot attribution and Sina news into the next-week outlook section.
  • Push the generated weekly market report to Feishu Docs using lark-cli.
  • Rerun only the Feishu push when the document link expires, without refetching all data.

Best For

  • Analysts who write weekly A-share reviews and share them with research teams
  • Users generating market outlook reports inside Codex or Claude sessions
  • Researchers publishing weekly reports to Feishu Docs
  • Engineers maintaining local market-data scripts and Feishu pushes