En

A Stock Daily Market Sense

Post-Market Structured Research Report Auto-Generator for A-Shares

Install & Use

Copy this prompt and send it to your AI assistant (Claude / Cursor / TRAE / Codex / WorkBuddy etc.) to auto-install:

Help me install this AI Skill: A Stock Daily Market Sense.
It is used for: Post-Market Structured Research Report Auto-Generator for A-Shares
Full Skill content: https://321skill.com/skills/a-stock-daily-market-sense-x-2/raw/index.md
Read that page and install it.

The prompt includes a link to the full Skill content. You can also view the full content.

This Skill addresses the low efficiency and fragmented information faced by A-share investors during post-market review. In actual trading, manually consolidating vast amounts of data—such as index trends, turnover concentration, profitability effect, and leading themes—after each market close is time-consuming and prone to missing key signals. Based on Tushare Pro daily data and Baostock style indices, this Skill automatically generates structured post-market research reports covering core modules like market trend, turnover concentration, profitability effect and leading themes, high-volume decline risks, and feature group analysis, compressing review work from hours to minutes.

Usage is simple: just instruct the AI with "Analyze today's A-share market" or specify a date. The Skill automatically runs the data pipeline to generate a deterministic evidence package (including individual stock K-lines, module-level JSON, etc.), followed by model-based module evaluation and writing. The process consists of: ① Identifying the trading day and generating the evidence package; ② Each module performing initial analysis using only its own JSON and methodology; ③ Determining theme star ratings via statistical scripts; ④ Aggregating module outputs and supplementing with catalyst and sub-theme deduction; ⑤ Finally generating a Markdown report and rendering to HTML as needed. The scripts do not invoke LLMs, and external messages cannot alter the theme star ratings determined by price and volume evidence, ensuring data objectivity.

It is well-suited for quantitative traders, data analysts, and investment researchers who require daily A-share review, especially teams or individuals already using Tushare or Baostock data sources and looking to automate their post-market analysis workflow. For users needing to quickly grasp market style shifts, changes in turnover concentration, and theme rotation rhythms, this Skill significantly enhances information processing efficiency while avoiding subjective bias inherent in manual summarization.

It is recommended to use this tool uniformly after each market close (post-15:00). Note that it primarily focuses on A-share post-market review and does not perform in-depth fundamental analysis of individual stocks, nor does it cover Hong Kong/US stocks, funds, futures, crypto analysis, ultra-short-term trading decisions, or automated order placement. All strength/weakness judgments are based on turnover evidence, without applying pre-existing industry/concept classifications. Leading themes are inducted by the model based on factual business data; when commonality is insufficient, it explicitly states "does not currently constitute a leading theme."

Key Features

Unlike automated reports from Wind or Tonghuashun, this Skill emphasizes turnover priority. All strength/weakness judgments are supported by turnover evidence, with rising themes confirmed by turnover depth. It avoids applying pre-existing industry/concept labels, instead having the model induct themes based on factual business data to prevent classification bias.

Limitations

Requires a Python environment along with Tushare Pro and Baostock data sources. Supports only A-share daily data; does not support Hong Kong/US stocks, funds, futures, or crypto analysis. Does not provide buy/sell recommendations.

FAQ

What data sources are required?

Requires Tushare Pro API and Baostock data. Tokens need to be configured on first use. The script automatically calls the data pipeline; no manual downloads are needed.

Does it support historical backtesting?

Supports generating reports for specified historical dates, but by default uses only data up to and including that date. For ex-post analysis, the --allow-future parameter is required.

Installation guide for AI assistants

If your AI coding assistant (Claude Code, Cursor, TRAE etc.) can see this page, send it this message to auto-install:

Visit https://321skill.com/skills/a-stock-daily-market-sense-x-2/raw/index.md to read the original Skill definition (Markdown format) for A Stock Daily Market Sense, and install it according to the instructions.