A Stock Daily Market Sense

This Skill addresses the inefficiency of daily post-market review for A-share investors. Manually analyzing market trends, concentration of trading volume, profitability effects, and leading market themes is time-consuming and prone to missing key data. Based on Tushare Pro daily data and Baostock style indices, this Skill automatically generates structured post-market reports covering core modules such as market trend, trading volume concentration, profitability effect, leading themes, risks of high-volume declines, and feature group analysis, reducing review work from hours to minutes.

Usage is straightforward: simply specify a trading date (e.g., "today" or a specific date). The Skill automatically runs scripts to generate a deterministic evidence package, after which the model writes the report module-by-module based on this package. The process includes: determining the trading date → generating the evidence package → modular generation (market overview, volume, profitability, feature groups) → star-rating of leading themes → report aggregation → lifecycle logging of themes → optional HTML visualization. It supports quantitative backtesting of feature groups and relative return factor mining, and can be invoked via natural language commands, such as "Analyze today's A-share market and generate a post-market report."

It is ideal for quantitative traders, data analysts, and financial researchers who need to review the A-share market daily, especially teams or individuals already using Tushare Pro data and looking to automate market report generation. For investors tracking market sentiment, trading volume concentration, and theme rotation, this Skill provides structured, evidence-driven analysis to avoid subjective assumptions.

It is recommended to use this tool uniformly after the market closes each day. By default, it uses only data from the specified date and earlier to prevent future information leakage. Note that it is primarily based on A-share daily data; it does not provide buy/sell recommendations or rely on pre-existing industry or concept classifications. Theme identification is strictly based on trading volume evidence. The scripts handle data retrieval and calculations, while report writing is performed by the model, so the model's quality directly impacts the readability of the final report.