simonlin1212/a-stock-data

What it solves

This project provides a comprehensive, unified data toolkit for the Chinese A-share market, designed specifically for AI programming assistants. It solves the fragmentation problem by aggregating data from 34 different sources—such as financial reports, K-line charts, market signals, and news—into a single, standardized format that AI models can easily parse and use for analysis.

How it works

It acts as a "Skill" file (structured Markdown with embedded Python) that can be injected into the context of AI coding assistants like Claude Code or Codex. It abstracts away the technical complexities of data retrieval, such as handling different authentication requirements, binary formats, and anti-scraping measures. The architecture is organized into 15 layers, prioritizing stable, zero-authentication sources (like Tencent and official exchanges) while providing fallback mechanisms if a primary source is blocked.

Who it’s for

This is for developers, quantitative researchers, and financial analysts who use AI agents to perform automated market research, stock analysis, or financial data processing on the Chinese stock market.

Highlights

• Unified Interface: Consolidates 87 data endpoints from 34 sources into a single, consistent toolset.
• AI-Ready: Designed for direct integration into AI coding assistants via context injection.
• Resilient Architecture: Features a 15-layer design with built-in fallback strategies to ensure data availability even if specific sources are blocked.
• Zero-Auth Focus: Prioritizes data sources that do not require API keys or user registration, simplifying deployment.
• Broad Coverage: Includes everything from real-time K-line charts and financial statements to sentiment analysis, news, and macro-economic indicators.