rootSunc/CNEquity

What it solves

CNEquity addresses the difficulty of building and maintaining a reliable, long-term financial dataset for the Chinese market. It eliminates the need for repetitive data engineering by providing a local, versioned data lake that handles incremental updates, error recovery, and data quality auditing, ensuring researchers and AI agents have consistent, reproducible data.

How it works

It acts as a local data infrastructure that fetches financial data (stock prices, financial reports, corporate events, etc.) from various public sources and organizes them into a structured Parquet-based data lake. Users can interact with this data via a Python API, SQL queries, or by connecting the data lake to AI agents (like Claude or ChatGPT) via the Model Context Protocol (MCP), allowing the AI to perform complex financial analysis directly on the local dataset.

Who it’s for

This project is designed for individual financial researchers, quantitative analysts, and developers building AI-driven financial agents who need a stable, self-hosted, and historically accurate dataset for backtesting and analysis.

Highlights

• Comprehensive Data Coverage: Manages 55 datasets ranging from basic market data and financial statements to sentiment scores and derivative instruments.
• AI-Ready: Includes native support for the Model Context Protocol (MCP), enabling LLMs to query local financial data securely.
• Research-Grade Integrity: Preserves historical context (e.g., delisted stocks, point-in-time financial reporting) to prevent common research pitfalls like survivorship bias.
• Robust Engineering: Features automated incremental updates, failure recovery, and data quality checks to ensure the local lake remains current and accurate.
• Flexible Access: Supports querying via Python, SQL, or a built-in web-based console.