simonlin1212/global-stock-data

What it solves

This project provides a comprehensive toolkit for accessing US and HK stock market data, eliminating the need for developers to manually handle complex authentication flows, ticker mappings, and fragmented data sources. It is specifically designed to be integrated into AI coding assistants (like Claude Code, Codex, or OpenClaw) as a "skill," allowing the AI to fetch real-time quotes, financials, and filings directly via Python code.

How it works

The project is structured as a single self-contained "Skill" file (SKILL.md) containing Markdown and embedded Python. It uses the requests library to fetch data from 11 different sources, including official government feeds (SEC EDGAR, US Treasury, CFTC) and financial platforms (Yahoo Finance, CBOE, FINRA, Eastmoney, Sina, Tencent). It includes a built-in thread-safe rate limiter to ensure compliance with source limits, such as the SEC's 10 requests per second cap.

Who it’s for

Developers and AI users who want to automate financial analysis, track options flow, or screen for companies based on SEC filings, and who are use AI coding assistants to execute these tasks.

Highlights

• Comprehensive Data Coverage: Access to real-time quotes, K-lines (back to 1984), fundamentals, and macro data (Treasury yield curve, COT reports).
• Official-Source-First: Includes high-fidelity data from CBOE (options Greeks and 0DTE flow), FINRA (market-wide short volume), and SEC EDGAR (filings and XBRL metrics).
• AI-Native Integration: Designed as a skill for AI assistants, allowing them to call endpoints for tasks like "Show NVDA's 0DTE options flow."
• Compliance Tiering: Clearly labels data sources by their commercial and redistribution terms (Tiers S, B, and C).
• Zero-Auth: Most endpoints require no API keys, with Yahoo's cookie/crumb authentication handled automatically.