daily_stock_analysis: an automated AI stock analysis system that delivers daily decision dashboards across global markets

What it solves

This project provides an automated system for analyzing stocks across multiple global markets (A-shares, Hong Kong, US, Japan, Korea, and Taiwan) and delivering daily "decision dashboards" to various messaging platforms. It eliminates the need for manual data gathering and synthesis by using AI to generate actionable insights, risk alerts, and catalyst identification for a user's selected stock list.

How it works

The system aggregates data from multiple sources, including real-time market quotes, K-line charts, technical indicators, news, announcements, and fundamental data. It then processes this information using Large Language Models (LLMs) via various API providers (such as OpenAI, Gemini, Claude, or DeepSeek) and search tools (like SerpAPI or Tavily) to perform sentiment analysis and trend prediction. Users can deploy the system via GitHub Actions for zero-cost automation, Docker, or local installation, and receive reports via platforms like WeChat, Feishu, Telegram, Discord, Slack, or email.

Who it’s for

Retail investors and traders who want a structured, AI-driven daily summary of their watchlists without manually tracking news and technical indicators across different markets.

Highlights

• Multi-Market Support: Covers A-shares, HK, US, Japan, Korea, and Taiwan stocks as well as ETFs.
• AI Decision Reports: Generates core conclusions, scores, trends, buy/sell points, and risk alerts.
• Agentic Querying: Includes a chat interface with 15 built-in strategies (e.g., Moving Average, Elliott Wave) for multi-turn questioning about specific stocks.
• Automated Delivery: Native integration with GitHub Actions for scheduled daily reports.
• Comprehensive Web UI: A workspace for manual analysis, task monitoring, history, and portfolio management.