oficcejo/aiagents-stock

What it solves

This project provides a comprehensive investment decision support system that simulates a professional securities analyst team. It helps retail investors overcome the difficulty of analyzing vast amounts of market data by automating stock selection, deep fundamental and technical analysis, market monitoring, and quantitative trading across A-shares, Hong Kong stocks, and US stocks.

How it works

The system uses a multi-agent AI architecture where specialized agents (Technical, Fundamental, Capital, Risk, Sentiment, and News analysts) collaborate to analyze stocks and generate final decisions. It integrates multiple data sources (including AKShare, Tushare, TDX, and web-scraping from platforms like iwencai) and provides a web interface via Streamlit. For decision-making, it can utilize standard LLMs (like DeepSeek) or a structured decision engine (TypeSafe Jev) to provide precise ratings, confidence levels, and price targets.

Who it’s for

Retail investors and traders who want professional-grade AI-assisted analysis, automated stock screening based on specific strategies (e.g., low-cap or high-growth), and real-time monitoring of their portfolios with automated notifications.

Highlights

• Multi-Agent Analyst Team: Simulates a team of 6 AI analysts who discuss and reach a consensus for a final investment decision.
• Diverse Selection Strategies: Includes pre-built strategies for main force capital flow, low-price "bull" stocks, small-cap growth, and value investing.
• Macro & Sentiment Analysis: Monitors 20+ news platforms for sentiment and analyzes macro cycles using the Merrill Lynch Investment Clock and official government data.
• Quantitative Integration: Supports automated trading and position management via the MiniQMT interface.
• Automated Monitoring: Real-time alerts for entry/exit prices sent via Email, DingTalk, or Feishu Webhooks.