Summary This project offers a Wyckoff analysis and trading system for A-shares, integrating quantitative screening, AI insights, and risk control. It targets retail investors seeking a transparent, AI-assisted approach, providing a cloud-deployed Streamlit web UI and an automated backend, minimizing local setup needs.

How It Works The system uses a Streamlit web UI and a GitHub Actions backend. Core components include a multi-layer Wyckoff "Funnel" for quantitative screening and AI deep dives via LLMs for stock assessment. RAG filters negative news sentiment. Results are pushed via webhooks or Telegram, with optional personal portfolio management via Supabase.

Quick Start & Requirements

• Primary Install: Python 3.10+ required. Clone repo, pip install -r requirements.txt.
• Prerequisites: Essential API keys: Supabase (URL, Key), Tushare Token, LLM provider key (e.g., Gemini). Optional: notification keys (Feishu, Telegram), news search key (Tavily).
• Run: streamlit run streamlit_app.py for web UI.
• Links: A cloud-deployed web page is directly accessible. Links to a Wyckoff skills repository and a Tushare registration page are provided within the project documentation.

Highlighted Details

• Wyckoff Funnel: Multi-layer screening (Layers 1-5) identifying A-share candidates via volume-price analysis and Wyckoff principles.
• AI Deep Dives: LLM-based analysis providing multi-party judgments and risk-off strategies.
• RAG News Filtering: Integrates news search (Tavily) to filter stocks with negative sentiment.
• Automated Workflows: GitHub Actions for scheduled daily screening, AI reports, and notifications.
• Personalized Trading: Dynamic portfolio management, AI stock replacement suggestions, and private Telegram recommendations.

Maintenance & Community No specific community channels (e.g., Discord, Slack) are listed. Contact is primarily via WeChat/Feishu QR codes or GitHub Issues.

Licensing & Compatibility Copyrighted by "youngcan" with "All Rights Reserved." Free for personal learning/research with attribution. Commercial use requires prior authorization and licensing fees. Unauthorized use is infringement.

Limitations & Caveats Full functionality depends on configuring multiple API keys (Supabase, LLM, Tushare), posing a setup barrier. Strict licensing prohibits commercial use without explicit, paid authorization. Advanced features may require local setup and secret management. The project's copyright date (2026) is unusual.