WyckoffTradingAgent: AI Trading Agent and Stock Screener Based on Volume-Price Analysis

WyckoffTradingAgent is an open-source intelligent trading agent system that combines classic Wyckoff trading theory with modern artificial intelligence technology to provide automated volume-price analysis and stock screening capabilities for the A-share market. The system supports command-line interface (CLI) operations and integration with MCP (Model Context Protocol) tools, making it a complete trading workflow platform.

The Wyckoff method was proposed by Richard Wyckoff in the early 20th century, with its core being the identification of market supply-demand balance through volume-price relationships. Its core concepts include the Law of Cause and Effect, Effort vs. Result, Relative Strength Analysis, and Volume-Price Relationships; market cycles are divided into four phases: Accumulation, Markup, Distribution, and Markdown.

The system includes a volume-price analysis engine (identifying Wyckoff features such as Spring and Upthrust), an AI stock screener (phase identification, relative strength ranking, etc.), CLI workflow support (batch scanning, custom conditions, etc.), and MCP tool integration (natural language queries, intelligent interpretation, etc.).

Data acquisition and processing cover historical K-lines, real-time market data, and fundamental data; pattern recognition algorithms include automatic identification of support and resistance levels, pattern matching, and multi-timeframe analysis; machine learning enhancements involve feature engineering, classification models, and anomaly detection.

The system can be used for trend following (identifying strong stocks in the Markup phase), bottom reversal capture (Spring effect in the Accumulation phase), top risk warning (Upthrust signal in the Distribution phase), and quantitative strategy development (as a strategy building block).

Open source value includes educational significance (coding classic theory), scalability (modular design), transparency (auditability), and community collaboration (contributing new indicators/strategies).

Limitations include: past performance does not guarantee future results; unique characteristics of A-shares (price limits/policy interventions); need for auxiliary human judgment; and dependence on data quality.

WyckoffTradingAgent is a typical case of integration between traditional technical analysis and modern AI. It transforms Wyckoff theory into an automated tool, providing A-share investors with a research and decision support platform to enhance market analysis capabilities.