TradingAgents-CN

Why we included this project

This is a Chinese-localized build of the TradingAgents multi-agent framework, and it shows what a full analytical workflow looks like when several specialized LLM agents are wired together instead of a demo. The system splits the work across market, fundamental, and news analysts, with a final layer that weighs their findings into a stock research report. The maintainers added a FastAPI backend, a Vue frontend, MongoDB and Redis storage, Docker images for amd64 and arm64, and support for multiple LLM providers such as OpenAI, Google, and VolcEngine Ark. It covers A-share, Hong Kong, and US markets, and the project is explicit that it is a learning and research platform, not a live-trading tool, so it suits developers and small teams who want to study or extend a complete multi-agent research pipeline. There is also a substantial Chinese learning center covering prompt engineering, model selection, and multi-agent analysis principles, which makes it a practical reference for anyone building agent-based financial analysis systems. One thing to check before deploying: the codebase uses a hybrid license, and the app and frontend directories require a commercial license.