Repolode
Submit
hsliuping

hsliuping/TradingAgents-CN

Multi-agent LLM research assistant for China's A-share stock market, with analyst agents that debate, natural-language screening, and paper trading.

View on GitHub
Stars
32,821
Forks
6,842
Open issues
298
Last commit
2 months ago

Repolode describes open-source projects and does not give investment advice. Trading and investing can lose money: assess the risk yourself and follow the laws where you live before using this project.

What is it?

TradingAgents-CN is a Chinese-market adaptation of a multi-agent LLM trading framework. Several AI agents take on different roles, such as market, sector, fundamentals and news analysts plus optimistic and cautious researchers, and collaborate through a debate-style research process on A-share stocks.

Beyond single-stock analysis at five depth levels, the system offers general research on natural-language questions, stock screening from a database of over 6,000 stocks, an AI assistant with web search and tool calling, and paper trading with review. A built-in learning center includes free courses and study guides. The backend is built on FastAPI with a Vue 3 frontend, backed by MongoDB and Redis, which the user installs separately. Market and financial data come from Tushare, AKShare and BaoStock, and users supply their own tokens and LLM API keys.

The project is for learning and research only and explicitly does not provide investment advice. The source is under a hybrid license: the core agent library is Apache-2.0, while the application layer is source-available with restrictions on redistribution and commercial use.

Key features

  • 01Debate-style multi-agent analysis of single stocks at five depth levels
  • 02Natural-language stock screening across more than 6,000 A-share stocks
  • 03AI assistant with web search, tool calling and persistent fact memory
  • 04Custom data-fetching skills through a Skill center
  • 05Paper trading with review, plus a free learning course
  • 06Market data from Tushare, AKShare and BaoStock

Use cases

  • Learning how multi-agent LLM frameworks are applied to stock research
  • Exploring A-share data with natural-language questions and screening
  • Practicing investment analysis workflows with paper trading
  • Studying prompt design and agent orchestration on a real finance task

Quick start

  1. 1

    Install and start MongoDB 7.0+ and Redis 6+ yourself, matching the credentials in .env

  2. 2

    Create a Python 3.11 virtual environment and install backend dependencies

    pip install -r requirements.txt
  3. 3

    Copy the example config to .env and fill in the required MongoDB, Redis and JWT_SECRET values

    cp .env.example .env
  4. 4

    Start the backend on port 8000

    uvicorn app.main:app

Things to know

  • The license is hybrid: the application layer (app/, frontend/, core/) is source-available and prohibits redistribution and commercial use.
  • It targets only the China A-share market.
  • You must install and run MongoDB and Redis yourself.
  • Using it requires your own LLM API key and market-data tokens such as Tushare.

In collections

Open-source AI agent frameworks & apps

Topics

#multi-agent#llm#stock-analysis#a-share#fintech#python#fastapi#vue

FAQ

What is TradingAgents-CN?
Multi-agent LLM research assistant for China's A-share stock market, with analyst agents that debate, natural-language screening, and paper trading. TradingAgents-CN is a Chinese-market adaptation of a multi-agent LLM trading framework. Several AI agents take on different roles, such as market, sector, fundamentals and news analysts plus optimistic and cautious researchers, and collaborate through a debate-style research process on A-share stocks.
Is TradingAgents-CN free to use? What license does it have?
TradingAgents-CN is open source under the Other license. Check the license terms before using it in commercial products.
Is TradingAgents-CN still maintained?
Updated within 6 months. The last commit to TradingAgents-CN was on Sep 22, 2026.
What can I use TradingAgents-CN for?
Learning how multi-agent LLM frameworks are applied to stock research Exploring A-share data with natural-language questions and screening Practicing investment analysis workflows with paper trading Studying prompt design and agent orchestration on a real finance task
How do I install TradingAgents-CN?
Install and start MongoDB 7.0+ and Redis 6+ yourself, matching the credentials in .env Create a Python 3.11 virtual environment and install backend dependencies: pip install -r requirements.txt Copy the example config to .env and fill in the required MongoDB, Redis and JWT_SECRET values: cp .env.example .env Start the backend on port 8000: uvicorn app.main:app
Where can I get TradingAgents-CN?
The source code and installation instructions are on GitHub: https://github.com/hsliuping/TradingAgents-CN

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects