Repolode
Submit
simonlin1212

simonlin1212/TradingAgents-astock

A-share-specialized multi-agent LLM investment research framework, forked from TradingAgents, with seven analyst roles and bull/bear debate.

View on GitHubWebsite
Stars
3,918
Forks
985
Open issues
2
Last commit
8 days 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-Astock is a fork of the TradingAgents multi-agent research framework, adapted for the Chinese A-share market. It replaces the US-focused data sources with free direct-connect feeds such as mootdx, Eastmoney, Sina and Tonghuashun, and adds A-share-specific analysts covering policy, hot-money (龙虎榜) activity and lockup expirations.

The pipeline runs seven analysts, then a bull-versus-bear debate, a research manager, a trader, a three-way risk debate and a portfolio manager. A-share trading constraints such as T+1, price limits and lot sizes are built into the trader's logic. Reports are written in Chinese, while internal debate stays in English. Users pick LLM providers through configuration, and the project includes a Streamlit web UI, a CLI, and a performance-statistics command that measures past directional accuracy against the CSI 300.

The project is positioned as an engineering implementation for research and teaching. It does not provide investment advice, and its outputs are AI-generated and may contain errors.

Key features

  • 01Seven analyst roles covering market, sentiment, news, fundamentals, policy, hot money and lockup expirations.
  • 02Bull vs. bear debate followed by a three-way risk debate before final decision.
  • 03Free data sources with no API key required, including mootdx, Eastmoney, Sina and Tonghuashun.
  • 04A-share trading rules such as T+1, price limits and lot sizes built into the trader role.
  • 05Configurable LLM providers, including OpenAI-compatible gateways and per-role model assignment.
  • 06Streamlit web UI, CLI, and a performance command reporting directional accuracy against CSI 300.

Use cases

  • Studying how multi-agent debate behaves on financial text in the A-share market.
  • Learning how to wire Chinese market data sources into an LLM agent pipeline.
  • Running batch analyses of A-share tickers to generate structured Chinese reports for review.
  • Backtesting the historical directional accuracy of an LLM decision flow against the CSI 300.

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/simonlin1212/tradingagents-astock.git
  2. 2

    Enter the project directory

    cd tradingagents-astock
  3. 3

    Install the package in editable mode (Python 3.10 or newer)

    pip install -e .
  4. 4

    Create a .env file with your LLM provider API key, then run the analysis via the CLI

    tradingagents

Things to know

  • Each analysis makes roughly 30 to 50 LLM calls, so it requires a paid API key by default.
  • Output is for research and education only and is not investment advice.
  • Eastmoney requests are rate-limited, so large batch runs are slow.

In collections

Claude Code tools, skills & pluginsOpen-source AI agent frameworks & apps

Topics

#multi-agent#llm#a-share#investment-research#quantitative-finance#langgraph#china-stocks#python#ai-agent#claude

FAQ

What is TradingAgents-astock?
A-share-specialized multi-agent LLM investment research framework, forked from TradingAgents, with seven analyst roles and bull/bear debate. TradingAgents-Astock is a fork of the TradingAgents multi-agent research framework, adapted for the Chinese A-share market. It replaces the US-focused data sources with free direct-connect feeds such as mootdx, Eastmoney, Sina and Tonghuashun, and adds A-share-specific analysts covering policy, hot-money (龙虎榜) activity and lockup expirations.
Is TradingAgents-astock free to use? What license does it have?
TradingAgents-astock is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is TradingAgents-astock still maintained?
Actively maintained. The last commit to TradingAgents-astock was on Oct 30, 2026.
What can I use TradingAgents-astock for?
Studying how multi-agent debate behaves on financial text in the A-share market. Learning how to wire Chinese market data sources into an LLM agent pipeline. Running batch analyses of A-share tickers to generate structured Chinese reports for review. Backtesting the historical directional accuracy of an LLM decision flow against the CSI 300.
How do I install TradingAgents-astock?
Clone the repository: git clone https://github.com/simonlin1212/tradingagents-astock.git Enter the project directory: cd tradingagents-astock Install the package in editable mode (Python 3.10 or newer): pip install -e . Create a .env file with your LLM provider API key, then run the analysis via the CLI: tradingagents
Where can I get TradingAgents-astock?
The source code and installation instructions are on GitHub: https://github.com/simonlin1212/TradingAgents-astock

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

Related projects