Repolode
KylinMountain

KylinMountain/TradingAgents-AShare

A-share (Chinese stock market) multi-agent investment research system where AI agents debate bull and bear cases and risk views before issuing trade suggestions.

Stars
890
Forks
244
Open issues
29
Last commit
9 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-AShare is an A-share stock analysis system built on a multi-agent design. Around 14 specialized agents cover fundamentals, sentiment, news, technicals, macro conditions and main-force capital flows, then pass their findings through structured bull-versus-bear debate and a three-way risk review before a portfolio manager issues a final decision with direction, confidence, target price and stop-loss.

The web app lets users enter requests in natural language, such as asking for a short-term view on a named stock. It also supports a watchlist with scheduled analysis, portfolio import with a tracking dashboard, and a history of stored reports. Each agent's debate can be followed in a visual timeline. LLM providers including OpenAI, Anthropic, Gemini, DeepSeek and Zhipu can be switched in the settings page.

The project ships as a Docker image or runs from source with Python and Node.js, and exposes a REST API protected by bearer tokens. It is also packaged as an OpenClaw skill. The project is licensed for noncommercial use and is intended for research and learning, not as investment advice.

Key features

  • 01Runs about 14 role-based agents covering fundamentals, sentiment, news, technicals, macro and capital flows.
  • 02Structures bull and bear debate and a three-way risk review before a final decision.
  • 03Accepts natural-language requests and detects the stock and holding period automatically.
  • 04Supports a watchlist with scheduled analysis that reuses portfolio context.
  • 05Imports portfolio positions and shows a tracking dashboard.
  • 06Works with multiple LLM providers, configured from the web settings page.
  • 07Exposes a REST API with token authentication and can be installed as an OpenClaw skill.

Use cases

  • Researching an A-share stock's short-term or medium-term setup with structured bull and bear arguments.
  • Running scheduled analyses on a watchlist and reviewing stored reports over time.
  • Integrating A-share analysis into custom scripts or dashboards through the REST API.
  • Studying how a multi-agent investment workflow is organized in code.

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/KylinMountain/TradingAgents-AShare.git
  2. 2

    Enter the project folder and set the app secret key

    cd TradingAgents-AShare && export TA_APP_SECRET_KEY=$(openssl rand -base64 32)
  3. 3

    Start the app with Docker Compose, then open http://localhost:8000

    docker compose up -d

Things to know

  • Focused on the A-share (Chinese) market only.
  • Licensed under PolyForm Noncommercial 1.0.0 for new modules, so commercial use is not permitted.
  • Needs your own LLM API key from a model provider to run analyses.
  • Outputs are for research only and are not investment advice.

In collections

Topics

FAQ

What is TradingAgents-AShare?
A-share (Chinese stock market) multi-agent investment research system where AI agents debate bull and bear cases and risk views before issuing trade suggestions. TradingAgents-AShare is an A-share stock analysis system built on a multi-agent design. Around 14 specialized agents cover fundamentals, sentiment, news, technicals, macro conditions and main-force capital flows, then pass their findings through structured bull-versus-bear debate and a three-way risk review before a portfolio manager issues a final decision with direction, confidence, target price and stop-loss.
Is TradingAgents-AShare free to use? What license does it have?
TradingAgents-AShare is open source under the Other license. Check the license terms before using it in commercial products.
Is TradingAgents-AShare still maintained?
Actively maintained. The last commit to TradingAgents-AShare was on Oct 30, 2026.
What can I use TradingAgents-AShare for?
Researching an A-share stock's short-term or medium-term setup with structured bull and bear arguments. Running scheduled analyses on a watchlist and reviewing stored reports over time. Integrating A-share analysis into custom scripts or dashboards through the REST API. Studying how a multi-agent investment workflow is organized in code.
How do I install TradingAgents-AShare?
Clone the repository: git clone https://github.com/KylinMountain/TradingAgents-AShare.git Enter the project folder and set the app secret key: cd TradingAgents-AShare && export TA_APP_SECRET_KEY=$(openssl rand -base64 32) Start the app with Docker Compose, then open http://localhost:8000: docker compose up -d
Where can I get TradingAgents-AShare?
The source code and installation instructions are on GitHub: https://github.com/KylinMountain/TradingAgents-AShare

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

Related projects