AI Inspiration WorldWhat it is
TradingAgents-astock is an open-source project by GitHub user simonlin1212, adapted from the upstream TauricResearch/TradingAgents and tuned for China A-shares. It is not the official upstream version. The README says it is an engineering implementation and research replication of a paper's framework, meant for research and teaching, and that it is not investment advice and offers no investment service. Several agents act as different analysts and write research reports on a stock, bull and bear researchers then debate, and a final rating with reasons is produced, with reports in Chinese. What sets it apart is that the data sources, analyst roles and trading rules were changed for A-shares.
Highlights:
- Built on the upstream framework, with data sources, analyst roles and trading rules adapted to A-shares
- 7 analyst roles, of which policy, hot-money tracking and lock-up monitoring were added for A-shares
- Bull and bear researchers debate before a rating and reasons are given
- Includes a web UI, with reports in Chinese
How to use:
- First, install it by following the quick start in the README and set up the Python environment.
- Then connect your own large language model service.
- Finally, enter a stock code to generate a research report, or use the web UI.
Who it's for:
People who want to study multi-agent research workflows, or do A-share research or teaching.
Before you start:
- You need Python 3.10 or later and a connection to a large language model service; see the README.
- Data comes from public market and news sites, so it goes online when used.
- This project is for research and teaching only. It is not investment advice; investing carries risk, so use your own judgment.
- If you use these features, weigh the safety risks yourself. AI Inspiration World is not responsible for any consequences.
- License: Apache-2.0.
Based on the project's public documentation. Not tested by AI Inspiration World.