simonlin1212

simonlin1212 / TradingAgents-astock

Public

A股多Agent投研框架 — 适配A股数据源(龙虎榜/游资/解禁等)，7位分析师基于A股规则的辩论决策，基于TradingAgents深度改造，适配大A。A-share multi-agent investment research framework — 7 AI analysts, bull/bear debate, risk assessment。

arxiv.orgpdf2412.20138
58
13
100% credibility
Found May 13, 2026 at 58 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

TradingAgents-Astock is a free, open-source AI tool that analyzes Chinese A-shares using teams of specialized agents to produce investment research reports via a web dashboard or command line.

How It Works

1
🔍 Discover Smart Stock Helper

You find this free tool on GitHub that uses AI teams to analyze Chinese stocks and give clear buy/hold/sell advice.

2
📦 Get It Running Fast

With one simple command, you add it to your computer—no complicated setup needed.

3
🔗 Link Your AI Brain

You connect a smart AI service (like a thinking helper) so the tool can do its magic.

4
🚀 Open the Dashboard

Click to launch a friendly web page where everything happens right in your browser.

5
📝 Pick Your Stock

Type in a stock code like 300750 and choose a date, then hit start.

6
⚙️ Watch the AI Team Work

See 7 expert AI analysts debate trends, news, and risks in real-time progress bars.

7
✅ Get Your Clear Signal

Receive a simple Buy, Hold, or Sell recommendation with full reasoning report.

📊 Download and Decide

Save the polished PDF report and feel confident making smarter stock choices.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 58 to 58 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is TradingAgents-astock?

TradingAgents-astock is a Python multi-agent framework for A-share investment research, forking the popular TradingAgents to handle China stocks like dragon-tiger lists, hot money flows, and lockups. It deploys 7 specialized analysts—covering market trends, sentiment, news, fundamentals, policy, capital flows, and reductions—that feed into bull/bear debates and risk assessments for buy/hold/sell signals. Users get instant reports via CLI (`tradingagents`) or Streamlit web UI (`tradingagents-web`), with PDF exports and free data pulls from local sources.

Why is it gaining traction?

Unlike generic TradingAgents tuned for US stocks, this astock version nails A-share quirks with tailored analysts and trading rules like T+1 and price limits, pulling real-time data without APIs or paywalls. The debate-driven workflow—bull/bear research clashes plus aggressive/conservative risk rounds—delivers nuanced decisions fast, and pip-install setup with Docker support means devs prototype in minutes. Early adopters praise the Chinese-output reports and zero-external-deps simplicity.

Who should use this?

Quant traders or researchers tracking A-shares who want AI to sift dragon-tiger boards, northbound flows, and policy noise before manual dives. Python scripters building automated stock screeners, or finance devs testing multi-agent setups on real China data. Skip if you're US/EU-focused or need production-grade backtesting.

Verdict

Solid experiment for A-stock multi-agent research (58 stars), but 1.0% credibility signals early maturity—light tests, nascent community. Try for prototypes; pair with your own validation before live trading.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.