> ## Documentation Index
> Fetch the complete documentation index at: https://kindling.birklid.com/llms.txt
> Use this file to discover all available pages before exploring further.

# TradingAgents

> Multi-agent LLM framework that replicates a trading firm — fundamental, sentiment, and technical analyst agents collaborate through structured debate.

<div style={{display: "flex", gap: "8px", marginBottom: "1.5rem", flexWrap: "wrap"}}>
  <Badge>Open Source</Badge>
  <Badge color="#F97316">Applications Layer</Badge>
</div>

# TradingAgents

**A simulated trading firm in code — specialized analyst agents debate investment decisions using real market data, news, and sentiment feeds.**

<Frame>
  <img src="https://mintcdn.com/tumbleweedlabs/QT0SlrwbzlJBSMcS/images/og-trading-agents.png?fit=max&auto=format&n=QT0SlrwbzlJBSMcS&q=85&s=7737cb08bf1a580c479dfd8bfddf60eb" alt="TradingAgents GitHub" width="1200" height="600" data-path="images/og-trading-agents.png" />
</Frame>

<CardGroup cols={4}>
  <Card title="Type" icon="code-branch">Open Source (Apache 2.0)</Card>
  <Card title="Stack Layer" icon="browsers">Applications</Card>
  <Card title="Language" icon="code">Python</Card>
  <Card title="Stars" icon="star">72k+</Card>
</CardGroup>

## What it is

TradingAgents is a multi-agent LLM framework that mirrors the structure of a professional trading firm. Specialized agents fill analyst roles — fundamental analysis, sentiment analysis, technical analysis, bull-case research, bear-case research — and a trader agent synthesizes their structured debates into investment decisions, subject to oversight from a risk management team. The system is built on LangGraph and supports multiple LLM backends including OpenAI, Anthropic, and Google.

Data ingestion covers historical price data, financial news, earnings reports, social media sentiment, and insider transaction filings. Compared to simple trading bots, the multi-agent debate structure provides interpretable reasoning: you can trace exactly why the system took a position. The 72k+ stars reflect broad interest from both the AI and quantitative finance communities.

<Tip>
  **Use this when** you want to prototype multi-agent approaches to financial research, build an explainable AI trading strategy that mirrors professional analyst workflows, or study how LLM agent collaboration performs on financial decision-making tasks.
</Tip>

## Get started

<CardGroup cols={2}>
  <Card title="tradingagents-ai.github.io ↗" icon="globe" href="https://tradingagents-ai.github.io/">
    Project site with architecture overview and demo.
  </Card>

  <Card title="GitHub ↗" icon="github" href="https://github.com/TauricResearch/TradingAgents">
    Source code, setup, and documentation.
  </Card>
</CardGroup>

## Related tools

<CardGroup cols={2}>
  <Card title="Dexter" icon="github" href="/library/finance/dexter">
    Autonomous financial research agent for deep equity analysis.
  </Card>

  <Card title="Alpaca" icon="globe" href="/library/finance/alpaca">
    Commission-free trading API for connecting strategies to live markets.
  </Card>
</CardGroup>