Read the source.
Everything you need to install, configure, and reason about Trading Agents Lab. The same documentation that ships inside the desktop app's Help menu, published here so anyone can learn from it, not just users.
Coming from the open-source project? Run TradingAgents as a desktop app. Designed as a teaching artifact for Clawdemy and other AI-education programs. If you're here to understand agentic systems, start with How it works and Reading the debate.
Start here
Install the desktop app, run your first diligence, and learn to read what the agents produce.
Getting Started
How to clone the repo, set up the engine, install the desktop, and run Trading Agents Lab for the first time.
How It Works
A conceptual walkthrough of the multi-agent debate pipeline, from ticker input to final decision.
Reading the Debate
A guided tour of the Analyze page, every element explained, from the input form to the committee assessment.
Models & providers
Pick a model, run local with Ollama or LM Studio, use a ChatGPT plan via OAuth, and keep token cost in check.
Configuring LLM Providers
How to connect OpenAI, Anthropic, OpenRouter, Google Gemini, xAI Grok, and MiniMax; how to pick which provider and model each debate uses; how keys are stored.
Local LLM (Ollama / LM Studio)
Run the multi-agent debate entirely on your own machine with Ollama or LM Studio: no API key, no per-token bill, no data leaves your computer.
ChatGPT OAuth
Use your paid ChatGPT subscription to power Trading Agents Lab debates, no API key, no per-token billing.
Cost Guard
Stacked daily / weekly / monthly USD caps and an optional sessions-per-day rate cap. Optional. Off by default. Override per-run.
Data & markets
Wire up market data, run any ticker including crypto, and fold sentiment into the debate.
Data Providers
Where market data comes from, what the engine fetches, and how to add Alpaca for power-user data access.
Crypto tickers
Analyze crypto pairs the same way you analyze equities. BTC, ETH, BTC/USD, BTC-USD all work; the engine routes to the right data source automatically.
Social sentiment (StockTwits + Reddit)
The sentiment_analyst agent grounds its analysis in real social data, bullish/bearish ratios from StockTwits and recent discussion from finance subreddits. No API key required.
Connect & automate
Push results to webhooks, connect the Clawless desktop agent, and move faster with shortcuts.
Webhooks
Push every completed debate's committee assessment to your own systems, Telegram, Slack, Discord, or any HTTPS endpoint.
Clawless Connector
What the optional Clawless gateway tap is, how it fits into Trading Agents Lab, and when it activates.
Keyboard Shortcuts
Every keyboard shortcut in Trading Agents Lab, menu accelerators and page-level shortcuts.
Operate
Where your keys and runs live, how to fix common issues, and answers to frequent questions.
Security and Storage
Where Trading Agents Lab stores data on disk, how secrets are protected, and what to do when you change machines.
Troubleshooting
Common problems and how to fix them.
Frequently Asked Questions
Posture, license, what Trading Agents Lab is and is not, and the relationships with upstream and Clawless.