Dexter is an autonomous agent for deep financial research that decomposes complex financial questions into structured research plans and executes them using live market data.
The tool addresses the challenge of conducting thorough financial analysis by automating the research process. Rather than requiring manual data gathering and analysis, Dexter takes a user's financial question and breaks it into step-by-step research tasks. It then autonomously executes these tasks using real-time financial data including income statements, balance sheets, and cash flow statements. The agent validates its own work through self-reflection and iterates on results until confident in the answer. Built-in safety features prevent runaway execution through loop detection and step limits.
Dexter suits developers and researchers who need to automate financial analysis workflows and can work with the Bun runtime and required API keys from OpenAI, Financial Datasets, and optionally Exa for web search. The tool is positioned for educational and informational purposes rather than live trading. It includes an evaluation suite using LangSmith and LLM-as-judge scoring to test the agent against financial question datasets, and supports interactive mode as well as watch mode for development. The project also offers WhatsApp integration for query submission.
The project maintains a substantial base of adopters, with almost all open issues raised by outside users rather than the core team. Maintainers typically respond to new issues and pull requests within a day. Development work is dominated by enhancement requests in the issue tracker.
