March 26, 2026AgentsOpen SourceTool

Dexter: Autonomous Financial Research Agent Trending on GitHub

Dexter, an autonomous agent for deep financial research, is trending on GitHub with 269 stars today. Built with TypeScript, Ink (React for CLI), and LangChain, it turns complex financial questions into clear, step-by-step research plans.

Dexter takes a financial question, breaks it into research tasks, executes them using live market data, checks its own work, and refines results until it has a confident, data-backed answer. It includes tools for financial metrics lookups, SEC filing analysis (10-K, 10-Q, 8-K), web search, and Playwright-based browser automation.

A key differentiator is self-validation: Dexter includes an evaluation suite that tests the agent against financial question datasets using LangSmith for tracking and LLM-as-judge scoring. All tool calls are logged to a scratchpad for full debugging and history tracking.

The agent supports multiple LLM providers β€” OpenAI (default), Anthropic, Google, xAI (Grok), OpenRouter, and Ollama for local execution. Built by Virat Singh, Dexter represents the growing category of domain-specific autonomous agents that go beyond general-purpose chat.

https://github.com/virattt/dexter
← Previous
oh-my-claudecode: Multi-Agent Orchestration Layer for Claude Code Hits GitHub Trending
Next β†’
T-MAP: Red-Teaming Framework for LLM Agents Targeting MCP Ecosystem
← Back to all articles

Comments

Loading...
>_