Stock Agent

Automated Stock Analysis Pipeline with MCP, LangGraph, FastAPI, PostgreSQL, MongoDB, and OpenAI LLM

This project provides an agentic framework for fetching, analyzing, and storing stock and news data using a multi-container, multi-agent Python pipeline.

Features

• Ticker Resolution: Swarm agent that resolves company names to tickers using heuristics and LLM.
• Data Pipeline: Downloads live stock data, fetches news, and stores both in PostgreSQL and MongoDB.
• LLM Recommendation: Summarizes data and news using OpenAI GPT-4o for buy/hold/sell recommendations.
• LangGraph: Orchestrates the pipeline with a modular, agentic graph.
• Dockerized: One-command launch of all backend services.
• Extensible: Easy to add more agents or pipeline branches.

Getting Started

1. Install Docker Desktop:
   https://www.docker.com/products/docker-desktop/

2. Clone this repo:

   git clone https://github.com/radozaprazny/stock-agent.git
   cd stock-agent