Medical Conversation Agent with LangGraph

This project implements a LangGraph-based conversational agent that simulates a medical consultation between a patient and doctor. The system uses two specialized subagents that interact in turns until the consultation is resolved.

🏥 Features

• Patient Agent: Asks medical questions, describes symptoms, and seeks clarification
• Multi Elyx Team Agent: Provides medical advice, asks follow-up questions, and can mark consultations as resolved
• LangGraph Orchestration: Manages the conversation flow and turn-taking
• Configurable: Support for different LLM models and conversation parameters
• Interactive & Batch Modes: Run single conversations or test with multiple scenarios

🚀 Quick Start

1. Install Dependencies

pip install -r requirements.txt

2. Set up Environment Variables

Copy the example environment file and add your API key:

cp .env.example .env

Edit .env and add your Groq API key:

GROQ_API_KEY=your_actual_api_key_here

3. Run the Agent

Interactive Mode (default):

python3 run_enhanced_synthesizer.py

• Add your groq api keys for robin round key method to avoid rate limits.