This repository demonstrates a customer service use case using the OpenAI Agents SDK, featuring a Python backend for agent orchestration and a Next.js frontend for visualization and chat interaction. It's designed for developers and researchers looking to build and experiment with multi-agent customer service systems.

How It Works

The system employs a multi-agent architecture where a Triage Agent routes user requests to specialized agents (e.g., Seat Booking, Flight Status, FAQ, Cancellation). This modular approach allows for efficient handling of diverse customer service inquiries by delegating tasks to agents with specific capabilities, enhancing response accuracy and user experience.

Quick Start & Requirements

• Backend: cd python-backend && source .venv/bin/activate && pip install -r requirements.txt
• Frontend: cd ui && npm install
• Running:
  • Backend only: cd python-backend && python -m uvicorn api:app --reload --port 8000
  • UI & Backend: cd ui && npm run dev
• Prerequisites: OpenAI API key (set via environment variable OPENAI_API_KEY or .env file in python-backend). Python 3.x, Node.js, npm.

Highlighted Details

• Demonstrates agent routing for flight-related inquiries and cancellations.
• Includes examples of Relevance and Jailbreak guardrails to maintain conversation focus.
• Allows customization of agent prompts, guardrails, and tools for tailored workflows.

Maintenance & Community

Contributions via issues and PRs are welcome, though review is not guaranteed.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

The project is explicitly stated as being for demonstration purposes, suggesting potential limitations for production-ready deployments without further customization. Review of community contributions is not guaranteed.