mcpField
E
Experimental Seat Booking Agent

Experimental Seat Booking Agent

github.com/minyang-chenFrameworks & SDKs8Updated 83w ago
Repository

About Experimental Seat Booking Agent

Experimental seat booking agent combining OpenAI Agents SDK, Autogen, and MCP tool support. Demonstrates multi-agent coordination for airline seat reservation services.

Key features

  1. 01Triage agent routes inquiries to specialized agents
  2. 02Retrieves user seating and airline preferences
  3. 03Uses MCP servers for web fetching and file system access
  4. 04QA agent answers plane and onboarding questions
  5. 05Seat booking agent handles reservations

Use cases

  1. 01Self-service airport seat booking kiosk
  2. 02Multi-agent coordination demonstration
  3. 03Integrate MCP tools for external operations
  4. 04Chat interface for customer self-service

MCP support

Leverages MCP servers as tools to fetch web content and read files from the file system.

Details

From the project README

Experimental Seat Booking Agent

Experimental seat booking agent implementation combines OpenAI Agents SDK, Autogen extensions, and MCP tools support.

seat-booking-workflow-graph

Use Case

The self-service airline seat reservation kiosk, also known as a flight seat booking service agent or passenger service agent, is designed to provide comprehensive customer support and assistance. This kiosk ensures a smooth and positive travel experience by handling various tasks, including reservations, through its call center interface.

Key Features:
  • Triage agent answer user general inquiry with tools
  • Triage agent should route customer question to specialized agents
  • Triage agent retrieve user preferences on seating, airlines and food
  • Triage agent use agents as tools on question routing
  • Triage agent use MCP server as tools to perform web fetch content and read files system
  • QA agent has knownledge to handling questions about Plane, Flight and Onboarding
  • Seat booking agent specifilized on plane seat reservation tasks
  • Web Chat interface for user self-service
Lab Technical Stack
  • Linux
  • OpenAI Agents SDK
  • Autogen 0.4 + extensions (MCP Tools,OpenAI, Ollama)
  • Chainlit (UI)
  • Ollama (local Qwen2,5)
Environment
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
Running
chainlit run app.py
Todos
  • save seat history at the end of the booking
  • improve FAQ agent to use RAG on actual documents
  • improve memory storage
  • add MCP tools to invoke actual booking service api
  • more to came...

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Commit your changes
  4. Push to the branch
  5. Create a Pull Request

Important: Please keep your pull requests small and focused. This will make it easier to review and merge.

Feature Requests

If you have a feature request, please open an issue and make sure it is tagged with enhancement.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.