How to Use the Duffel Flights MCP in OpenAI Agents SDK

Q: How do I initialize the Duffel Flights MCP Server in my OpenAI Agents SDK project?

Install the package using pip install openai-agents and set up the server streamable HTTP parameters. Instantiate MCPServerStreamableHttp pointing to your Vinkius endpoint to connect the MCP Server, then pass it directly into your Agent constructor within an async context manager.

Q: How does Duffel Flights handle token expiration with the OpenAI Agents SDK?

Flight offers returned by listflightoffers are valid for a limited window, so your agent must execute createflightorder before they expire. If a delay occurs, the OpenAI Agents SDK can catch the API error and automatically run a new createflightofferrequest to refresh the pricing.

Q: Is caching supported for static data like airline codes?

You can set cacheToolsList=True in your server parameters to keep tool definitions cached. For static endpoints like listaircraft, you should cache the results locally in your Python application to avoid hitting Duffel rate limits.

Build reliable, guarded booking agents with the OpenAI Agents SDK and Duffel Flights.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Duffel Flights MCP to OpenAI Agents SDK

Create your Vinkius account to connect Duffel Flights to OpenAI Agents SDK and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Duffel Flights with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Live inventory searches via OpenAI Agents SDK

Your agent uses `create_flight_offer_request` to query Duffel for real-time ticket availability, origin airports, and destination routing. It feeds the raw search parameters directly from conversational user input into the API, pulling live seat options without manual form filling. Once the initial query runs, the agent calls `list_flight_offers` to display the exact ticket options, pricing, and baggage allowances. This pipeline matches your OpenAI Agents SDK schema validation, keeping flight search payloads structured and clean before displaying options to travelers.

Book passenger tickets securely with this MCP Server

The `create_flight_order` tool acts as the core booking mechanism, turning selected flight offers into confirmed reservations. By passing passenger details and flight keys directly through the OpenAI Agents SDK, you bypass the need for external database lookups during the booking phase. After reserving the seats, the agent triggers `pay_for_flight_order` to finalize the transaction. This two-step process fits perfectly into the SDK's built-in guardrails, letting you inspect the booking state before charging the customer's card.

Resolve airport codes and airline fleets instantly

Resolving three-letter IATA codes into full airport names and locations is handled instantly by the `list_airports` tool. Instead of hardcoding static airport databases, your OpenAI Agents SDK system queries this live endpoint to verify destination codes during a chat. To give travelers complete context, the agent runs `list_airlines` and `list_aircraft` to detail which carrier is operating the route. This metadata lets your agent system use this MCP tool to explain exactly what plane they will fly on, right inside the conversational thread.

Setup guide

Set up Duffel Flights MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all Duffel Flights tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives Duffel Flights tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Duffel Flights tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="Duffel Flights Agent",
            instructions="You have access to Duffel Flights tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Duffel. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Duffel Flights now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Duffel Flights MCP in OpenAI Agents SDK

Install the package using `pip install openai-agents` and set up the server streamable HTTP parameters. Instantiate `MCPServerStreamableHttp` pointing to your Vinkius endpoint to connect the MCP Server, then pass it directly into your `Agent` constructor within an async context manager.

Yes, you can control tool access by defining specific agent handoffs or by filtering the tools exposed during initialization. This prevents a customer-facing agent from calling sensitive billing tools like `pay_for_flight_order` without human approval.

Flight offers returned by `list_flight_offers` are valid for a limited window, so your agent must execute `create_flight_order` before they expire. If a delay occurs, the OpenAI Agents SDK can catch the API error and automatically run a new `create_flight_offer_request` to refresh the pricing.

You can set `cacheToolsList=True` in your server parameters to keep tool definitions cached. For static endpoints like `list_aircraft`, you should cache the results locally in your Python application to avoid hitting Duffel rate limits.

Your passenger names, passport numbers, and payment tokens are sent directly to Duffel through the MCP pipeline into Vinkius's secure, zero-trust V8 sandbox. The OpenAI Agents SDK merely acts as the orchestration layer, meaning sensitive booking payloads are never stored on OpenAI's servers or used for model training.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript