Connect your AI Agents to Duffel in minutes

What is a Duffel MCP?

It's an MCP server that connects your agents to Duffel's travel booking API via tools. Your agents can invoke these tools to search for flights, book orders, manage itinerary changes, process cancellations, and more.

Duffel doesn't offer an MCP server purpose built for its platform, but you can use one from a third-party platform, like Merge Agent Handler.

How can I use the Duffel MCP server?

The use cases naturally depend on the agent you've built, but here are a few common ones:

• Automated flight booking: When a traveler submits a trip request via an internal tool or form, an agent searches for available flights through Duffel, selects the best option based on company travel policy, and completes the booking automatically

• Trip change handling: When a traveler reports a schedule conflict via Slack, an agent retrieves their current order from Duffel, fetches available change offers, and presents options before confirming the modification

• Cancellation and refund processing: When a trip is cancelled in a travel management system, an agent creates and confirms the cancellation in Duffel, then logs the refund details to the expense platform

• Itinerary lookup for travel assistants: A corporate travel assistant agent retrieves a traveler's confirmed order from Duffel, including flight segments, seat assignments, and passenger details, and surfaces a clean summary in a chat interface

What are popular tools for Duffel's MCP server?

Here are some of the most commonly used tools:

• create_offer_request: initiates a flight search by submitting origin, destination, dates, and passenger details to Duffel. Call this at the start of any booking workflow to retrieve a set of available fare options

• list_offers: retrieves the available fares returned from a completed offer request. Use this when an agent needs to evaluate and rank flight options before selecting one to book

• create_order: books a selected offer by creating a confirmed order with passenger details and payment. Call this when the agent is ready to finalize a booking after a fare has been chosen

• get_order: fetches the full details of a confirmed booking, including flight segments, passenger info, and order status. Helpful when an agent needs to surface itinerary details or verify order state before taking a downstream action

• create_order_cancellation: initiates a cancellation request for an existing order. Use this when an agent detects a trip cancellation event in another system and needs to begin the Duffel-side process

• confirm_order_cancellation: finalizes a pending cancellation and processes the refund. Call this after reviewing cancellation conditions to complete the flow and record the outcome in the booking system

What makes Merge Agent Handler's Duffel MCP server better than alternative Duffel MCP servers?

Several factors make Merge Agent Handler's Duffel MCP server worth choosing over a custom integration or community build:

• Enterprise-grade security and DLP: Merge Agent Handler includes built-in data loss prevention controls that let you block or redact sensitive fields before they reach an agent. For Duffel, this means you can prevent passenger details, payment information, and booking records from being exposed even when the agent has broad access to the orders API

• Managed authentication and credentials: Merge stores and refreshes Duffel credentials on your behalf. You never expose raw API keys or OAuth tokens to an agent or manage token rotation manually

• Real-time observability and audit trail: Every tool call made against Duffel is logged with timestamp, tool name, input parameters, and response metadata. Travel ops and finance teams can audit exactly what an agent booked, changed, or cancelled without any custom instrumentation

• Tool Packs and controlled access: Tool Packs let you bundle specific Duffel tools with tools from other connectors into a single MCP endpoint, scoped to a specific use case. An agent gets exactly the tools it needs, nothing more

How can I start using Merge Agent Handler's Duffel MCP server?

You can take the following steps:

1. Create or log into your Merge Agent Handler account.

2. Install the Merge CLI by running pipx install merge-api, then run merge configure to link the CLI to your Merge account and merge login to authenticate your session.

3. Register the Agent Handler MCP server with Claude Code by running claude mcp add --transport http agent-handler https://ah-api.merge.dev/mcp, then open Claude Code and run /mcp to confirm agent-handler appears with a connected status.

4. Select agent-handler from the MCP list. This opens a browser window where you select which integrations to authenticate. Choose Duffel and complete the auth flow. Merge stores and manages the credentials going forward.

5. Open a Claude Code session and start querying Duffel data directly. The first time you use a Duffel tool, a Magic Link may appear to complete connector authentication.

If you want to connect Merge Agent Handler's Duffel MCP with internal or customer-facing agentic products, you can follow the steps in our docs.

Can employees use Merge Agent Handler to connect their AI tools to Duffel?

Yes, Agent Handler for Employees lets your employees connect Claude, ChatGPT, Microsoft Copilot, Cursor, and other MCP-compatible AI tools to Duffel without bypassing IT governance.

Instead of setting up direct connections with personal credentials that IT can't monitor or revoke, each employee authenticates through Agent Handler and gets individual credentials tied to their identity.

IT also provisions access by role or group via SCIM. A travel manager, for example, gets Duffel access to book and manage flights, Expensify to track and reconcile travel costs, and Slack to coordinate itinerary approvals; while a travel coordinator gets Duffel access to handle booking changes and cancellations, Google Calendar to sync trip dates with team availability, and Gmail to manage traveler communications.

Every tool call an employee's AI makes to Duffel is also inspected against your DLP rules and logged to a searchable audit trail, giving security teams full visibility into what data was accessed and by whom.