The NextPax MCP is live: our integration knowledge, inside your AI tools

The NextPax MCP is available now. It gives the AI tools your team already works in direct access to official NextPax integration knowledge: documentation, API specifications, step-by-step integration journeys and the real meaning behind our error responses. For both the Supply API and the Distribution API.

The slow part is rarely the code

Ask a developer who has connected a property management system to a distribution network where the time actually went, and it is almost never the code.

It went on a field whose purpose wasn't obvious from its name. On an error response that needed someone who knows the API by heart to decode. On a certification checklist that took longer to satisfy than the implementation it was checking.

A PMS provider on the Supply API and an OTA on the Distribution API hit different endpoints, but they lose time identically. The knowledge they need already exists: in documentation, in specifications, in the heads of our integration team. They just have to leave their work to go and find it.

The NextPax MCP closes that gap, not with more documentation, but by putting the documentation we already have inside the tool your developer has open right now.

What MCP actually is

Model Context Protocol (MCP) is an open standard that lets AI tools connect to external knowledge and systems through one consistent interface. It launched in late 2024 and has since been adopted across the industry, becoming the default way AI tools talk to business systems, much like the way solid API documentation became a baseline expectation a decade ago, rather than a differentiator.

For hospitality connectivity, that shift has a very concrete consequence. The AI tools your developers already work in can read official NextPax knowledge directly, instead of your team switching between a documentation portal, their editor and a support ticket, holding the thread together manually across all three.

The distinction that matters: socket, not chatbot

This is the part worth being precise about, because "we added AI" has become an easy thing to claim and a hard thing to make useful.

A chatbot sits between you and the system. It works from its own separate knowledge base, it is a second interface to learn, and when it doesn't know something it tends to produce a confident answer anyway.

The NextPax MCP is not that. It is a socket.

It plugs official NextPax knowledge straight into the AI tool your team has already chosen. Your AI then holds two things at the same time: your code and our documentation. That combination is where the value sits. Your AI doesn't quote a docs page back at you: it reads the error your integration just returned, reads what that error means in our specification, looks at the code that produced it, and tells you what to change.

There is no new dashboard, no separate portal, no additional login. One install, in whichever MCP-capable tool your team prefers.

Both sides of the exchange

The MCP covers the full NextPax connectivity surface, which means it serves two distinct audiences.

On the supply side: the Supply API. PMS providers connecting their system to NextPax distribution, and companies building their own property management system who want channel connectivity integrated from day one rather than bolted on later.

On the distribution side: the Distribution API. Channels and OTAs integrating with NextPax to source supply: properties, content, availability and rates.

And once you are live, it continues to support people who never touch the API at all. Support and operations teams get instant answers on how flows work, which steps to follow, and how specific situations should be handled, without filing a ticket or making an API call. It starts as a developer tool during integration and naturally becomes a knowledge tool for the wider team afterwards.

What your AI can do today

Answer integration questions. Endpoints, fields, flows, requirements: everything in the official NextPax documentation, in plain language, inside your own AI tool.

Explain and fix errors. Your AI combines the error meaning from our documentation with your actual code and suggests exactly what to change.

Guide you to certification. Step-by-step integration journeys take you from the first API call through to a certified, live connection.

Work in your tools. Claude, ChatGPT, Cursor, or any tool that speaks the open Model Context Protocol.

Channel onboarding journeys, included

Connecting to NextPax is one integration. Going live on a specific channel has its own requirements, its own sequence and its own recurring mistakes, and that knowledge is included too.

The MCP includes step-by-step onboarding journeys for all supported channels, including Booking.com, Airbnb, Vrbo, Expedia, Trip.com, Agoda, HomeToGo, Holidu, Homes & Villas by Marriott Bonvoy, and GDS connectivity through Amadeus, Sabre, and Travelport. Your AI walks your team through the requirements, the steps and the common errors for each of them, alongside the Supply API and Distribution API documentation.

Behind that sits the wider network: one integration, 150+ global channels, 650,000+ properties connected.

Get access

The NextPax MCP is rolling out to NextPax partners now. Plug it into Claude, ChatGPT or Cursor and give your team instant answers, from the first API call to daily operations.

Request access or talk to our integration team.