By Jonathan Nyborg
May 21, 2026
The WURFL Gateway MCP connector puts live WURFL device data inside your AI assistant. You ask questions in plain English; it queries the database and hands you answers. No API setup, no need to learn the query syntax, no need to be the team member who usually interacts with WURFL.
The following examples show the connector in action. Below, we’ll dive into the broader context and the steps to get you up and running.
“I have a User-Agent string. Tell me what it is.”
There are plenty of reasons you might need to know what device a User-Agent string actually represents. Maybe you’re trying to make sense of a log file, or troubleshooting a customer ticket where the device type is the missing piece.
Either way, if you don’t have a WURFL API set up somewhere, decoding a UA string by hand isn’t straightforward. Hand it to your AI assistant instead, and the WURFL Gateway MCP handles the query.
Example: What device is this, and what are its form factor, screen resolution, OS version, and whether it’s a smartphone: Mozilla/5.0 (Linux; Android 13; SM-S918B) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/112.0.0.0 Mobile Safari/537.36
“I have criteria. Find me the devices.”
If you do not have a specific User Agent string but have an idea of the capabilities of the device(s) you are looking for, you can ask WURFL to find matching device UA strings. Imagine a scenario where you’re building a feature gated behind display density. Before you write a test plan, you need to know which devices actually qualify: recent Android flagships, excluding brands your team doesn’t test against. Getting that information used to mean writing an API query yourself, with some time spent in the capability documentation to figure out which filters exist and what values they accept. The WURFL Gateway MCP connector handles that complexity for you. You describe what you want in plain language; it runs the query and returns the matches. You can refine from there: tighten the filters if needed, change the form factor, add additional brand constraints, all in natural language, in the same session.
Example: Find Android smartphones released after 2022 with a screen density above 400 PPI, excluding Xiaomi. Show the top results with their brand, model, screen resolution, and release date.
“I want to understand a brand’s presence in WURFL.”
Sometimes you need a quick read on what a brand’s footprint actually looks like. How many devices, which form factors, which OS versions? It’s the kind of question that’s easy to answer roughly and hard to answer precisely.
Brand-presence questions usually come from people who don’t run WURFL queries themselves: product trying to scope a feature, sales responding to an RFP, support sizing the impact of a customer report. Until now, getting a precise answer meant pinging an engineer and waiting. The Gateway MCP changes that. Anyone with access to the AI assistant can request WURFL information directly and get the answer they need.
Example: Give me a breakdown of Samsung’s devices in WURFL by form factor and operating system.
What’s in the rest of the box
The examples above only scratch the surface. The WURFL Gateway MCP exposes a full set of tools: device lookups, capability filters, variant discovery, and side-by-side comparisons. Your AI assistant chooses among them based on what you’re asking. You don’t need to know the tool names. You don’t need to know WURFL. You can just ask.
Here are two more examples to round out the picture:
Variants. Given a device, pull all its hardware variants. That’s going to include regional SKUs, carrier models, and various storage tiers. The same “phone” often appears in WURFL as a dozen distinct entries, each with different capabilities. When your analytics and WURFL results don’t agree, this is where you start.
Example: Find all regional variants of the Samsung Galaxy S24 Ultra.
Comparisons. Two or more devices, side by side, on whatever capabilities you care about. The kind of thing that used to mean multiple browser tabs and mental math. Handy when you’re scoping how a feature will behave across a target device set.
Example: Compare the Samsung Galaxy S24 Ultra and Google Pixel 9 Pro: screen resolution, physical screen size, and OS version.
The shift, and what we’re doing about it
AI is changing how developers work. That is not news anymore, it’s just the environment. And it’s no longer only developers using these tools. We’re meeting that shift with MCP connectors that put WURFL where the questions are getting asked.
How to get started
A quick note on context: this is the second MCP connector we’ve released. The first, the WURFL Documentation MCP Server, taught your AI assistant to talk about WURFL: capability definitions, User-Agent decoding, and setup help. The WURFL Gateway MCP lets your assistant talk to WURFL: live catalog search, filters, comparisons. It’s powered by QuWURFL, so a subscription is required. Existing QuWURFL customers can jump straight to the setup guide. Not yet a customer? We’d love to chat to get you access.