Thrive launches MCP: your learning platform just got a lot smarter

There's a new standard reshaping how AI tools work together, and Thrive is already ahead of it.

MCP (Model Context Protocol) is an emerging global standard for how AI agents communicate with software platforms. Rather than requiring a separate, bespoke integration every time an AI tool needs to talk to another system, MCP creates a shared language. A standardised way for AI to ask questions and get answers, in plain English, across different platforms.

Thrive has already launched its MCP server, and the implications for L&D teams are huge.

Hear it straight from our Chief Product and Technology Officer: Frankie Woodhead sat down with Learning News to talk through what MCP means for workplace learning.

‍

What MCP does

Until now, connecting AI tools to a learning platform meant navigating technical integrations that required developer time and a fair bit of patience. MCP changes that equation.

With Thrive's MCP server live, an L&D administrator using any AI assistant (whether that's Microsoft Copilot, Claude, or something else entirely) can now interact with Thrive through natural language alone. Ask it to:

• Surface all content that hasn't been viewed in two years. 
• Flag everything rated one out of five. 
• Make someone an admin. 

…or any other natural language commands you can think of. The AI carries out the action without the user needing to click through multiple screens or wait on a developer.

Put simply: if you can write the request, you can action it.

‍

Why this matters for learning teams

L&D teams have been asked to do more with less for years now. MCP is one of the more meaningful ways technology has stepped up to meet that challenge.

Content creation workflows that previously took weeks could soon involve AI agents working in parallel on behalf of your team. Imagine one agent drafting content based on what learners are actively searching for, while a second reviews it for quality and alignment with your wider content strategy. 

There's also a bigger shift at play. Learners increasingly expect to access information wherever they already are, without having to open a separate platform to find it. MCP makes Thrive available in those moments. Whether a learner is working inside Copilot or another AI-powered environment, they can pull up their upcoming learning, check their progress, or find a relevant course — all without leaving the tool they're already in.

‍

Built with control in mind

Thrive's MCP server is fully configurable. Organisations can choose what actions they're comfortable enabling from day one. This might look like read-only data retrieval to start, with broader capabilities switched on when the team is ready. Authentication, security and governance have been part of the design from the beginning, not added as an afterthought.

Getting set up is similar to implementing an API. You'll need some support from someone on your technical team, but once configured, organisations can typically be up and running within a day or two.

Early appetite from customers has been strong. Platforms that can show they're evolving alongside AI are the ones people want to be on. Thrive's single-architecture platform means AI agents can work across the full breadth of what Thrive offers, and that's a genuine advantage as MCP becomes the standard way the world of work gets things done.

Catch up on Frankie Woodhead’s interview here, and if you’re ready to see MCP in action, book a demo today. 

‍