Sandbox release disclaimer: Some content in this article outlines how to use functionalities that will be publicly released to all clients at the end of July 2026. If you see differences between this article and your platform, check back at that time. Learn more about the release cycle.
Introduction
The Model Context Protocol (MCP) server allows you to seamlessly integrate Docebo with enterprise AI assistants (MCP clients) like ChatGPT, Claude, Microsoft Copilot, and Gemini.
By leveraging this integration, you can interact with your platform's data using natural language — asking questions, pulling reports, and managing learning without navigating through menus.
This article outlines who can benefit from this integration, what you can do, and how to prompt your AI assistant effectively.
Requirements and limitations
- The Docebo MCP server must be configured by your administrator before you can use it. Contact your administrator if the Docebo connector is not available in your AI assistant.
- The integration operates entirely within your existing Docebo permissions. The visibility of content and the actions available to you are the same as in the platform. If an action cannot be completed, you will receive an error message.
- MCP clients can perform actions in Docebo on your behalf. Review requests carefully before confirming.
Getting started
Before you can use the integration, you need to connect your AI assistant to Docebo. This is a one-time step.
Open a browser tab and log into Docebo as you normally would. Then, in your AI assistant, find the Docebo connector in the list of available enterprise connectors and click the authorization link to start the authentication flow. Once authorized, your AI assistant is linked to your Docebo account and you can start using the integration right away.
Note for SSO users: make sure you are already logged into Docebo in your browser before clicking the authorization link in your AI assistant.
Interacting with Docebo through your AI assistant
Once your AI assistant is connected to the Docebo MCP server, you can use natural language to trigger actions. The following sections outline the primary use cases and example prompts you can try.
The current version of the MCP server is for learners looking to quickly check their learning progress, find specific content, or get personalized recommendations.
Try asking:
- "What courses have I completed?"
- "Based on my completed courses, what advanced leadership courses should I take next?"
- "I am interested in data privacy. What courses do we have available?"