Skip to content

Deploy via MCP

What a connected AI assistant can do with publishing and sharing your Coassemble courses, and how learners still reach the course.

Yes. An AI assistant connected to your workspace over MCP (the Model Context Protocol) can publish a course and get you a share link, alongside creating and editing the course in the first place. Learners open the course the same ways they always have: through that link, an embed, a Slack or Teams message, an email invitation, or a SCORM package in your LMS.

What you can ask a connected assistant to do

  • Publish a course once it is ready.
  • Get a share link you can send on or paste wherever you need it.
  • Organise courses into collections and folders so a programme hangs together.
  • Restructure a course, reordering screens or reworking sections before it goes out.
  • Read analytics afterwards, including course and screen level results, so you can ask what is working.

How learners reach the course

The connection sits between your assistant and your workspace, not between your assistant and your learners. Once a course is published, delivery works exactly as it does for any Coassemble course: a share link, an embed in a tool like Notion, Confluence or your intranet, a Slack or Microsoft Teams share, an email invitation, or a SCORM package uploaded to your existing LMS.

What is MCP?

MCP, the Model Context Protocol, is an open standard for connecting AI assistants and agents to external content and tools. It lets an AI tool work with your Coassemble workspace through a documented, permissioned connection rather than a one-off custom integration.

Frequently asked questions

Can learners take a course inside an AI assistant?

No. A connected assistant works on your side of the course: building it, publishing it, and reading results. Learners take the course through a share link, an embed, or your LMS.

Which assistants can do this?

Any assistant that supports connecting to a remote MCP server and signing in. MCP is an open standard rather than a private integration, so this is not a fixed list, and support depends on the assistant rather than on Coassemble.

Can my own product do this rather than a chat assistant?

Yes. If you embed Coassemble, the same tools are available to your application, with the caller identified using the same client and user identifiers your existing Embed integration already uses rather than a Coassemble login. The capability is the same, only the authentication differs.