Claude, which learners signed up in the last 30 days?

We are excited to announce our new Model Context Protocol (MCP) server that makes it easier to manage your Edbound account. Available today through Claude, Cursor, Windsurf, and VS Code, you can now query and act on your learner data, course structures, and live sessions through plain conversation.

Through a simple chat, look up users, track course progress across cohorts, manage enrollments, generate password reset and auto-login URLs, and pull live session data, all without opening the Edbound dashboard.

Note: Edbound MCP is generated from inside your account at Settings → Integrations → MCP. Paste the endpoint URL into your AI tool and you are live. See below for setup instructions per tool.

The Edbound MCP server is built for content-driven businesses especially those looking to speed up daily ops, reduce dashboard hopping and put their AI assistant to work on real account data.

This is just the beginning. We are shipping deeper coverage soon across our Webinar, Podcast, Event, Blog, Case Study, and Community hubs, plus GTM-side workflows that make growing your business easier. The vision is simple. Every aspect of a content hub on Edbound becomes addressable through your AI assistant of choice.

What is MCP?

MCP stands for Model Context Protocol. It is an open standard built by Anthropic that lets AI assistants like Claude talk to external tools and data through one shared language.

Before MCP, every AI tool needed a custom integration with every app. Now any MCP-compatible AI assistant can connect to any MCP server and use the tools that server exposes. Same standard, same connection pattern, no glue code.

What is Edbound MCP?

Edbound MCP is our hosted server that gives AI tools secure, scoped access to your Edbound account. Once you connect it, your AI assistant can read across your users, courses, bundles, and live sessions, and take actions like enrollments and password resets, all through chat.

You generate the key from inside Edbound. You decide which tools each key can use. You can spin up a separate key for every AI tool you connect, so your Cursor key and your Claude key carry different permissions.

For coaches running their academy on Edbound, this means your daily ops happen inside the chat window. For enterprise teams running corporate training, channel enablement, or partner education hubs, this means your learner success team has a 24/7 AI ops layer sitting on top of your hub.

How it works: an example flow with Claude

Once Edbound MCP is toggled on inside your Claude chat, the entire workflow runs through simple conversational requests.

You ask: "Help me analyze the content completion rates of my users." Claude calls GetUsers and pulls the list with profile details.

You follow up: "Which courses are these learners enrolled in?" Claude runs GetCourseUsers for the course names and returns a filtered list of course names.

You dig deeper: "Which modules are they getting stuck on?" Claude calls GetUserCourseProgress for each learner, cross-references against GetCourseSections to map progress to modules, and returns a breakdown of where the drop-offs are happening.

You wrap up: "Could you create a comprehensive report for me" Claude calls GetCourseLiveSessions and returns the upcoming sessions list.

That whole flow used to take forty-five minutes of dashboard hopping. It now takes ninety seconds.

The same pattern works whether you are a coach managing a 50-learner cohort or an enterprise L&D team managing 50,000.

Available on Claude, Cursor, Windsurf, and VS Code

Edbound MCP follows the open MCP standard, so it works wherever MCP works. We have native setup guides for the four most common AI tools our customers use:

1. Claude: Add Edbound as a custom connector in Settings, toggle it on inside any chat
2. Cursor: Drop the JSON config into your MCP Tools settings, switch your chat to Agent mode
3. Windsurf: Paste the config into mcp_config.json, the server shows up inside Cascade
4. VS Code: Add the server through the Command Palette, run it through GitHub Copilot in Agent mode

Step-by-step setup for each tool is in our Connecting to Edbound MCP guide.

If your AI tool supports MCP but is not on this list, you can still connect using the generic setup. The endpoint URL and the JSON config work with any MCP-compatible client.

Built with security at the centre

Your MCP key is scoped, rotatable, and granular by design.

1. Per-key tool permissions. Every key you generate has its own toggle list. Turn off RemoveUser, RemoveCourse, and the other destructive tools by default. Turn them on only for keys used in audited workflows.
2. One key per AI tool. Use a separate MCP key for Claude, Cursor, and any other client. Rotate or revoke without touching the others.
3. Treat it like a password. Never share it in public channels, screenshots, or version control. If a key is exposed, delete it from Settings → Integrations → MCP and generate a new one.
4. Human confirmation on. Most AI tools let you approve each step before it runs. Keep this on, especially for tools that can permanently delete data.

Full security guidance is in our MCP security best practices article.

Get started

If you are already on Edbound, head to Settings → Integrations → MCP and generate your first key. Pick your AI tool, paste in the endpoint URL, and you are live in under five minutes.

If you are not yet on Edbound and you run a content-driven business, whether you are a coach scaling your academy or an enterprise launching B2B content hubs, come take a look. We have helped over 1.5 million users across 27+ countries build, run, and scale their content engines. The MCP layer is the newest reason to join.

Frequently Asked Questions

Who can use Edbound MCP?
Any Edbound customer on the Power Plan or above. If you are on a lower-tier plan and want access, get in touch and we will help you upgrade.

Which AI tools does Edbound MCP work with?
Claude, Cursor, Windsurf, and VS Code have native setup guides today. Edbound MCP follows the open MCP standard, so it works with any MCP-compatible client. The full setup guide is here.

Is it safe to give an AI assistant access to my Edbound account?
Yes, when you set it up correctly. Each MCP key has its own tool permissions, so you can disable destructive tools by default, use a separate key for each AI tool, and rotate keys whenever you want. Read our full security best practices before connecting.

Can I create more than one MCP key?
Yes. We recommend a separate key for each AI tool or use case, with only the tools each one actually needs. This makes rotation and access control much cleaner.

What happens if I lose my MCP key?
We do not store your endpoint URL in a recoverable form once it is generated. If you lose the key, delete the existing entry in Settings → Integrations → MCP and click Create to generate a new one. Update your AI tool's config with the new endpoint.

Can I control which tools an AI agent can access?
Yes. Click the settings gear next to any MCP key, choose Tools, and toggle each capability on or off. We recommend enabling only what that key needs and keeping destructive tools off unless your workflow specifically needs them.

What can I actually do with Edbound MCP today?Manage learners, run reports, update account data, and automate operational work across users, courses, bundles, and live sessions. The full tool list with example prompts is in Supported MCP Tools.