Edusign's MCP (Model Context Protocol) integration lets you connect your compatible AI assistants (such as Claude, ChatGPT, Cursor, etc.) directly to your Edusign data. Instead of relying solely on its general knowledge, your AI can query your Edusign platform in real time to help you view your attendance sheets, courses, groups, or students, and automate some of your administrative work.
The Edusign MCP now uses OAuth via a new dedicated URL: https://mcp.edusign.com. Each user logs in securely using their own Edusign account.
📺 Setup Video
All of the steps below are explained step by step in this short video:
💡 Note: In your Edusign account, you'll find information about the MCP connection under Settings -> API and MCP tab!
🧑🏫 Prerequisites: Before you begin, make sure you have:
Access to your Edusign account.
The permissions required to access your platform's settings (for the initial installation of the connector, which is reserved for an administrator).
An AI tool compatible with the MCP standard, such as Claude or ChatGPT.
⚙️ How do I set up the MCP connection?
The installation takes place in two steps: a one-time configuration performed by an administrator at the organizational level, followed by individual authentication that each user performs on their own.
1. Connector Installation (One-time setup)
An administrator must add Edusign as a custom connector in the settings of the AI tool used by the team (for example, in Claude's Settings ). This step needs to be performed only once for the entire organization: once the connector is installed, it becomes available to all authorized users.
To configure the connector, use the following URL:
https://mcp.edusign.com
Enter this URL in the field for adding a custom MCP server or connector, then save.
2. Authentication (to be performed by each user)
Once the administrator has installed the connector, each user must authenticate individually via OAuth using their own Edusign account the first time they use it. This personal authentication ensures that the AI only accesses the data and permissions specific to each user, and does not need to be set up by the administrator for each new user: all a user needs to do is log in once from their own account.
3. Test your assistant
Once you're logged in, start a new conversation in your AI and try a simple query, such as: "List my last three classes on Edusign" or "Show me this month's excused absences". If the response is accurate and matches your data, your connection is up and running!
🎯 Use Cases: What Should You Ask Your AI?
Once the MCP integration is active, this tool proves to be particularly useful on a day-to-day basis for administrators, academic leaders, and instructors. Here are a few examples of prompts you can use:
"Give me the list of classes for the week for Group X."
"Summarize a student's attendance and absences for the past month."
"Which attendance sheets haven't been signed yet?"
"Prepare a summary of excused absences for approval."
🔒 Security and Best Practices
To ensure the security of your platform, follow these guidelines when using the MCP integration:
Each user logs in with their own account: never share your login credentials with a coworker, even if the connector has already been installed by an administrator.
Use only trusted AI clients that are officially MCP-compatible.
If you have any doubts about the security of your login credentials, log out of the connector and log back in through your Edusign account.
🆘 Is the connection not working?
If your AI is unable to connect, check the following:
Was the connector properly installed beforehand by an administrator using the URL
https://mcp.edusign.com?Did you log in individually using your own Edusign credentials?
Does your AI tool (and its current version) fully support the MCP feature?
The quickest solution: Log out of the connector, then log back in using your Edusign account.
To learn more, visit our help center.
💬 If you can't find what you need, contact us via chat. 💬