Terms of Service

Last updated: March 2026

1. Overview

Canvas MCP is a personal-use integration that connects your Canvas LMS account to AI assistants via the Model Context Protocol (MCP). By using this service, you agree to these terms.

This MCP is not affiliated with Canvas or Instructure. It is an independent project that strives to operate within Instructure's and Canvas's acceptable use and API policies. We do not speak for or represent Instructure or your institution.

2. Acceptable Use

You agree to:

  • Only connect Canvas accounts you are authorised to access.
  • Not use this service to scrape, bulk-export, or redistribute Canvas data.
  • Comply with your institution's Canvas acceptable use policy.
  • Keep your Personal Access Token confidential.

3. Credential Storage

Your Canvas Personal Access Token (PAT) is encrypted using AES-256-GCM and stored in a serverless Redis database. It is only decrypted server-side to make API calls on your behalf and is never transmitted to the AI assistant. You can revoke access at any time by deleting your PAT in Canvas (Account → Settings → Access Tokens).

4. No Warranty

This service is provided "as is" without warranties of any kind. It is not affiliated with or endorsed by Instructure (Canvas), Anthropic, or any educational institution. We aim to stay within Canvas/Instructure's policies; compliance with your institution's rules remains your responsibility. Use at your own risk.

5. Limitation of Liability

To the maximum extent permitted by law, the operators of Canvas MCP are not liable for any loss, damage, or unauthorised access resulting from your use of this service.

6. Changes

These terms may be updated at any time. Continued use of the service after changes constitutes acceptance.