Skip to content

LTI Connection (Canvas)

This guide connects OnMicro.AI to Canvas as an LTI 1.3 tool. Once a Canvas administrator approves OnMicro.AI for an account, instructors can insert any of their OnMicro apps directly from within Canvas (the Rich Content Editor, an assignment, or a module link). When a student finishes the app, a score is passed back to the Canvas gradebook.

Note: The grade is pass/fail based on whether or not the student successfully reached the end of the exercise.

  1. A Canvas admin registers OnMicro.AI once as a Developer Key (LTI tool) for the account.
  2. An instructor inserts an OnMicro app using Canvas deep linking and picks the app from a list of their OnMicro apps.
  3. A student opens the app, completes it, and a score is sent back to Canvas via LTI Advantage (AGS).

Instructor requirement: the picker matches you by email. You must have an OnMicro.AI account using the same email address as your Canvas account, and the app you want to insert must belong to that account.

# IN CANVAS (administrator)

  1. Go to your Canvas account admin → Apps.

  2. Choose ‘Install a New App’.

  3. Select LTI Version ‘1.3’ and set the install method to “Enter URL” and paste the OnMicro.AI configuration URL:

    https://onmicro.ai/lti/canvas-config/

    This URL returns the tool configuration (launch, login, and JWKS endpoints, the grade-passback scopes, and the deep-linking placements). Canvas will fill in the rest of the form for you.

  4. The app form should be populated with the correct OnMicro.AI URLs and values. Select ‘Next’ to proceed.

  5. Leave the default permissions. Select ‘Next’ to proceed.

  6. For User Data Shared With this App, select ‘All User data’ to share names and emails with Onmcro.AI

  7. Leave the default Placements selected. Select ‘Next’ to proceed.

  8. Leave the default Override URI values. Select ‘Next’ to proceed.

  9. Give your app a nickname (e.g. OnMicro) and select ‘Next’ to proceed.

  10. Review your changes and select ‘Install App’ to complete the connection.

  11. Choose the Modify Availability button (the pencil icon) and switch the app to ‘Available’

  12. Note the generated values. From the app’s settings record:

FieldWhere to find it
Client IDThe ‘Copy Client ID’ button
Deployment IDThe Deployment ID is listed on the screen

# IN ONMICRO.AI (administrator, via Django admin)

Tool consumers are configured from the Django admin panel.

  1. Go to the Django admin → LTI → LTI configs → Add LTI config.

  2. Leave “Microapp” blank. This makes the registration account-level, so instructors choose an app via deep linking.

  3. Fill in the platform credentials from Canvas:

    FieldValue
    IssuerYour canvas URL with https prefix (e.g. https://mysite.instructure.com)
    Client IDThe Developer Key Client ID from Canvas
    Auth Login URLYour instance’s value (e.g. https://mysite.instructure.com/**api/lti/authorize_redirect**)
    Auth Token URLYour instance’s value (e.g. https://mysite.instructure.com/**login/oauth2/token**)
    Keyset URLYour instance’s value (e.g. https://mysite.instructure.com/**api/lti/security/jwks**)
    Deployment IDThe Deployment ID from Canvas

    Tip: the exact Issuer / Auth / Keyset URLs depend on your Canvas environment. They are shown in Canvas’s LTI documentation for your region and match the values your Canvas instance uses.

  4. Save the configuration.

# IN CANVAS (instructor)

  1. Open a course where the OnMicro.AI app is available.

  2. Insert the app using deep linking. For example, in the Rich Content Editor open the apps/plugins menu and choose OnMicro.AI, or add an assignment with Submission Type → External Tool → Find and select OnMicro.AI.

  3. Pick your app from the list shown by OnMicro.AI and confirm. Canvas inserts a link to the app (and creates a gradebook column when added as an assignment).

  4. Publish the content as usual.

# AS A STUDENT

  1. Open the OnMicro.AI app from the course.

  2. Complete the exercise and reach the end.

  3. Check the gradebook. A score is passed back to Canvas automatically when the app is completed.