Uploaded image for project: 'Moodle'
  1. Moodle
  2. MDL-82731

AI: Provider Plugin - AWS Bedrock

XMLWordPrintable

    • Icon: New Feature New Feature
    • Resolution: Unresolved
    • Icon: Minor Minor
    • None
    • Future Dev
    • AI
    • main_MDL-82731
    • Hide

      Testing

      1. Login as admin.
      2. Disable any other configured and enabeld providers providers (OpenAI, Azure...)
      3. Create a new instance of the AWS Bedrock provider and enter valid aws key and secret credentials.
      4. Click "Create Instance"
      5. Enable the instance you have just created.
      6. Click on the Settings link for the instance.
      7. Click on the generate text settings link
      8. Choose the first model that isn't custom in the list.
      9. Choose US East (N. Virginia) as the region
      10. Save changes.
      11. Navigate to Site admin > AI >AI Placements and enable the text editor placement
      12. Edit your user profile.
      13. Click on the AI button in the TinyMCE text editor.
      14. Select AI Generate Text.
      15. Agree with the Policy (If any).
      16. Input some text in the Describe the text you want AI to create. Example: Write a short introduction for Moodle LMS.
      17. Press Generate Text.
      18. Verify you will see the Generated text on the right.
      19. Navigate back to the settings for configured AWS Bedrock provider
      20. Click on the Settings link for action Generate text.
      21. Choose the next model in the list..
      22. Click Save changes.
      23. Repeat step 12 to 22. For all the models
      24. Navigate back to the settings for configured AWS Bedrock provider
      25. Click on the Settings link for action Generate image.
      26. Choose the first model that isn't custom in the list.
      27. Choose US East (N. Virginia) as the region
      28. Save changes.
      29. Edit your user profile.
      30. Click on the AI button in the TinyMCE text editor.
      31. Select AI Generate image.
      32. Agree with the Policy (If any).
      33. Input some text in the Describe the image you want AI to create. Example: a mountain landscape.
      34. Press Generate image.
      35. Verify you will see the Generated image.
      36. Repeat step 24 to 35. For all the models

       

       

      Show
      Testing Login as admin. Disable any other configured and enabeld providers providers (OpenAI, Azure...) Create a new instance of the AWS Bedrock provider and enter valid aws key and secret credentials. Click "Create Instance" Enable the instance you have just created. Click on the Settings link for the instance. Click on the generate text settings link Choose the first model that isn't custom in the list. Choose US East (N. Virginia) as the region Save changes. Navigate to Site admin > AI >AI Placements and enable the text editor placement Edit your user profile. Click on the AI button in the TinyMCE text editor. Select AI Generate Text. Agree with the Policy (If any). Input some text in the Describe the text you want AI to create. Example: Write a short introduction for Moodle LMS. Press Generate Text. Verify  you will see the Generated text on the right. Navigate back to the settings for configured AWS Bedrock provider Click on the Settings link for action Generate text . Choose the next model in the list.. Click Save changes. Repeat step 12 to 22. For all the models Navigate back to the settings for configured AWS Bedrock provider Click on the Settings link for action Generate image . Choose the first model that isn't custom in the list. Choose US East (N. Virginia) as the region Save changes. Edit your user profile. Click on the AI button in the TinyMCE text editor. Select AI Generate image. Agree with the Policy (If any). Input some text in the Describe the image you want AI to create. Example: a mountain landscape. Press Generate image. Verify you will see the Generated image. Repeat step 24 to 35. For all the models    
    • Hide

      Code verified against automated checks with warnings.

      Checked MDL-82731 using repository: https://github.com/mattporritt/moodle

      Should these errors be fixed?

      Built on: Mon Apr 28 04:32:20 UTC 2025

      Show
      Code verified against automated checks with warnings. Checked MDL-82731 using repository: https://github.com/mattporritt/moodle main (0 errors / 1 warnings) [branch: main_MDL-82731 | CI Job ] overview (0/1) , phplint (0/0) , phpcs (0/0) , js (0/0) , css (0/0) , phpdoc (0/0) , commit (0/0) , savepoint (0/0) , thirdparty (0/0) , externalbackup (0/0) , grunt (0/0) , shifter (0/0) , mustache (0/0) , gherkin (0/0) , Warn: The main_ MDL-82731 branch at https://github.com/mattporritt/moodle has not been rebased recently (>20 days ago). Should these errors be fixed? Built on: Mon Apr 28 04:32:20 UTC 2025
    • Show
      Launching automatic jobs for branch main_ MDL-82731 https://ci.moodle.org/view/Testing/job/DEV.02%20-%20Developer-requested%20PHPUnit/18561/ PHPUnit (sqlsrv) https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/63593/ Behat (NonJS - boost and classic) https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/63594/ Behat (Firefox - boost) https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/63595/ Behat (Firefox - classic) https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/63596/ App tests (stable app version) Built on: Thu Mar 13 00:25:17 UTC 2025

      Create an AI provider plugin that interfaces with AWS Bedrock: https://aws.amazon.com/bedrock/

      AWS Bedrock is a bit different to the other Provider plugins. AWS Bedrock is a fully managed service from Amazon Web Services (AWS) designed to simplify the deployment and management of generative AI models at scale.
      This means you can deploy one of many AI models into an AWS environment and then communicate to the model via a consistent API. This has the advantage of being quiet flexible and able to accommodate many AI models. It also adds a layer of security and privacy, as the model (and all the associated data) in contained within the orgs AWS account and not shared across different users/orgs etc.

      This plugin will interface with the upstream AI service and will make available AI Actions to the subsystem.

            matt.porritt@moodle.com Matt Porritt
            matt.porritt@moodle.com Matt Porritt
            David Woloszyn David Woloszyn
            Votes:
            2 Vote for this issue
            Watchers:
            9 Start watching this issue

              Created:
              Updated:

                Estimated:
                Original Estimate - Not Specified
                Not Specified
                Remaining:
                Remaining Estimate - 0 minutes
                0m
                Logged:
                Time Spent - 6 hours, 24 minutes
                6h 24m

                  Error rendering 'clockify-timesheets-time-tracking-reports:timer-sidebar'. Please contact your Jira administrators.