[Meta] Explosion of AI Agent Tests. (Easier to build, export, import and run tests]

[Tracker]
Update Summary: [One-line status update for stakeholders]
Short Title: Explosion of AI Agent Tests
Short Description: Easier to build, export, import and run tests
Check-in Date: MM/DD/YYYY
Due Date: MM/DD/YYYY
Blocked by: [#XXXXXX] (New issues on new lines)
Additional Collaborators: @username1, @username2
Metadata is used by the AI Tracker. Docs and additional fields here.
[/Tracker]

Problem/Motivation

We have some tests written for Drupal CMS AI Agents. However, whilst it is possible to create tests, its not easy. We want to be able to make an explosion of tests so many people can create, submit and share those tests.

Then we can find ways of running tests more easy and we start working on many different metrics for reporting those tests.

This will eventually allow for the training data needed for:

#3561040: [Meta] Fine-tune small Open Source Drupal specific LLMs

Steps to reproduce

Proposed resolution

Tasks to get us there.

• #3541336: Create Tests from a log of an AI Assistant Chat History.
• #3541338: Create a Central Store of tests such as on github.
• #3560677: Add metadata to test and test group exports
• #3555753: Make it possible to export groups to recipes
• #3541323: Store the Test completion time for Tests and Test Groups in results.
• #3537161: Write a suite for Canvas AI and check any issues
• #3541324: Run more than one test group in Bulk - Test Collections?
• #3541333: Allow for Test Groups to contain Recipies with content and config.
• #3541329: Run Multiple + All result Averages
• #3541319: Create workflow for AI Agent testing to be run on an external website.

Remaining tasks

User interface changes

API changes

Data model changes