To improve performance of AI content-rating, we should use a KV binding to store R2 asset ratings. This decision was made based off the discussion in a previous PR: #207 (comment).
To improve rating "accuracy", should also see if we can send data in metrics.json or the test title and description to the test-rating as well.
To improve performance of AI content-rating, we should use a KV binding to store R2 asset ratings. This decision was made based off the discussion in a previous PR: #207 (comment).
To improve rating "accuracy", should also see if we can send data in
metrics.jsonor the test title and description to the test-rating as well.