Before integrating the Guardrails API, use the online testing feature to validate detection behavior against your content. Submit text, select a detection model, and inspect results — all from the console, without writing code.
Prerequisites
Before you begin, ensure that you have:
Activated Guardrails on the Guardrails service activation page
Running tests incurs charges. For details, see Billing overview.
Run a test
Log on to the Guardrails console.
In the left-side navigation pane, click Online testing.
Select the Text template or Text & image template tab based on the type of content you want to test.
In the input box, enter the content to test. The maximum length is 2,000 characters. You can also click a preset template to populate the input field.
Below the input box, select the detection model:
Input Detection Model: Select AI Input Content Moderation_Pro (query_guard_pro_ec) to detect content submitted to the large model.
Output Detection Model: Select AI Output Content Moderation_Pro (response_guard_pro_ec) to detect content generated by the large model.
Click Run test.
Review the Test Results panel.
Enable detection features
To enable or adjust detection features, go to Protection Configuration:
In the left-side navigation pane, choose .
In the configuration list, locate the target Service and click Management.
In the Protection Dimension area, use the toggles to enable or disable features such as Sensitive content detection and Prompt injection detection. Some features incur separate charges. For details, see Billing overview.
Click Configuration Management on a Protection Dimension card to further configure detection rule toggles.