The Test Results section of the Guardrails console provides two views for analyzing your moderation traffic: Result Query for inspecting individual records and Risk Report for tracking violation trends across detection types.
Query moderation results
Use Result Query to search and inspect the moderated text your service has processed.
Log on to the the Guardrails console.
Go to Test Results > Result Query. The page lists each moderated text entry with its hit tags and request time.
Filter the list using any combination of the following criteria:
The page retains up to 50,000 records from the last 30 days. For longer retention or higher volume, save the results returned from API calls.
Filter Description Time range Narrow results to a specific date range Request ID Look up a specific API call Text Search by moderated text content Tags Filter by hit tag To view full moderation details for an entry, find the text in the list and click Details in the Operation column.
View risk reports
Use Risk Report to monitor violation trends and distribution across your detection traffic.
Go to Test Results > Risk Report.
Select the Content security detection, Sensitive content detection, or Prompt injection detection tab.
View Trend Statistics and Risk Distribution for the selected detection type. Break down the data by day, month, account, or service. Risk Distribution shows the top five hit tags and all hit tags.