Hook Execution Logs
View logs for each Hook interception or modification, control whether original message content is recorded, and trace Hook execution details for individual messages from conversations.
Last updated
Was this helpful?
View logs for each Hook interception or modification, control whether original message content is recorded, and trace Hook execution details for individual messages from conversations.
Every Hook interception, modification, and pass-through is logged. For regulated industries (finance, healthcare, government), "traceability" is itself a compliance requirement — when issues arise, you need to trace "which message, triggered by which rule, with what action, at what time."
Navigate to AgentOps → Hook Execution Logs from the left menu. You can see the following for each execution entry:
Time — When it occurred
Assistant — Which AI assistant triggered it
Matched Rule — Which Hook handled it
Action — What was done (mask/rewrite, block, or pass-through)
Status — Processing result
You can filter by date range, assistant, and keywords, and export reports for further audit analysis. The top of the page also displays statistics for total executions, matches, and match rate.
Click on any log entry to see the complete context of that action: basic information (Hook name, type, trigger point, action, result), related information, and message content, giving you a clear picture of what the system did.
By default, execution logs only record the "action result" without retaining the original sensitive content — this prevents PII or sensitive messages from being stored again in the logs.
If you need to see "what the original content looked like" during debugging, go to the Hook settings page and change the Record original values before masking toggle from off to on. When enabled, logs will additionally retain the original content before masking/blocking, making it easier to investigate issues.
Privacy trade-off: Enabling "Record original values" means the original PII/message content will be stored in execution logs. For security reasons, it is recommended to keep this off by default and only enable it temporarily when troubleshooting is needed.
In addition to centralized execution logs, you can also trace back from actual conversations: Navigate to Customer Service → All Conversations, open any conversation, and you can see which messages in that conversation were processed by Hooks (masked or blocked), along with the corresponding actions taken.
This allows you to understand "why this message was masked/blocked" directly while reviewing customer conversations, without having to cross-reference logs separately.
Hook Message Interception Overview — What Hooks are and the five standard templates
Create and Configure Hooks — How to create, configure parameters, and link to AI assistants
Last updated
Was this helpful?
Was this helpful?