Monitor Genie spaces usage with audit logs and alerts
This feature is in Public Preview.
Due to an outage, some dashboard audit logs might be missing for the period October 16, 2025 – November 19, 2025.
This page has sample queries that workspace admins can use to monitor activity associated with Genie spaces. All queries access the audit logs table, which is a system table that stores records for all audit events from workspaces in your region.
See Monitor account activity with system tables. For a comprehensive reference of available audit log services and events, see Audit log reference.
For information about monitoring dashboard usage, see Monitor dashboard usage.
Track Genie space interactions
The examples in this section demonstrate how to retrieve audit logs for common questions about Genie space activity.
Query for feedback
The following query returns feedback ratings submitted for the Genie spaces in your workspace from the past 30 days. The columns in the query include the space_id and the email address of the user who submitted the feedback, along with all other columns from the system table.
SELECT
user_identity.email as user_email,
action_name,
request_params.space_id,
request_params.feedback_rating,
*
FROM
system.access.audit
WHERE
service_name = 'aibiGenie'
AND action_name = 'updateConversationMessageFeedback'
AND event_date >= current_date() - interval 30 days
Return requests for review
The following query returns request for review activity from Genie spaces over the past 30 days. It includes the space_id, the email address of the user who added the comment, the type of action, and all other columns from the source table.
SELECT
user_identity.email as user_email,
action_name,
request_params.space_id,
*
FROM
system.access.audit
WHERE
service_name = 'aibiGenie'
AND action_name = 'createConversationMessageComment'
AND event_date >= current_date() - interval 30 days
API response codes
The following query returns a count of Genie API requests grouped by HTTP response status code for the past 30 days. Use it to monitor success rates, spot errors (such as 429 or 5xx), and track activity for conversation start and create message actions. Results include the date and time to the minute so you can correlate 429 rate-limit responses with request volume per minute (for example, more than 5 requests in a minute).
SELECT
date_trunc('minute', event_time) as event_minute,
response.status_code,
COUNT(*) as request_count
FROM
system.access.audit
WHERE
service_name = 'aibiGenie'
AND action_name IN ('genieStartConversationMessage', 'genieCreateConversationMessage')
AND event_date >= current_date() - interval 30 days
GROUP BY
date_trunc('minute', event_time),
response.status_code
ORDER BY
event_minute,
response.status_code
Set up alerts
You can set alerts to automate this type of monitoring. See Create an alert to learn how to set an alert on a specific threshold.