Anomaly Action

The Anomaly action manages anomalies that have gone inactive for a period you set. Archive moves them out of the active queue but keeps the record, and Delete removes them permanently. They share the inactivity rule described below, but not the same pool of candidates.

Why Use an Anomaly Action

Use an Anomaly action when findings nobody is acting on should leave the active queue on their own, instead of accumulating until someone cleans them up by hand. Common cases:

Keep the queue meaningful: A queue holding only findings under triage keeps the team's attention; stale entries bury the new ones.
Enforce a retention habit: Archive anomalies untouched for a quarter, on a monthly schedule, so cleanup happens on a cadence instead of on memory.
Clear known noise: Delete short-lived findings from datastores where the record has no audit value, keeping Archive for everything worth revisiting.
Make automation auditable: Assign a tag and a comment on the Archive action, so anyone reviewing an archived anomaly sees it was a cleanup run and why.

Evaluation

When the parent Flow runs, the action looks at the anomalies in scope. Any anomaly that has gone the full Inactivity Period untouched, meaning no updates and no comments during that whole period, qualifies. The period is counted in days, and the action offers presets of a week, a month, a quarter (the default), or a year:

An Archive action archives it with the reason you chose. It only considers anomalies that are still open, so anything already archived is left alone.
A Delete action removes it permanently. It considers every anomaly, including the ones a previous Archive run moved out, so a Delete also purges archived findings that went quiet.

The action runs whenever the Flow's trigger activates. In most setups that means on a schedule defined on the Flow's trigger, so the cleanup repeats automatically at the cadence you set.

Filter Conditions

Filter conditions narrow the list of candidates by source datastore, anomaly status, or anomaly tag. All filters must match (AND logic). An anomaly is affected only if it meets every filter you fill in.

Filters add on top of the Inactivity Period, not in place of it. An anomaly that matches every filter is still left alone until it has been inactive for the full period.

What a Delete Removes

Running a Delete action permanently removes matching anomalies and has the following effects:

Effect	Description
The anomaly record	The anomaly is permanently removed from the platform.
Comments	All comments on the anomaly are also removed.
Tag associations	The anomaly's tag associations are removed. The tags themselves remain available for other anomalies.
Assignees	Assignment links between the anomaly and users are removed.
Linked tickets	Links between the anomaly and external tickets (Jira, ServiceNow) are removed. The tickets in those external systems are not affected.
Quality check and datastore counts	Active anomaly counts on the related quality checks and datastores update automatically.

Warning

This action cannot be undone. Anomalies removed by a Delete action cannot be restored. If you might need the record later, use Archive instead.