Agent-assisted follow-up

Agent-assisted follow-up is available only when enabled for your workspace. Start it in Workbench from an authorized project; this is not a CLI scan workflow.

#Start and review work

1. Open the authorized project in Workbench.
2. Open Agent assessments and use the workflow available to your workspace.
3. Review each resulting finding in Findings register. Inspect attached HTTP Records or OAST callbacks before deciding whether the result is actionable.

Agent-assisted work stays within the project and its authorization boundary. It does not replace scope review or operator triage. A promoted finding still needs evidence that a reviewer can inspect.

For the standard assessment path, see Projects and scans. For evidence review, see Findings and evidence.