Agent-assisted follow-up

Agent-assisted follow-up is a bounded Workbench workflow for an authorized project. It is not an unrestricted chat interface, a general-purpose intrusion agent, or a replacement for operator review.

Entry boundary

Start from the project that already contains the approved target or source, identities, and authorization context. Workspace entitlement controls whether agent-assisted work is available.

What the agent can contribute

The agent can map context, propose a focused assessment plan, and help prioritize follow-up. Concrete modules still execute checks. A promoted result remains in the same project evidence trail as native assessments.

What it does not claim

  • It does not expand the project authorization boundary.
  • It does not replace a pentest engagement.
  • It does not turn an explanation into a finding without inspectable evidence.
  • It does not remove the need for an operator to review a promoted result.

Evidence and review

Review the concrete assessment evidence attached to a promoted finding before remediation, export, or disclosure. HTTP records and OAST callbacks remain relevant when the agent-assisted work produced them; source locations remain relevant for source analysis.

Availability

Agent-assisted DAST is available from the Business package. See Plans and Agent-assisted follow-up for the operating model.

LoginRequest access

Need a workflow that is not documented here? Email contact@eresussec.com.