Agent-assisted follow-up

Agent-assisted work, when enabled for your workspace, runs inside Workbench on an authorized project. It is not an interactive TUI, not eresus-guard olium chat, and not a CLI scanner mode.

What it is allowed to do

  • Stay inside the same project, identities, and authorization gate as native modules
  • Write findings into the same evidence trail
  • Attach HTTP Records and OAST callbacks the same way other assessments do

What it is not

  • Not a pentest product
  • Not a restore product
  • Not an unbounded “hack everything” agent
  • Not a replacement for operator review

Every promoted finding still needs reviewed evidence.

How to start

Open Workbench, select the project, and use the approved agent workflow from that workspace. Do not start agent work from the CLI.

See Projects and scans and Scope and safety.

LoginRequest access

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