Skip to content

Pentest Engine

The Pentest Engine runs an end-to-end penetration test against authorized targets — from reconnaissance through exploitation to a finished report. It combines automated tooling with an AI exploitation layer and a library of exploit validators that confirm whether a vulnerability is truly exploitable.

BenefitCapabilityBusiness value
Full pipelineOSINT → recon → DAST → exploitation → reportOne workflow instead of stitching tools together
Confirmed riskExploit validatorsSeparates real, exploitable issues from noise
ScaleAI-assisted payloads and orchestrationCover more ground than manual testing alone

A pentest moves through stages: OSINT (open-source intelligence on the target), recon (surface and service discovery), DAST (dynamic scanning of running web apps), exploitation (validating findings, with an AI layer that generates and adapts payloads), and reporting.

Rather than reporting a “possible” vulnerability, the engine includes validators for common classes — SQL injection, XSS, SSRF, RCE, LFI, XXE, SSTI, IDOR, CORS, JWT issues, insecure deserialization, file upload, GraphQL, and more — that attempt safe confirmation so you act on proven issues.

The engine orchestrates a large library of integrated tools (network scanning, web testing, and more), gated by a per-phase permission matrix so only approved tools run at each stage. Dedicated modules cover web pentesting, Active Directory attacks, and cloud-specific techniques.

  1. Define authorized targets (URLs / hosts) and confirm scope.
  2. Launch the pentest from the Security Test Hub or the New Scan Web PT option.
  3. Monitor progress through the pipeline stages.
  4. Review validated findings and export the pentest report.
  • Only test assets you’re authorized to test.
  • Start with recon/DAST, then enable exploitation on confirmed targets.
  • Use exploit-validated findings to drive the most credible remediation conversations.