Skip to content

Whitebox Agent

Whitebox pentesting is a Cascade capability. When you attach repository source code to an assessment, Cascade reasons over that code alongside the running application, so findings can point to the exact file and function behind a vulnerability.

What It Does

  • Consumes uploaded source: Uses repository archives (and related code artifacts) as whitebox context
  • Pre-seeds recon from the repo: Maps architecture, auth model, entry points, and high-risk sinks before deep exploitation
  • Combines static triage with dynamic validation: Uses source-aware findings to prioritize payloads and endpoints, then proves issues against the live target
  • Keeps evidence dynamic: Static hits stay hypotheses until Cascade validates them on the running application

How It Is Used

In the New Pentest form, open Fine-Tune (Optional), then Artifacts. Upload your source code (typically as an archive) along with any other useful context.

Uploaded artifacts are attached to the pentest workflow and surfaced to the scanner. When source is present, Cascade switches into its source-aware whitebox workflow automatically. You cannot enable or disable whitebox as a separate toggle: it runs when source code is attached.