Evidence-first security operations / interactive learning
A five-minute, interactive walkthrough of two harmless lab findings: controlled investigation, explicit human approval, bounded fixes, independent retests and an auditable outcome.
Select an illustrative saved-state scenario. All results below are examples, not live receipts.
Both synthetic findings have been investigated, but neither has approved remediation and independent retest proof.
A pinned package advisory is shown separately from a harmless policy-mode fixture. No JWT exploitation occurs.
A fixed local, non-secret sentinel at /lab/debug demonstrates an independent second finding.
Select each beat to see what to say, what it proves and what it does not prove.
Security operations must connect findings to explicit decisions and trustworthy evidence—not simply claim that a tool ran.
Two harmless simulations make a controlled workflow understandable.
No live vulnerability exploit, Internet scan or live AI model.
The lab's real controls are separate from this public educational page.
Approving policy does not authorize debug. A Reject decision never authorizes a fix. The public scenario buttons are not approvals.
Same-fixture validation, bounded execution and independent retest must agree. Past-run evidence cannot close a new run.
Paused, blocked, stale or unknown outcomes stay unproven. No automatic replay or assumption of success.
Source code, the full five-minute rehearsal and sanitized test evidence are available in GitHub. The actual Docker UI remains local-only.