Skip to content

Control Effectiveness

Every SiderealProbeResult carries a controlEffectiveness field that normalizes the 12 possible probe outcomes into four values. This is the primary field for dashboards, reports, alerting, and incident creation.

Effective — The security control is working as intended. The probe verified that the enforcement layer is active and correctly denying or detecting the test action.

Ineffective — The security control failed. The probe attempted an action that should have been blocked, denied, or detected, and the control did not respond. This represents a real gap in your security posture.

Degraded — The probe could not fully evaluate the control. The enforcement layer may be partially working, the backend may be unreachable, or the result was inconclusive. Investigation is recommended.

Compromised — The integrity of the probe result itself is in question. This occurs when HMAC verification fails, indicating the result may have been tampered with. A SystemAlert is created and the affected probe surface is suspended.

Probe OutcomeControl Effectiveness
PassEffective
DetectedEffective
BlockedEffective
RejectedEffective
FailIneffective
UndetectedIneffective
AcceptedIneffective
NotEnforcedIneffective
NotApplicableDegraded
BackendUnreachableDegraded
IndeterminateDegraded
TamperedResultCompromised

The controlEffectiveness field is:

  • Set on every SiderealProbeResult by the result reconciler
  • Available as a label (sidereal.cloud/control-effectiveness) for filtering
  • Used by the incident reconciler to determine whether to create incidents (enforce mode)
  • Used by the report generator for continuous monitoring summaries and POA&M entries
  • Exposed as a Prometheus metric label for dashboards