Skip to main content

Vulnerability Management process

Process

  1. Discover: accept findings from code review, dependency review, provider advisories, VAPT, authenticated testing, monitoring, incidents, researchers, and supplier notices.
  2. Validate: confirm the affected asset, version, reachability, reproducibility, evidence, and whether the finding is a duplicate or false positive.
  3. Prioritise: consider severity, exploitability, exposure, data sensitivity, tenant isolation, known exploitation, compensating controls, and business impact.
  4. Assign: record an accountable technical owner, risk owner, target date, and interim controls.
  5. Remediate: patch, reconfigure, restrict, redesign, remove, or accept the risk through a dated exception.
  6. Verify: retest the exact attack path and relevant regression behavior; a code change alone is not closure.
  7. Close and learn: retain evidence, update related controls, and report trends and overdue risk.

Target treatment times

These are initial operating targets, measured from validated finding to verified remediation.

PriorityTargetEscalation
Critical72 hoursImmediate management escalation and containment; shorter where actively exploited
High14 daysWeekly risk-owner review until closure
Medium30 daysEscalate when due date is at risk
Low90 daysAddress through planned maintenance or accepted risk

Contractual or regulatory deadlines override these targets when stricter.

Exceptions

Risk acceptance requires the finding, impact, reason, compensating controls, approver, expiry, and review date. Critical actively exploited exposure cannot be accepted without immediate containment and accountable management decision.

Vulnerability record schema

FieldPurpose
Finding identityStable reference used across scanner, review, remediation, deployment, exception, and retest evidence
Affected asset and versionExact application, dependency, provider configuration, data domain, endpoint class, or release in scope
Discovery source and dateSAST, dependency scan, VAPT, provider advisor, incident, review, or researcher provenance
Technical conditionReproducible preconditions, trust boundary, vulnerable behavior, and evidence without publishing exploit-enabling secrets
Business impactConfidentiality, integrity, availability, tenant-isolation, privacy, financial, and operational consequences
PriorityBase severity adjusted for reachability, exploitation, data sensitivity, known exploitation, and compensating controls
OwnershipTechnical remediation owner, risk owner, due date, and escalation path
TreatmentPatch, configuration, restriction, redesign, removal, mitigation, or time-bounded risk acceptance
Release linkageCorrected revision, migration, provider change, deployment, and rollback dependency
RetestTester, date, exact attack path, related regression scope, result, and residual risk
ClosureRisk-owner approval, evidence reference, closed date, and any follow-up monitoring

Prioritisation rules

Known exploitation, exposed authentication bypass, cross-tenant access, privileged credential exposure, unauthorised financial action, or broad sensitive-data impact triggers immediate containment regardless of the nominal scanner score. A finding is not downgraded only because the current UI hides the vulnerable operation. Compensating controls reduce residual risk only when they are verified on the exact attack path and have an accountable owner and monitoring signal.

Current-state limitation

The process now exists, but historical closure evidence, automated SAST/DAST/dependency scanning, independent retest evidence, a consolidated vulnerability register, and demonstrated adherence to these targets were not available. Accordingly, checklist items requesting past reports remain unavailable rather than being backfilled with unperformed tests.