Detection & correlation
Normalize events, apply versioned detection rules, correlate signals and calculate risk.
SECURITY OPERATIONS PLATFORM · V17
Turn security telemetry into actionable defense.
SentraX brings detection, correlation, UEBA, threat intelligence, incident workflows and evidence integrity into one security operations platform.
Hardened prototype / production-candidate architecture. Independent security validation is still required before production certification.
SECURITY STATUS
Defense plane online
RBACTenant isolation
UEBABehavior baseline
AIHuman approval
AUDITEvidence chain
ONE OPERATING PLANE
Built around the SOC workflow
From raw events to an auditable incident lifecycle, SentraX separates detection, analysis and action while keeping security boundaries explicit.
UEBA
Transparent historical baselines and anomaly scoring help surface behavior that deviates from normal patterns.
Threat intelligence
Bring threat context into the security event plane and connect indicators to investigations.
Incident response
Use incidents, playbooks and SOAR-style workflows with human approval before consequential actions.
Evidence integrity
AES-GCM protection, audit-chain controls and a KMS/HSM integration boundary support protected evidence.
AI SOC Analyst
AI operates on explicitly untrusted telemetry, with output controls, secret filtering and advisory-only behavior.
SECURITY BY DESIGN
Defenses are layered, not assumed.
SentraX documents security boundaries and validation limits instead of making unsupported “100% secure” claims.
Identity & access
Authentication, RBAC, active-user checks, session invalidation and tenant isolation.
API hardening
Rate limiting, received-body size enforcement, input limits and secure response headers.
AI isolation
Telemetry is treated as untrusted data; instructions embedded in telemetry are not treated as policy.
Data protection
Authenticated encryption and an adapter boundary for KMS/HSM-backed key management.
ARCHITECTURE
A clear security plane from ingest to evidence
Designed so each stage has a defined responsibility and trust boundary.
Users / Integrators
→Auth · RBAC · Tenant Isolation
→Ingest · Normalize · Detect · Correlate · Risk
→UEBA · TI · ATT&CK · Attack Graph
→Incidents · Playbooks · Human Approval
→AES-GCM · Audit · Evidence
AI SOC AnalystUntrusted telemetry boundaryOutput allow-listSecret filteringHuman-in-the-loop
VALIDATION
What is verified — and what is not
Verified in the prototype
- Python compile checks
- Automated test suite: 62 tests passed
- Frontend runtime smoke checks
- CSP and security response headers
- No frontend innerHTML / localStorage / sessionStorage usage
Still required before production
- Docker build/runtime in CI
- PostgreSQL/Redis integration & concurrency tests
- AWS Marketplace sandbox validation
- Load, failover and disaster-recovery tests
- Independent penetration test and source review
SENTRAX
See the platform in action.
Request a technical demonstration and review the architecture, security controls and deployment path.