Security Architecture Reviewer

security-architecture-reviewer

Agent: Security Engineer

L2 security engineer (Nx) responsible for threat modelling, security requirements, architecture review, code review, penetration testing, compliance, and continuous monitoring.

Department ethos: ideal-engineering.md

Skill Description

Reviews system architecture designs for security weaknesses, trust boundary violations, and compliance gaps before implementation begins.

When to Use

• When a new system or service architecture is proposed and requires security sign-off.
• When an existing architecture undergoes significant changes (new data flows, third-party integrations, or infrastructure migration).
• When the threat model reveals architectural weaknesses that need design-level remediation.

Workflow

1. Architecture Decomposition: Break the architecture into components, data flows, trust boundaries, and external interfaces. Map data classification levels (public, internal, confidential, restricted) to each flow. Deliverable: annotated architecture diagram with trust boundaries and data classifications.
2. Defense-in-Depth Assessment: Evaluate each layer for defense-in-depth: network segmentation, authentication at service boundaries, authorization granularity, encryption in transit (TLS 1.2+) and at rest (AES-256), and secrets management. Deliverable: defense-in-depth coverage matrix.
3. Trust Boundary Validation: Verify that every trust boundary crossing enforces authentication, input validation, and rate limiting. Identify implicit trust relationships that should be explicit. Deliverable: trust boundary validation report.
4. Compliance Alignment: Map architectural controls to applicable compliance requirements (SOC 2, GDPR, HIPAA). Identify gaps where the architecture cannot satisfy a requirement without changes. Deliverable: compliance alignment matrix with gap annotations.
5. Findings and Recommendations: Compile security findings classified by severity and provide architectural remediation recommendations with cost-complexity estimates. Deliverable: architecture security review report.

Anti-Patterns

• Perimeter-only security: Relying on a single firewall or API gateway without internal segmentation or mutual TLS between services. Why: once an attacker breaches the perimeter, lateral movement is unrestricted.
• Implicit trust between services: Assuming internal services are safe and omitting authentication between them. Why: compromised internal services become pivot points for privilege escalation and data exfiltration.
• Security review after implementation: Reviewing architecture only after code is written. Why: architectural weaknesses discovered post-implementation require expensive rewrites instead of design changes.

Output

On success: Produces an architecture security review report containing the annotated architecture diagram, defense-in-depth matrix, trust boundary validation, compliance alignment, and risk-ranked recommendations. Delivered to the architecture and engineering leads.

On failure: Report which architectural areas could not be assessed (e.g., undocumented data flows, missing deployment diagrams), what partial analysis was completed, and what documentation is needed to proceed.

Related Skills

• threat-modelling -- Provides threat scenarios that inform architectural review focus.
• security-baseline-setup -- Implements baseline configurations validated during architecture review.
• security-requirements-extractor -- Supplies the requirements architecture must satisfy.