mmaideveloper/aiarchitecttoolkit / skills/security-review
security-review
UnverifiedReview a proposed or implemented solution for security risks, threat coverage, control effectiveness, and security evidence. Use for focused architecture security reviews, threat-informed release checks, or security findings; do not use as penetration testing or as proof of compliance.
Security Review
Produce an evidence-backed security assessment without inventing controls, classifications, test results, risk acceptance, or approval.
Workflow
- Read
AGENTS.md,architecture/toolkit-profile.yamlwhen present, the requested review scope, architecture artifacts, code/configuration, deployment evidence, and prior security findings. - Establish assets, identities, data classifications, trust boundaries, entry points, privileged operations, external dependencies, deployment environments, and likely threat actors. Mark missing classifications or ownership as
To verify. - Build the threat-and-control matrix in references/security-review-method.md. Use a recognized threat method when required by the project; otherwise use a concise STRIDE-informed analysis without claiming exhaustive coverage.
- Assess preventive, detective, responsive, and recovery controls. Verify implementation and test evidence rather than relying on design statements.
- Report exploitable weaknesses, control gaps, unsafe defaults, unverifiable claims, and accepted residual risks separately. Include exact evidence and practical remediation.
- Persist the review only when requested, using the configured review path or
architecture/reviews/SEC-NNN-<slug>.md.
Review Areas
- Identity, authentication, authorization, privilege boundaries, service identities, and emergency access
- Secrets, keys, certificates, rotation, and credential exposure
- Network boundaries, ingress/egress, private connectivity, segmentation, and administrative access
- Data minimization, encryption, integrity, retention, deletion, backup, restoration, and tenant isolation
- Input/output handling, injection, deserialization, file processing, SSRF, supply chain, and dependency provenance
- Logging, alerting, incident response, forensic usefulness, abuse detection, and security test coverage
- AI-specific threats when applicable: prompt injection, unsafe tool use, data exfiltration, model or retrieval poisoning, excessive agency, insecure output handling, and model/vendor trust
Rules
- Treat scans and checklists as evidence inputs, not proof of security.
- Do not expose secrets, exploit live systems, perform destructive tests, or expand testing scope without explicit authorization.
- Distinguish design gaps from implementation defects and operational misconfiguration.
- Map each finding to an affected asset or boundary, credible threat, evidence, consequence, remediation, and owner when known.
- A risk exception requires scope, rationale, compensating controls, accountable owner, expiry/review date, and approval evidence.
- Never claim certification, compliance, or release approval.
Output
Lead with the security posture and release-blocking findings. Include scope and limitations, assets and trust boundaries, threat-and-control matrix, prioritized findings, positive verified controls, residual risks, unverified items, and recommended owner/action.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 464k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 464k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 464k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 192k | Updated | Claude CodeCodex | Verified | Install |