The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected in this skill. It defines a set of instructional steps for designing secure system architectures. The skill includes commendable defensive guardrails for the agent, such as requiring Infrastructure-as-Code (IaC) for review before making IAM changes, forbidding root-level permission assignments, and enforcing MFA and fail-closed mechanisms in the resulting designs. No external code, network operations, or sensitive file access patterns were found.

secure-architecture