The Agent Skills Directory

[SAFE]: The skill provides a robust multi-agent orchestration framework for software development. It organizes tasks through specialized roles with distinct responsibilities and clear handoff protocols, ensuring a structured development lifecycle.
[SAFE]: Security considerations are explicitly integrated into the agent instructions. The Reviewer agent (Luna) is directed to perform security reviews targeting injection flaws, authentication bypasses, and secret exposure. The Builder agent (Mason) is instructed to implement security baselines such as query parameterization and input sanitization.
[SAFE]: The DevOps agent (Dep) promotes industry-standard security practices, including the use of multi-stage Docker builds, non-root users, and automated security scanning tools like trivy, npm audit, and pip audit.
[SAFE]: Analysis of the skill instructions revealed no evidence of prompt injection, obfuscated code, or unauthorized data exfiltration. The Orchestrator's project state management and reference-based artifact handling are designed for efficiency and context preservation without introducing security risks.

agent-squad