The Agent Skills Directory

[SAFE]: The skill contains procedural instructions for a security persona and does not include any executable scripts, network operations, or hardcoded credentials.
[SAFE]: The instructions incorporate strong security guardrails, specifically directing the agent to avoid running applications, builds, tests, or installing dependencies, which mitigates risks associated with Remote Code Execution (RCE) and Privilege Escalation.
[SAFE]: The workflow relies on a subagent delegation model with read-only limits, ensuring that the auditing process remains contained within a source-analysis boundary.

security-review