performing-security-audits

SUSPICIOUS: the skill's purpose and capabilities are broadly aligned, but it gives an AI agent unrestricted shell access for security auditing, a high-risk dual-use capability. I found no direct evidence of malware, credential harvesting, or deceptive third-party routing in the supplied skill, but scope is powerful and under-specified.