offensive-cloud-role
Warn
Audited by Socket on Sep 5, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill is internally consistent and does not show third-party credential exfiltration or suspicious installers, but it is an offensive cloud-operations skill that enables an AI agent to use acquired cloud credentials and metadata tokens to enumerate IAM, storage, and lateral-movement paths, then pass follow-on access to other offensive skills. High operational risk comes from the offensive capability itself, not from covert malware behavior.
Confidence: 90%Severity: 81%
Audit Metadata