email-protocol-attack
Audited by Socket on Sep 19, 2026
2 alerts found:
Securityx2SUSPICIOUS: the skill is a coherent offensive email-attack playbook, but that purpose itself gives an AI agent high-risk capabilities including phishing, brute forcing, spoofing, mailbox access, and Exchange attack workflows. There is no strong evidence of hidden malware or deceptive credential harvesting, but the autonomous security-attack scope is disproportionate for general agent use and poses high operational risk.
This fragment is an offensive mail-server assessment and credential-attack guide, not normal application or npm library functionality. It presents commands capable of unauthorized password guessing, mailbox access, service enumeration, and vulnerability reconnaissance. It does not itself demonstrate malware such as persistence, local data theft, reverse shells, or cryptomining, but it creates a substantial security risk if executed against systems without authorization. Because only part of a larger file is shown, the assessment is limited to the visible fragment.