Privilege Escalation Methods
Audited by Socket on Feb 16, 2026
1 alert found:
Malware[Skill Scanner] URL pointing to executable file detected All findings: [CRITICAL] command_injection: URL pointing to executable file detected (CI010) [AITech 9.1.4] [HIGH] data_exfiltration: Credential file access detected (DE002) [AITech 8.2.3] This skill is an explicit, actionable post-exploitation playbook that instructs operators how to escalate privileges, harvest credentials, and perform domain compromise. Its capabilities are coherent with its stated purpose but are inherently malicious when used outside an authorized penetration test context — the document includes direct examples of exfiltration to attacker-controlled endpoints and uses known offensive tools. Treat as malicious/offensive content: do not include in a general-purpose agent or public skill store without strict access controls and lawful-use gating. LLM verification: This Skill contains explicit, actionable post-exploitation instructions that enable local privilege escalation, credential harvesting, persistence, and domain compromise. While framed for legitimate penetration testing and red-team use, the content is highly dangerous if used outside controlled, authorized environments. The presence of an example attacker URL and direct guidance to copy sensitive artifacts (NTDS.dit, SSH keys) and set SUID on binaries demonstrates high potential for abuse. Treat