audit

Warn

Audited by Gen Agent Trust Hub on Nov 21, 2026

Risk Level: MEDIUMDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONPERSISTENCE
Full Analysis
  • [DATA_EXFILTRATION]: The skill contains commands that access highly sensitive system files, such as /etc/shadow and /etc/passwd, to perform security audits and check for misconfigurations like empty passwords.
  • [COMMAND_EXECUTION]: Provides instructions for system-level operations including service management, audit rule modification, and network scanning with Nmap.
  • [EXTERNAL_DOWNLOADS]: Includes commands to install several external security tools (auditd, lynis, OpenSCAP, AIDE) from standard package repositories.
  • [PERSISTENCE]: Includes a script to be placed in /etc/profile.d/ to send email alerts upon SSH login, which executes automatically for every login session.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from untrusted sources, specifically system authentication logs (/var/log/auth.log). 1. Ingestion points: /var/log/auth.log, /etc/shadow, /etc/passwd (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Package installation (apt/yum), system audit configuration (auditctl), and network scanning (nmap) (SKILL.md). 4. Sanitization: Absent; the skill uses standard text processing tools like grep and awk to parse log content.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 21, 2026, 11:44 AM