Linux Production Shell Scripts

Warn

Audited by Gen Agent Trust Hub on Nov 1, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONPERSISTENCEINDIRECT_PROMPT_INJECTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: Provides templates using powerful system utilities including rm, rsync, mysqldump, and systemctl for administrative tasks.
  • [PRIVILEGE_ESCALATION]: Includes scripts that require or use sudo for tasks like package installation (apt-get install), service management (systemctl restart), and user management (useradd).
  • [PERSISTENCE]: Provides a Task Scheduler script that modifies the system's crontab (crontab -l ... | crontab -) to schedule recurring tasks.
  • [REMOTE_CODE_EXECUTION]: Includes a Remote Script Execution template that uses ssh to pipe local script content to a remote shell (ssh ... "bash -s" < ...).
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: Shell scripts use unvalidated variables such as $1, $2, $package, $username, and $remote_server across various phases in SKILL.md.
  • Boundary markers: No delimiters or instructions are provided to prevent the agent from including malicious shell metacharacters in these variables.
  • Capability inventory: The templates have access to file deletion (rm), remote access (ssh, rsync), and privilege escalation (sudo).
  • Sanitization: There is no evidence of escaping or validation logic for the values interpolated into the shell scripts.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 1, 2026, 11:35 PM