implementing-backup-strategies

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by converting user-specified parameters into shell scripts.\n- Ingestion points: User-provided requirements and triggers for backup strategies processed by the agent in SKILL.md.\n- Boundary markers: Absent; no instructions are provided to the agent to delimit user data or ignore instructions embedded within the inputs.\n- Capability inventory: Shell execution via Bash (tar, rsync, aws) and extensive file system access (Read, Write, Edit, Grep, Glob) as defined in SKILL.md and referenced scripts.\n- Sanitization: Absent; there are no instructions for validating, escaping, or sanitizing user parameters before their interpolation into generated shell scripts.\n- [SAFE]: The use of persistence mechanisms such as cron jobs and systemd timers is the intended primary function for scheduling recurring backups and does not represent a security risk in this context.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 10:09 PM
Security Audit — agent-trust-hub — implementing-backup-strategies