safety-guard

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill contains multiple shell scripts in payloads.md designed for security auditing, compliance verification, and runtime monitoring. These scripts leverage the Bash tool to perform tasks such as scanning command history for violations, checking time window compliance, and analyzing request logs for rate limit breaches.
  • [PRIVILEGE_ESCALATION]: The runtime monitoring scripts in payloads.md utilize commands that require elevated privileges, such as sudo tcpdump for packet capture and find with permission flags to detect SUID binaries. These operations are intended for legitimate security oversight and are categorized for human confirmation within the skill's "Freeze" mode classification.
  • [INDIRECT_PROMPT_INJECTION]: As a security interceptor that processes commands generated by other skills or agents, this skill represents a potential target for indirect prompt injection. An attacker might attempt to bypass the safety guard using techniques like encoding, long-context dilution, or multi-turn attacks. The skill explicitly identifies these as "Defense Evasion Techniques" and provides a framework (Scope Lock Protocol, Pre-Action Checklist) to mitigate such risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 04:36 PM