system-info-hardened

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill is intentionally hardened with restrictive instructions.
  • [PROMPT_INJECTION]: The skill includes defensive instructions that explicitly prevent the agent from ignoring safety boundaries, even when presented with social engineering or claims of authority.
  • [DATA_EXFILTRATION]: Strict guardrails prevent the redirection or transmission of infrastructure-identifying diagnostics to external services.
  • [COMMAND_EXECUTION]: The skill prohibits shell operator chaining (e.g., pipes, semicolons) to ensure diagnostics remain standalone and do not become execution vectors.
  • [EXTERNAL_DOWNLOADS]: External links point to the author's official domains (faberlens.ai and GitHub) for documentation and reporting purposes, which are considered legitimate for this skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 06:32 PM
Security Audit — agent-trust-hub — system-info-hardened