magic-intelligence-system

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a template-driven system for authoring game lore. All code is local and performs only basic JSON validation.
  • [REMOTE_CODE_EXECUTION]: No remote code execution patterns were detected. The only script present (scripts/validate.mjs) is a dependency-free Node.js script used for internal data validation.
  • [DATA_EXFILTRATION]: No network operations or sensitive file access patterns were found. The skill operates entirely on its own project data.
  • [PROMPT_INJECTION]: The instructions in SKILL.md and CLAUDE.md are focused on maintaining consistency with the established fictional canon ('Arcanea') and do not attempt to bypass AI safety guardrails.
  • [COMMAND_EXECUTION]: The skill uses standard development commands (node scripts/validate.mjs) which are appropriate for its stated purpose of validating data files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 11:06 AM
Security Audit — agent-trust-hub — magic-intelligence-system