geo-fundamentals

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Data Exposure & Exfiltration] (SAFE): The analysis tool reads local files with specific extensions (.html, .jsx, .tsx) and ignores sensitive system or configuration directories. No data is transmitted externally.
  • [Remote Code Execution] (SAFE): The skill does not download external scripts or execute code from remote sources.
  • [Prompt Injection] (SAFE): No instructions designed to override agent safety or extract system prompts were detected in the skill documentation.
  • [Dynamic Execution] (SAFE): The script uses static regular expression patterns for its audit and does not utilize dynamic execution functions like eval or exec.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:20 PM