skill-creator-openai

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • [SAFE] (SAFE): No malicious behavior or security vulnerabilities detected. The scripts use yaml.safe_load() to prevent arbitrary code execution during metadata parsing and Path.resolve() for secure file path handling. Validation logic is present to ensure metadata follows a strict schema.
  • [Category 7: Metadata Poisoning] (LOW): The quick_validate.py script proactively mitigates metadata poisoning by enforcing a strict hyphen-case regex for skill names and stripping angle brackets (<, >) from descriptions to prevent HTML/XSS injection in downstream UIs.
  • [Category 10: Dynamic Execution] (SAFE): The code avoids all forms of eval(), exec(), or unsafe deserialization, opting for the PyYAML SafeLoader for processing skill configurations.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 12:49 PM