cro-methodology

Pass

Audited by Gen Agent Trust Hub on Feb 23, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill is purely documentation-based and contains no executable logic.\n
  • Evidence: All analyzed files (SKILL.md and reference guides) are Markdown documents providing procedural instructions and frameworks.\n
  • Risk Assessment: Without scripts, binaries, or configuration files, there is no surface for direct code execution or automated malicious behavior.- [SAFE]: Analysis of the methodology guidelines and instructional text revealed no security threats.\n
  • Analysis: The instructions focus on user research and A/B testing methodologies without attempting to override agent safety protocols or exfiltrate sensitive data.\n
  • Observation: The skill contains an informational link to a book on Amazon.com. As Amazon is a well-known service and the link's affiliate tag aligns with the author's metadata, this is documented as normal vendor functionality and does not impact the safety verdict.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 23, 2026, 11:09 PM