evaluation-criteria

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [Prompt Injection] (SAFE): No instructions attempting to override agent behavior or bypass safety filters were detected. The markdown content provides standard guidelines for objective assessment.
  • [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, hardcoded credentials, or network operations were found. The skill does not access the local file system or external APIs.
  • [Remote Code Execution] (SAFE): No external dependencies, package managers (pip/npm), or remote script execution patterns (curl|bash) were found.
  • [Persistence & Privilege Escalation] (SAFE): No commands for gaining elevated privileges or maintaining access across sessions (e.g., cron jobs, shell profiles) were found.
  • [Indirect Prompt Injection] (SAFE): The skill defines a framework for evaluating external deliverables but lacks any executable components or unsafe interpolation patterns that could trigger code execution from processed data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:36 PM