NYC

testing-anti-patterns

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [PROMPT_INJECTION] (SAFE): The content consists of educational guidance for a language model to follow when assisting with software testing. No attempts to override core safety instructions or bypass system constraints were found.- [DATA_EXFILTRATION] (SAFE): No access to sensitive file paths (~/.ssh, .env, etc.) or network exfiltration patterns were detected.- [REMOTE_CODE_EXECUTION] (SAFE): No remote scripts, package installations (npm/pip), or dynamic execution of untrusted code were found.- [NO_CODE] (SAFE): The skill is entirely descriptive, containing markdown instructions and illustrative code snippets that are not meant to be executed directly by the system shell.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:21 PM