The Agent Skills Directory

[SAFE]: The skill provides instructional content for a 'judge' agent to evaluate the thoroughness of a security audit. It enumerates standard web application vulnerability categories such as SQL Injection, XSS, and Authentication flaws as a checklist for the agent to follow.- [SAFE]: The decision logic is designed to ensure maximum coverage and prevents premature termination of security testing by requiring verification against known vulnerability databases and checking for untested attack surfaces. This promotes thorough testing rather than bypassing safety or operational standards.- [SAFE]: No evidence of prompt injection, data exfiltration, or obfuscation was found in the skill metadata or body. The instructions support legitimate security research and evaluation workflows and do not involve unauthorized network operations or file access.

judge-pentest