The Agent Skills Directory

[SAFE]: The skill is a legitimate security auditing utility designed to help researchers identify Denial of Service vulnerabilities in XML and YAML parsers. No malicious code or exfiltration patterns were detected.\n- [COMMAND_EXECUTION]: The skill includes shell command templates using grep to identify vulnerable code sinks in local files. These are benign search operations consistent with the skill's stated purpose.\n- [PROMPT_INJECTION]: The skill contains example payloads for Billion Laughs attacks. These are provided as static documentation and testing templates rather than instructions to bypass agent safety guidelines.

entity-expansion