The Agent Skills Directory

[SAFE]: The skill implements a meta-review framework for the agent to verify its own output for logical consistency, scope, and assumptions before delivery.
[SAFE]: Explicit instructions require verification of CVE IDs and CVSS scores against authoritative primary sources such as the National Vulnerability Database (NVD).
[SAFE]: Promotes security best practices by distinguishing between illustrative pattern-level payloads and potentially harmful version-specific exploit chains.
[SAFE]: References established industry standards and organizations, including NIST, FIRST, and MITRE ATT&CK, for technical validation.

verification-loop