dns-recon

This skill is internally consistent and uses trusted package sources, with no credential harvesting or third-party proxying. However, it equips an AI agent with offensive DNS reconnaissance and AXFR vulnerability-testing behavior, so it should be classified as high security risk but not malware.