m365-entra-attack

SUSPICIOUS. The skill is internally coherent as a red-team attack skill, but its actual purpose is offensive credential attack and access-validation against M365/Entra targets. There is no clear credential exfiltration or hidden malware behavior in the provided text, yet the capability is inherently high risk for an AI agent because it facilitates unauthorized-seeming security testing, browser-submitted credentials, and attack-chain expansion via related skills and unpinned third-party tooling.