macos-security-bypass

High-risk offensive security skill. Its actual footprint is coherent with a red-team bypass playbook, but that footprint is inherently dangerous for an AI agent because it teaches endpoint protection bypass, privilege abuse, persistence, and transitive loading of other offensive skills.

This fragment is highly suspicious and strongly indicative of malicious intent: it is an offensive playbook for bypassing macOS TCC privacy controls and abusing PPPC/MDM to pre-approve sensitive permissions, includes a PPPC XML permission-grant template, and provides concrete local reconnaissance commands targeting TCC.db and entitlements. If present in a supply-chain dependency, it should be treated as a security-critical risk and excluded/reviewed with priority.