The Agent Skills Directory

[REMOTE_CODE_EXECUTION] (HIGH): The script scripts/scan-infrastructure.sh downloads and executes a Kubernetes manifest directly from raw.githubusercontent.com/aquasecurity/kube-bench/main/job.yaml. As the aquasecurity organization is not in the trusted list, this constitutes an unverifiable remote code execution risk.
[COMMAND_EXECUTION] (HIGH): The script scripts/scan-infrastructure.sh performs privilege escalation using sudo to run lynis. It also executes docker-bench-security with high-privilege flags (--net host, --pid host, --cap-add audit_control), which allows the container to bypass host isolation.
[EXTERNAL_DOWNLOADS] (HIGH): The skill fetches resources from an external GitHub account (aquasecurity) not included in the predefined trusted scope.

security-hardening