The Agent Skills Directory

[SAFE]: The skill provides comprehensive security guidance for Docker management and uses local scripts for linting without external dependencies. It includes explicit warnings against high-risk actions like mounting the Docker socket into untrusted containers or using the --privileged flag.
[COMMAND_EXECUTION]: The skill uses standard Docker CLI commands and a bundled Python script for validating Dockerfiles and Compose files. These operations are essential for the skill's functionality and do not involve executing untrusted remote code or using unsafe shell interpolation.
[DATA_EXFILTRATION]: No suspicious network activity or exfiltration patterns were identified. The skill includes instructions to prevent secrets from being baked into image layers through the use of BuildKit secret mounts and .dockerignore files.

docker