The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard utility commands such as grep, test, and docker within its checkpoints to perform static analysis and linting of local Docker configuration files. These operations are diagnostic in nature and restricted to evaluating the local project structure.
[SAFE]: The skill explicitly instructs users on how to avoid security pitfalls, such as hardcoding credentials in ENV or ARG instructions, and provides checkpoints (DC-10 through DC-13) to detect such leaks. It also promotes the use of non-root users, BuildKit secrets, and .dockerignore optimization to prevent accidental data exposure.
[EXTERNAL_DOWNLOADS]: Reference materials mention well-known CI/CD components, such as official GitHub Actions (actions/checkout, docker/setup-buildx-action) and established Ansible roles. These are documented as part of standard development workflows and do not constitute risky external dependencies or unauthorized downloads.
[PROMPT_INJECTION]: No patterns associated with prompt injection or agent behavior overrides were detected. The instructions are focused on technical guidance for containerization and follow standard instructional formatting.

docker-development