The Agent Skills Directory

[SAFE]: The skill serves as a documentation resource for Docker best practices and does not contain any executable code or malicious instructions. It explicitly promotes security hardening techniques, such as:
Using non-root users in Dockerfiles (USER appuser).
Dropping container capabilities (cap_drop: ALL).
Implementing read-only filesystems.
Using .dockerignore to prevent sensitive files like .env from being included in image layers.
Explicitly warning against hardcoding secrets (e.g., API_KEY) in image layers or configuration files.
[CREDENTIALS_UNSAFE]: While the skill contains example credentials (e.g., POSTGRES_PASSWORD: postgres), these are used strictly as local development placeholders in documentation and do not represent a security risk. The skill also explicitly identifies hardcoding real secrets as an anti-pattern.

docker-patterns