The Agent Skills Directory

[SAFE]: The skill functions as an educational resource and template library for Docker and Docker Compose. It actively promotes security hardening by including examples of non-root user configuration, dropping capabilities, and using read-only filesystems.
[COMMAND_EXECUTION]: Includes standard Docker CLI and Docker Compose commands for container management, debugging, and health checks (e.g., docker compose logs, wget for internal health checks). These are standard operational commands for the described use case.
[EXTERNAL_DOWNLOADS]: References official and well-known Docker images from trusted registries (e.g., node:22-alpine, postgres:16-alpine, redis:7-alpine).
[CREDENTIALS_UNSAFE]: Uses common placeholder credentials for local development database examples (e.g., postgres:postgres). These are recognized as safe boilerplate for instructional purposes and the skill explicitly warns against hardcoding real secrets.

docker-patterns