The Agent Skills Directory

[PROMPT_INJECTION] (SAFE): The skill contains no instructions intended to override agent behavior or bypass safety filters.\n- [DATA_EXFILTRATION] (SAFE): No credentials or sensitive data access patterns were found. The skill includes advice on using .dockerignore to prevent accidental exposure of .env files. Hardcoded credentials in Docker Compose examples are clearly placeholders for local development.\n- [EXTERNAL_DOWNLOADS] (SAFE): The skill references standard Docker base images and package managers (pip, npm, apt). These are considered safe in the context of a Docker pattern guide.\n- [COMMAND_EXECUTION] (SAFE): All command examples provided are standard for Dockerfile construction and container management.\n- [PRIVILEGE_ESCALATION] (SAFE): The skill explicitly demonstrates how to create and use non-root users within containers, which is a recommended security practice.

docker-patterns