The Agent Skills Directory

[COMMAND_EXECUTION]: The skill employs standard Docker and shell commands (e.g., docker info, docker ps, find, docker build) to analyze project structure and verify build integrity. These operations are restricted to the container environment and project directory, which is essential for its function as a containerization expert.- [CREDENTIALS_UNSAFE]: The skill actively promotes secure secret management. It provides templates demonstrating the use of Docker Secrets (e.g., POSTGRES_PASSWORD_FILE) and BuildKit secret mounts, which are preferred over insecure environment variables for production environments.- [DATA_EXPOSURE]: It scans for standard configuration files like Dockerfile and docker-compose.yml within the current project directory to provide context-aware optimization advice. No attempts to access sensitive system files outside the project scope (e.g., SSH keys, AWS credentials) were identified.- [REMOTE_CODE_EXECUTION]: The skill does not download or execute scripts from untrusted external sources. All referenced base images originate from well-known, trusted registries like Docker Hub official images or Google Container Registry (gcr.io).

docker-expert