docker-essentials

Arbitrary command execution via docker exec with user input

SKILL.md documents patterns like 'docker exec container_name ls -la' and 'docker exec -it container_name bash' that would allow injection if container_name or commands are taken from user input without validation. An attacker could inject shell metacharacters to execute arbitrary commands on the host.

Path traversal and arbitrary file access via volume mounts

Documentation shows 'docker run -v /host/path:/container/path' and '-v $(pwd):/app' patterns. If /host/path is constructed from user input without validation, attackers could mount sensitive directories like /etc, /root, or / to exfiltrate data or modify system files.

Credential exposure via environment variables

SKILL.md demonstrates 'POSTGRES_PASSWORD=secret' passed via -e flag. These appear in 'docker inspect', process lists, and logs. The documentation encourages this insecure pattern without warnings about credential management.

No input validation guidance for container names or image tags

All examples assume trusted input for container_name, image_name parameters. No guidance on validating these inputs against command injection (e.g., 'container; rm -rf /' or image names with embedded commands).

Privileged operations without authorization checks

Commands like 'docker system prune -a' (deletes all images), 'docker rm -f' (force remove), and 'docker-compose down -v' (delete volumes with data) are documented without any discussion of access control or confirmation requirements.

No guidance on network isolation or security boundaries

Exposing all container logs without filtering sensitive data