The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses sudo to start the Docker daemon and run containers, which grants the agent root-level permissions. Evidence includes sudo dockerd and sudo docker run commands in both README.md and SKILL.md.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes output from external commands. Ingestion points: Docker container output from sudo docker run hello-world in SKILL.md. Boundary markers: None present. Capability inventory: System-wide administrative access via sudo and background process execution. Sanitization: No filtering or validation of command output is performed before the agent processes it.

docker