The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts scripts/agent.py and scripts/process.py utilize the subprocess.run function to execute docker and kubectl commands. These calls are implemented using a list of arguments, which is a secure method that prevents shell-level command injection. The commands are used strictly for gathering system state for security analysis.
[INDIRECT_PROMPT_INJECTION]: The skill processes output from container runtime and Kubernetes management tools. While this constitutes an ingestion point for external data, the scripts are designed for observational reporting and do not perform automated high-privilege actions or control-flow changes based on the data. The ingestion is handled through structured parsing (JSON), providing a layer of validation.

detecting-container-drift-at-runtime