The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes instructions for the agent to execute shell commands for building and running Docker containers, such as docker build and docker run. These commands facilitate the execution of instructions and code within a containerized environment.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its automated testing and optimization commands. If the agent processes an untrusted application, malicious code or comments within that application could influence the agent's behavior during execution steps like docker run --rm app python -c "import app".
Ingestion points: SKILL.md (Agentic Optimizations section)
Boundary markers: None provided to isolate the agent from container output.
Capability inventory: Uses Bash to execute docker and pip commands.
Sanitization: No sanitization is performed on the application code prior to building or running the container.
[EXTERNAL_DOWNLOADS]: Fetches Docker base images from official repositories (Python Slim) and Google Container Registry (gcr.io/distroless). These are recognized as trusted and well-known sources.

python-containers