The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard Docker CLI commands such as docker build, docker run, and docker history to validate project configurations and verify container integrity. These operations are performed on the local system to assist in debugging and optimization.
[INDIRECT_PROMPT_INJECTION]: The skill processes project-specific files to provide tailored advice and validation.
Ingestion points: The skill automatically detects and reads Dockerfile, docker-compose.yml, and .dockerignore files in the current working directory.
Boundary markers: There are no explicit delimiters or instructions to ignore potential injection attempts within the ingested project files.
Capability inventory: The skill can execute the build and runtime logic defined in the project files using the docker command suite.
Sanitization: No input validation or sanitization is applied to the content of the discovered Dockerfiles before they are processed by the build engine.

docker-expert