The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard system utilities including docker, docker-compose, find, and grep to analyze the local development environment and validate container builds. These operations are essential for the skill's primary purpose of Docker optimization and troubleshooting.
[EXTERNAL_DOWNLOADS]: The provided templates reference official and well-known images from Docker Hub and Google Container Registry (e.g., node:18-alpine, postgres:15-alpine, gcr.io/distroless/nodejs18-debian11). These are recognized as safe, industry-standard sources for container base images.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted local project files like Dockerfile and docker-compose.yml which are then processed by the agent.
Ingestion points: Files identified via find . -name "Dockerfile*" and find . -name "*compose*.yml".
Boundary markers: None explicitly defined for file content ingestion.
Capability inventory: Executes docker build, docker run, and docker exec as part of its validation logic.
Sanitization: No specific sanitization or escaping of file content is mentioned before processing.

docker-expert