The Agent Skills Directory

[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface. 1. Ingestion points: The skill searches for local Dockerfiles and docker-compose.yml files using find in SKILL.md. 2. Boundary markers: No boundary markers or 'ignore instructions' are used when processing these files. 3. Capability inventory: The skill has the capability to build and execute containers using docker build, docker run, and docker exec in SKILL.md. 4. Sanitization: No sanitization or validation of the file contents is performed before execution. Remediation: Implement human-in-the-loop checkpoints before building or running discovered containers.
[COMMAND_EXECUTION]: The skill executes powerful shell commands like docker build, docker run, and docker exec on files detected in the workspace. This automated validation process can lead to local code execution if the workspace contains maliciously crafted Docker configurations.

docker-expert