The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes locally bundled Python scripts to perform static analysis on Docker files. Specifically, it uses python3 scripts/dockerfile_analyzer.py and python3 scripts/compose_validator.py to audit configurations. These scripts use standard Python libraries and do not perform network operations.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided content (Dockerfiles and Docker Compose files) to generate optimization suggestions, which creates a potential surface for indirect prompt injection.
Ingestion points: The agent reads user-provided files such as Dockerfile and docker-compose.yml for analysis.
Boundary markers: The instructions do not explicitly define boundary markers to separate user data from instructions, though the use of static analysis scripts provides a degree of isolation.
Capability inventory: The skill can execute bundled Python scripts and generate text-based suggestions or modified configuration files.
Sanitization: The included analysis scripts utilize regular expression pattern matching to identify issues, which naturally sanitizes the input by treating it as static text rather than executable instructions.

docker-development