The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documents and provides templates for using the platform's shell command execution feature (!command). These examples (e.g., !git status) are standard development operations and are correctly placed within markdown code blocks to prevent execution when the skill itself is loaded.
[DATA_EXFILTRATION]: No network calls, unauthorized external connections, or sensitive data harvesting patterns were identified.
[PROMPT_INJECTION]: The instructions focus on structured content generation and follow professional conventions without attempting to override system safety guidelines or core agent behavior.

create-command