The Agent Skills Directory

[SAFE]: The skill acts as a read-only reference card. It has no tools enabled in the allowed-tools configuration, ensuring it cannot perform any actions on the system.
[SAFE]: While the skill mentions file paths (e.g., ~/.origin/), these are documented as the plugin's own storage directory rather than sensitive system paths or credentials.
[SAFE]: The instructions to the agent to output the help menu "verbatim" are standard for formatting documentation and do not attempt to bypass safety constraints or override core agent behaviors.
[SAFE]: The skill mentions local installation and configuration commands (e.g., /init, origin model install), but these are provided as informational text for the user and are not executed by this skill.

help