The Agent Skills Directory

[COMMAND_EXECUTION]: The verify-loop.py script retrieves command strings from a task.json file and executes them using subprocess.run(shell=True). This mechanism allows the execution of arbitrary shell commands on the host if the configuration file is modified.
[COMMAND_EXECUTION]: The install.py script modifies the user's global ~/.claude/settings.json file to register a Stop hook. This hook ensures that stop-hook.py is executed at the end of every agent session to manage the skill's state.
[COMMAND_EXECUTION]: The task.py script executes git commands, specifically git worktree add, via subprocess calls to manage isolated development environments.

do