The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to facilitate iterative, autonomous shell command execution through the paseo loop run command. It explicitly instructs the agent to use user-provided input ($ARGUMENTS) to construct CLI calls, which creates a significant surface for command injection if shell metacharacters are included in the user's request.
[COMMAND_EXECUTION]: The skill utilizes a --verify-check flag that accepts and executes arbitrary shell commands. This capability allows the agent to run code persistently in a background daemon process, increasing the impact of any potential malicious instructions or compromised iterations.
[COMMAND_EXECUTION]: The $ARGUMENTS variable is directly interpolated into the skill's instructions, and the agent is tasked with translating this intent into execution. There are no instructions provided for escaping, validation, or sanitization of this input before it is passed to the shell environment.

paseo-loop