The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs process management by directing the agent to execute system shell commands lsof -ti :<port> and kill -9 <PID>.
[COMMAND_EXECUTION]: User input from the $ARGUMENTS variable is directly interpolated into the lsof command template. This introduces a command injection vulnerability where a user could provide input containing shell delimiters (e.g., ;, &, |) to execute arbitrary secondary commands.
[COMMAND_EXECUTION]: The skill utilizes the kill -9 (SIGKILL) command to forcefully terminate processes. If the port-to-PID lookup is manipulated through injected input, an attacker could potentially terminate unauthorized or critical system processes.

kill-process