The Agent Skills Directory

[SAFE]: No security issues were identified during the analysis of the skill's instructions or scripts.
[COMMAND_EXECUTION]: The skill uses local tmux commands to manage terminal sessions and send keystrokes. This is the core intended functionality. The scripts find-sessions.sh and wait-for-text.sh use robust variable quoting and standard safety flags to prevent command injection.
[DATA_EXPOSURE]: The skill utilizes a configurable socket directory (defaulting to /tmp/clawdbot-tmux-sockets) for managing tmux sessions. This is standard operational procedure for terminal multiplexers and does not involve exposure of sensitive user data or credentials.
[PROMPT_INJECTION]: The skill instructions do not contain attempts to bypass agent safety filters or override system constraints. While the documentation provides examples of running other coding agents in autonomous modes (e.g., using --yolo), this is presented as a workflow example for external tools and does not affect the safety posture of the analyzing agent.

tmux