The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions to execute arbitrary shell commands within TMUX panes using the send-keys function. While intended for service management, this provides a mechanism for running any command. Evidence: The pattern tmux send-keys -t "target" "command here" Enter is explicitly listed in the Command Reference in SKILL.md.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it reads and analyzes terminal output which is considered untrusted data. 1. Ingestion points: Terminal history is read into the agent's context using tmux capture-pane in SKILL.md. 2. Boundary markers: No delimiters or "ignore" instructions are used when processing the captured terminal text. 3. Capability inventory: The agent has the ability to execute further shell commands via tmux send-keys based on its analysis of the captured output. 4. Sanitization: There is no evidence of sanitization or validation of the captured terminal data before analysis.

tmux-aware