The Agent Skills Directory

[DATA_EXPOSURE] (LOW): The use of tmux capture-pane allows the agent to read contents from active terminal panes. This poses a risk of exposing sensitive data such as API keys, secrets, or private environment variables if they are currently visible in those terminal sessions.
[INDIRECT_PROMPT_INJECTION] (LOW): The skill ingests untrusted text from tmux panes, which could contain malicious instructions designed to manipulate the agent's behavior.
Ingestion points: Terminal output read via tmux capture-pane in SKILL.md.
Boundary markers: No delimiters or specific instructions are provided to distinguish captured content from the agent's system instructions.
Capability inventory: Execution of tmux commands and execution of local Python scripts.
Sanitization: The skill mentions a function strip_ansi_and_control to remove terminal formatting, but it does not perform semantic sanitization of the content itself.
[COMMAND_EXECUTION] (SAFE): The skill utilizes standard system utilities including tmux, grep, and nix for its intended debugging functionality.

tmux-status-debug