skills/lidge-jun/cli-jaw-skills/tmux/Gen Agent Trust Hub

tmux

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes tmux send-keys to provide input to terminal sessions. This capability allows the agent to execute arbitrary commands or interact with any CLI application running within the tmux environment.
  • [DATA_EXFILTRATION]: Through tmux capture-pane, the agent can read the entire scrollback buffer of a terminal pane. This exposes any sensitive data displayed during the session—such as credentials, configuration secrets, or private data—to the agent's context.
  • [PROMPT_INJECTION]: The skill ingests untrusted data by scraping terminal output from active sessions. If a terminal session displays attacker-controlled content (such as log files or web output) containing instructions, the agent may interpret that content as a valid command or behavioral override. The skill lacks boundary markers to distinguish between legitimate terminal data and embedded instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 06:32 PM