NYC

superpowers-lab

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [Command Execution] (LOW): The tmux-wrapper.sh script facilitates the execution of arbitrary interactive commands and keystrokes. While this is the skill's primary purpose, it provides a high-privilege interaction channel with the host system.
  • [Indirect Prompt Injection] (LOW): The skill is susceptible to indirect prompt injection from data ingested via terminal panes.
  • Ingestion points: skills/using-tmux-for-interactive-commands/tmux-wrapper.sh uses tmux capture-pane to feed terminal content back to the agent.
  • Boundary markers: Output is prefixed with a simple header but lacks robust delimiters or safety instructions.
  • Capability inventory: The skill allows full terminal session control, including command execution and keystroke injection.
  • Sanitization: No sanitization is performed on the captured terminal content.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:30 PM