tmux-processes

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (SAFE): The skill provides standard patterns for managing long-lived processes in tmux sessions. This is a common and intended developer workflow.
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection surface identified through terminal monitoring.
  • Ingestion points: The skill uses tmux capture-pane -p (SKILL.md) to read process output into the agent's context.
  • Boundary markers: Absent; output is ingested as raw text.
  • Capability inventory: The skill facilitates arbitrary shell command execution via tmux send-keys.
  • Sanitization: No validation or escaping of the captured terminal output is performed.
  • [COMMAND_EXECUTION] (LOW): Dynamic command construction using shell variables. The skill generates bash commands from templates which are executed in a persistent shell session.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:05 PM