skills/goldengrape/agent-company/tmux/Gen Agent Trust Hub

tmux

Pass

Audited by Gen Agent Trust Hub on Feb 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill is designed to execute arbitrary shell commands via tmux send-keys. While this is high-risk functionality, the implementation uses the -l (literal) flag in examples and follows best practices for script arguments. The severity is lowered as this is the primary intended purpose of the skill.
  • [PROMPT_INJECTION] (LOW): Detected potential for Indirect Prompt Injection (Category 8).
  • Ingestion points: Terminal output is ingested into the agent context via tmux capture-pane (utilized in scripts/wait-for-text.sh and documented in SKILL.md).
  • Boundary markers: Absent. There are no explicit delimiters or instructions provided to the agent to ignore command-like patterns within the captured terminal text.
  • Capability inventory: The skill possesses high-privilege capabilities including arbitrary shell command execution (tmux send-keys) and session management (tmux kill-server).
  • Sanitization: Absent. Output is captured and processed as raw text.
  • [DATA_EXPOSURE] (SAFE): While capture-pane can read sensitive information if it is printed to the terminal, the scripts target private sockets in /tmp by default, reducing the risk of accidental exposure to other users on a multi-user system.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 20, 2026, 05:24 PM