tmux

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (SAFE): The skill utilizes the Bash tool to run standard tmux commands for process management (new-window, send-keys, capture-pane). These are consistent with the skill's stated purpose of managing background tasks.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill implements a pattern where the agent reads output from a running process, creating an attack surface for indirect prompt injection.
  • Ingestion points: The tmux capture-pane -p -t "server-log" command in SKILL.md brings untrusted process output directly into the agent's context.
  • Boundary markers: There are no delimiters or warnings to the agent to ignore instructions contained within the captured output.
  • Capability inventory: The agent has access to the Bash tool, which allows for broad command execution should it be tricked by captured output.
  • Sanitization: No sanitization or filtering of the captured output is suggested or performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:25 PM