skills/ajoslin/dot/tmux/Gen Agent Trust Hub

tmux

Fail

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: HIGHCOMMAND_EXECUTION
Full Analysis
  • Command Execution (HIGH): The script scripts/tmux_run_job.py accepts a --command argument which is interpolated into a shell command string and executed via bash -lc within a tmux session. This allows for arbitrary command execution on the host machine.
  • Persistence Mechanisms (HIGH): The script scripts/tmux_healthcheck_cron.py provides functionality to automatically modify the user's crontab (crontab - and crontab -l) to install recurring background tasks. This is a common persistence vector used in malicious activities.
  • Indirect Prompt Injection (LOW): The skill exhibits a significant attack surface for indirect injection. Ingestion points: Data enters via the --command argument in scripts/tmux_run_job.py. Boundary markers: None; the command is interpolated into a shell script. Capability inventory: subprocess.run is used to execute tmux commands that spawn bash processes; crontab is used to modify system schedules. Sanitization: While shlex.quote is used for the outer shell wrapper, the actual payload command is directly interpolated, leaving it vulnerable to shell injection if the source data is untrusted.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 21, 2026, 07:24 AM