team-agents

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill employs multiple bash scripts to manage the lifecycle of agent teams. Scripts like cleanup.sh, killshot.sh, and panes.sh interact with tmux panes using the maw utility to provide visual orchestration. shutdown-worktrees.sh utilizes standard git worktree commands to ensure that temporary development branches created for agents are properly removed.
  • [COMMAND_EXECUTION]: System diagnostic scripts such as doctor.sh and team-ops.sh perform routine maintenance tasks. They inspect the local process tree via /proc to identify active Claude processes and cross-reference them with the skill's state stored in ~/.claude/teams/ to detect and resolve orphaned agent sessions.
  • [COMMAND_EXECUTION]: The spawn-skills.sh script dynamically generates ephemeral markdown instruction files in ~/.claude/skills/. This allows the lead agent to create specific interfaces for communicating with sub-agents in a manual control mode, facilitating structured message passing between team members.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 01:15 PM
Security Audit — agent-trust-hub — team-agents