skills/quickcall-dev/skills/dag-fleet/Gen Agent Trust Hub

dag-fleet

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs advanced orchestration using shell scripts to manage tmux sessions, windows, and lifecycles. It programmatically executes worker tasks using the claude and codex CLI tools.
  • [COMMAND_EXECUTION]: Implements comprehensive process management, utilizing pgrep -f and kill -9 to ensure that workers and their descendant subprocesses are properly terminated during fleet cleanup or individual worker stops.
  • [COMMAND_EXECUTION]: Dynamically generates and executes temporary shell runner scripts (.run.sh) to manage environment variables, directory contexts, and tool restrictions for sub-agents running within tmux panes.
  • [SAFE]: Utilizes inline Python scripts via python3 -c to handle complex logic such as Kahn's algorithm for topological sorting of worker dependencies and real-time cost estimation from token usage metrics.
  • [SAFE]: Manages a local registry of fleet metadata at ~/.claude/fleet-registry.json, facilitating easy resolution of fleet roots and management of active sessions without external data transmission.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 11:08 AM