Manages and monitors parallel Claude Code agents executing task-based workloads with dependency management.
The Swarm Orchestrator skill enables seamless coordination of multiple Claude Code agents working in parallel on 'beads' (tasks with complex dependencies). It provides a robust set of commands to start, stop, and monitor execution via a centralized HTTP API, bypassing the need for complex tmux configurations. This skill is essential for developers looking to scale their AI-assisted coding workflows, offering real-time status updates, escalation management for human intervention, and automated health checks to ensure a stable multi-agent environment.
주요 기능
010 GitHub stars
02Real-time monitoring and status reporting
03Graceful and forced swarm termination
04Automated health checks for CLI and Git environments
05Human-in-the-loop escalation management
06Parallel agent coordination for dependent tasks
사용 사례
01Monitoring and resolving conflicts in high-volume AI coding sessions
02Scaling complex code migrations across multiple parallel agents
03Orchestrating multi-step development sprints with task dependencies
