clawteam

Orchestrate autonomous multi-agent teams across Claude Code, Codex, OpenClaw, and Cursor.

40|9|Updated Jan 4, 2026
One-click install
npx skills add https://github.com/akillness/oh-my-skills --skill clawteam
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: clawteam
Source: https://github.com/akillness/oh-my-skills/tree/main/.agent-skills/clawteam
Command: npx skills add https://github.com/akillness/oh-my-skills --skill clawteam

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires clawteam, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Orchestrates a fleet of autonomous agents to run parallel tasks across diverse tooling and environments, simplifying complex coordination and reducing manual overhead.

Core Features & Use Cases

  • Leader-driven orchestration with per-agent workspaces via git worktrees and tmux-backed sessions.
  • Supports multiple agent types (Claude Code, Codex, OpenClaw, nanobot, Cursor, and custom scripts) to optimize task execution.
  • Dependency-aware task planning and real-time monitoring through an isolated, self-contained workflow.

Quick Start

Spawn a leader agent to orchestrate a team of workers and monitor progress using clawteam.

Frequently Asked Questions about clawteam

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I orchestrate multiple autonomous agents to run parallel tasks across different coding tools?

Multi-agent orchestration coordinates a leader agent that spawns workers into isolated git worktrees, enabling parallel tasks across Claude Code, Codex, OpenClaw, Cursor, and custom scripts with dependency-driven planning and real-time monitoring.

Can I use tmux and git worktrees to isolate workspaces for multiple AI agents?

Yes, tmux-backed sessions and git worktrees provide per-agent workspace isolation, ensuring each autonomous worker operates in a separate environment without conflicting with parallel tasks or shared repository states.

What's the best way to manage task dependencies in a multi-agent coding pipeline?

Dependency-aware task planning enforces execution order by having the leader agent coordinate workers, ensuring downstream tasks wait for prerequisite completion before triggering new agent work in isolated workspaces.

Does clawteam support custom agent scripts alongside Claude Code and Codex?

clawteam supports multiple agent types including Claude Code, Codex, OpenClaw, nanobot, Cursor, and custom scripts, allowing you to optimize task execution by matching agent capabilities to specific pipeline requirements.

How do I monitor real-time progress of parallel agents working in separate git worktrees?

Real-time monitoring tracks each agent's progress through per-agent inboxes and a self-contained workflow, providing visibility into worker status across complex experiments, API development, or research pipelines.

When should I avoid using multi-agent orchestration for my development workflow?

Multi-agent orchestration adds coordination overhead suited for complex parallel tasks and research pipelines; simpler sequential development or single-agent workflows may not benefit from leader-driven worker spawning and dependency management.