longtho638-jpg

Openclaw Worker — AI skill for Claude Code

AI community

🦞 Autonomous AI Swarm Engine — Drop a file, watch 3 AI agents execute in parallel.

How to install Openclaw Worker

This entry records only its repository, not the path inside it, so there is no exact command to give. Open longtho638-jpg/openclaw-worker and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Openclaw Worker does

🦞 Autonomous AI Swarm Engine — Drop a file, watch 3 AI agents execute in parallel. Built on Claude Code CLI + tmux.

Alternatives in AI

Open Multi Agent — TypeScript multi-agent orchestration engine — one runTeam() call from goal to result 5.8k ★
Golutra — Multi-agent AI orchestration platform for automation, workflows, and developer tools 3.8k ★
Nodeterm — Node-based terminal manager for AI coding agents — tmux-backed terminals and parallel agent sessions as dragga 1.3k ★

README

🦞 OpenClaw Worker — Autonomous AI Swarm Engine

**Drop a file. Watch AI agents execute.** No manual intervention needed.

OpenClaw Worker turns Claude Code CLI into an autonomous swarm — 3 AI workers running in parallel, processing tasks from a simple file-drop queue. Built for developers who want AI agents that **just work**.

⚡ 30-Second Quick Start

# Prerequisites: Node.js 18+, tmux, Claude CLI
brew install tmux                              # macOS
# sudo apt install tmux                        # Linux

npm install -g @anthropic-ai/claude-code       # Claude CLI

# Clone & Start
git clone https://github.com/longtho638-jpg/openclaw-worker.git
cd openclaw-worker && npm install

# Launch the swarm (4 tmux panes inside your terminal)
bash restore_swarm.sh

You'll see 4 panes:

┌──────────────────┬──────────────────┐
│ P0: Mission Log  │ P1: AI Worker 1  │
│ (real-time feed) │ (Claude CLI TUI) │
├──────────────────┼──────────────────┤
│ P2: AI Worker 2  │ P3: AI Worker 3  │
│ (Claude CLI TUI) │ (Claude CLI TUI) │
└──────────────────┴──────────────────┘

🚀 How It Works

**Drop a mission file → AI workers auto-execute → Results logged.**

# Create a task (that's it!)
echo "Refactor the auth module to use JWT tokens" > tasks/mission_myapp_auto_auth_refactor.txt

The task-watcher daemon:

Detects new files in tasks/ directory
Dispatches to the next available worker (round-robin)
Monitors execution (busy/idle/timeout detection)
Archives completed missions to tasks/processed/

📁 Architecture

openclaw-worker/
├── task-watcher.js              # 🧠 Main orchestrator
├── config.js                    # ⚙️ Configuration
├── restore_swarm.sh             # 🦞 One-command launcher
├── lib/
│   ├── brain-tmux.js            # Tmux interactive mode (default)
│   ├── brain-headless-per-mission.js  # Headless claude -p mode
│   ├── brain-vscode-terminal.js # VS Code terminal mode
│   ├── task-queue.js