Claude Code tmux Runner

Overview

Run multiple Claude Code tasks in parallel using tmux sessions. Non-blocking, background execution with status monitoring.

Usage

Start a Task

~/.openclaw/scripts/claude-tmux.sh start "Your task description" [task-name]

Examples:

~/.openclaw/scripts/claude-tmux.sh start "Write a Python auth module" auth
~/.openclaw/scripts/claude-tmux.sh start "Build a React component" ui
~/.openclaw/scripts/claude-tmux.sh start "Create a Go microservice" api

List All Tasks

~/.openclaw/scripts/claude-tmux.sh list

Check Task Status

~/.openclaw/scripts/claude-tmux.sh status <task-id>

View Generated Files

~/.openclaw/scripts/claude-tmux.sh files

Stop Tasks

~/.openclaw/scripts/claude-tmux.sh stop <task-id>
~/.openclaw/scripts/claude-tmux.sh stop-all

Workflow

• User says "use Claude Code xxx" → Optimize prompt → Show for approval
• User confirms → Start background task → Return immediately
• User can continue chatting → Non-blocking
• User asks "how is Claude doing" → Check status → Return progress

Industrial Code Standards

Generated code must include:

• Complete error handling (try-catch, boundary checks)
• Type annotations (TypeScript, Python type hints)
• Documentation (JSDoc, docstring)
• Unit tests (pytest, jest)
• Performance considerations (time/space complexity)
• Security best practices (input validation)
• Maintainability (clear naming, modular)

File Locations

• Script: ~/.openclaw/scripts/claude-tmux.sh
• Logs: ~/.openclaw/logs/claude/
• Files: ~/.openclaw/workspace/
• State: ~/.openclaw/state/claude/