coding-agent

Orchestrate Codex CLI and Claude Code via tmux for parallel development tasks.

1|Updated Feb 6, 2026
One-click install
npx skills add https://github.com/rarestg/rarestg-skills --skill coding-agent-rarestg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/rarestg/rarestg-skills/tree/main/skills/coding-agent
Command: npx skills add https://github.com/rarestg/rarestg-skills --skill coding-agent-rarestg

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill streamlines complex coding tasks by orchestrating multiple AI coding assistants, enabling parallel work, research delegation, and end-to-end feature development.

Core Features & Use Cases

  • Parallel Task Execution: Delegate tasks to Codex CLI and Claude Code simultaneously for faster results.
  • Phased Development: Manage complex projects through distinct research, planning, implementation, and review phases, each handled by specialized AI agents.
  • Issue Fixing: Utilize git worktrees and AI agents to concurrently fix multiple issues.
  • Use Case: Build a new feature by first having an agent research existing code, then another plan the implementation, followed by parallel coding agents for each phase, and finally a review agent.

Quick Start

Use the coding-agent skill to research how authentication tokens flow through the codebase.

Frequently Asked Questions about coding-agent

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

FAQPage Schema
How do I run multiple AI coding assistants in parallel for software development?

You can run parallel AI coding assistants by orchestrating Codex CLI and Claude Code as background processes via tmux. This setup manages concurrent execution, allowing you to delegate complex software tasks to multiple agents simultaneously for faster results.

Can I use git worktrees to fix multiple issues concurrently with an AI agent?

Yes, you can use git worktrees to fix multiple issues concurrently. The system orchestrates AI agents across separate worktrees, enabling concurrent issue resolution and phased implementation without branch conflicts during your development workflow.

What is the best way to manage phased implementation pipelines using AI coding assistants?

The best way to manage phased implementation pipelines is by delegating distinct research, planning, coding, and review phases to specialized AI agents. This orchestrates end-to-end feature development by routing each phase to the appropriate assistant.

Does this parallel coding workflow require tmux to manage background AI agents?

Yes, tmux is required to manage background AI agents. The system orchestrates Codex CLI and Claude Code through tmux sessions, enabling detailed session management, progress monitoring, and multi-turn conversations with the coding assistants.

How do I monitor progress and manage sessions for background coding agents?

You monitor progress and manage sessions for background coding agents through the tmux orchestration layer. This approach provides detailed session management and progress monitoring capabilities for multi-turn conversations with your AI coding assistants.