ntm

Manage and orchestrate AI coding agents within tmux sessions via a Go-based CLI and TUI.

419|61|Updated Dec 10, 2025
One-click install
npx skills add https://github.com/Dicklesworthstone/ntm --skill ntm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ntm
Source: https://github.com/Dicklesworthstone/ntm/tree/main
Command: npx skills add https://github.com/Dicklesworthstone/ntm --skill ntm

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill streamlines the management of multiple AI coding agents by providing a unified interface for spawning, monitoring, and coordinating them within tmux sessions, eliminating window chaos and context switching.

Core Features & Use Cases

  • Multi-Agent Orchestration: Spawn and manage multiple AI agents (Claude, Codex, Gemini) simultaneously in a single tmux session.
  • TUI & Command Palette: Offers an interactive terminal UI and a command palette for efficient session management and agent interaction.
  • Robot Mode: Provides a machine-readable JSON API for programmatic control and integration with other AI tools.
  • Use Case: Coordinate a team of AI agents for a complex coding task: use Claude for architecture, Codex for implementation, and Gemini for documentation, all managed seamlessly within NTM.

Quick Start

Use the ntm skill to spawn a new session named 'myproject' with two Claude agents and one Codex agent.

Frequently Asked Questions about ntm

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

FAQPage Schema
How do I manage multiple AI coding agents in tmux without losing track of sessions?

You can manage multiple AI agents in tmux by using a CLI and TUI to spawn, monitor, and send prompts to different models within a single terminal session. This eliminates window chaos and context switching by providing a unified orchestration interface.

Can I programmatically control AI agents in the terminal for automated workflows?

Yes, you can programmatically control AI agents using a robot mode that exposes a machine-readable JSON API. This allows you to automate agent lifecycles and integrate orchestration with other AI tools for advanced workflows.

How do I coordinate different AI models like Claude and Codex for a single coding project?

You can coordinate different AI models by spawning them simultaneously in a tmux session. Use Claude for architecture, Codex for implementation, and Gemini for documentation, managing all agents seamlessly through an interactive terminal UI and command palette.

What is the best way to orchestrate AI agents for complex coding tasks in a command line environment?

The best way to orchestrate AI agents in a command line environment is using a Go-based CLI that manages agent lifecycles within tmux. It provides an interactive terminal UI and integrates with tools like Agent Mail, Beads, and UBS for advanced coordination.

Do I need tmux installed to orchestrate AI coding agents with this tool?

Yes, tmux is required as the tool manages and orchestrates multiple AI coding agents specifically within tmux sessions. The entire architecture relies on tmux to spawn, monitor, and coordinate agent lifecycles without window chaos.