Agent Management
- 291 installs
- 30.1k repo stars
- Updated August 4, 2026
- davila7/claude-code-templates
agent-management is a Claude Code skill that manages the full AI agent lifecycle via the AI Maestro CLI for developers who need to create, hibernate, wake, and orchestrate multiple tmux-based coding agents.
About
agent-management is part of the AI Maestro skill suite in davila7/claude-code-templates, providing CLI-driven lifecycle control for multiple AI coding agents. Through aimaestro-agent.sh commands, developers create agents with project directories and task tags, list and inspect agent status, hibernate agents to save resources, wake them with optional tmux attach, rename, export and import configurations, and install or uninstall plugins per agent. Agent statuses track online, offline, and hibernated states across tmux sessions. The skill requires AI Maestro running locally with tmux 3.0 plus and sits alongside five sibling skills for messaging, memory search, docs search, graph queries, and planning within the six-skill AI Maestro orchestration platform. Reach for agent-management when running parallel backend and frontend agents, pausing overnight work without losing state, or installing specialized plugins on individual agent instances.
- role-based subagents
- handoff contracts
- tool and context limits
- supervision loops
- failure recovery
Agent Management by the numbers
- 291 all-time installs (skills.sh)
- Ranked #2,302 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/davila7/claude-code-templates --skill agent-managementAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 291 |
|---|---|
| repo stars | ★ 30.1k |
| Last updated | August 4, 2026 |
| Repository | davila7/claude-code-templates ↗ |
How do you manage multiple Claude agents with tmux?
Orchestrate multiple Claude agents with roles, handoffs, and guardrails when building autonomous coding or research workflows inside Claude Code.
Who is it for?
Developers running multiple parallel Claude Code agents via AI Maestro who need create, hibernate, wake, and plugin lifecycle management in tmux sessions.
Skip if: Single-agent Claude Code sessions without AI Maestro installed, or cloud-only agent platforms that do not use local tmux orchestration.
When should I use this skill?
A user asks to create an agent, list agents, hibernate or wake an agent, install a plugin on an agent, or manage AI Maestro agent lifecycle.
What you get
Running tmux agent sessions with task assignments, hibernated state snapshots, exported agent JSON configs, and per-agent plugin installations.
- running tmux agent sessions
- exported agent JSON config
By the numbers
- Part of the AI Maestro 6-skill orchestration suite
- Requires tmux 3.0+ for agent session management
Files
AI Maestro Agent Management
Create, manage, and orchestrate multiple AI agents through a unified CLI. Handles the full agent lifecycle: create, hibernate, wake, rename, export/import, and plugin management. Part of the AI Maestro suite.
Prerequisites
Requires AI Maestro running locally with tmux 3.0+.
# Install the CLI
git clone https://github.com/23blocks-OS/ai-maestro-plugins.git
cd ai-maestro-plugins && ./install-agent-cli.shCore Commands
Agent Lifecycle
| Command | Description |
|---|---|
aimaestro-agent.sh list | List all agents with status |
aimaestro-agent.sh show <agent> | Detailed agent information |
aimaestro-agent.sh create <name> --dir <path> | Create new agent |
aimaestro-agent.sh update <agent> --task "..." | Update task/tags |
aimaestro-agent.sh delete <agent> --confirm | Delete agent |
aimaestro-agent.sh rename <old> <new> | Rename agent |
aimaestro-agent.sh hibernate <agent> | Save state, free resources |
aimaestro-agent.sh wake <agent> | Resume hibernated agent |
aimaestro-agent.sh restart <agent> | Hibernate then wake |
Plugin Management
| Command | Description |
|---|---|
aimaestro-agent.sh plugin install <agent> <plugin> | Install plugin |
aimaestro-agent.sh plugin uninstall <agent> <plugin> | Remove plugin |
aimaestro-agent.sh plugin list <agent> | List installed plugins |
aimaestro-agent.sh plugin marketplace add <agent> <source> | Add marketplace |
Export/Import
| Command | Description |
|---|---|
aimaestro-agent.sh export <agent> | Export agent config |
aimaestro-agent.sh import <file> | Import agent from file |
Usage Examples
# Create a backend API agent
aimaestro-agent.sh create backend-api \
--dir ~/projects/backend \
--task "Build REST API with TypeScript" \
--tags "api,typescript"
# End of day -- save resources
aimaestro-agent.sh hibernate frontend-ui
aimaestro-agent.sh hibernate data-processor
# Resume next morning
aimaestro-agent.sh wake frontend-ui --attach
# Install a plugin on an agent
aimaestro-agent.sh plugin install backend-api my-plugin
# Backup before risky changes
aimaestro-agent.sh export backend-api -o backup.jsonAgent Statuses
| Status | Meaning |
|---|---|
online | Running in tmux session |
offline | Registered but no active session |
hibernated | Saved state, session killed |
Full AI Maestro Experience
This skill is part of the AI Maestro platform, which provides 6 skills for AI agent orchestration: messaging, memory, docs, graph, planning, and agent management.
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10How it compares
Pick agent-management over generic agent prompts when you need persistent tmux-based agent lifecycle with hibernate, wake, and per-agent plugin control via AI Maestro CLI.
FAQ
What CLI does agent-management use?
agent-management uses the AI Maestro aimaestro-agent.sh CLI for all lifecycle operations. Commands include list, create, show, hibernate, wake, restart, delete, rename, export, import, and plugin install or uninstall per agent.
What are AI Maestro agent statuses?
agent-management tracks three agent statuses: online means running in an active tmux session, offline means registered without a session, and hibernated means state is saved and the tmux session was killed to free resources.
What prerequisites does agent-management need?
agent-management requires AI Maestro running locally with tmux 3.0 or newer. Install the CLI by cloning ai-maestro-plugins and running install-agent-cli.sh before creating agents with project directories and task tags.