ai-fleet

🥈Silver

AI agent fleet manager for parallel agentic systems like claude code and codex. Operations teams use it to spin up and command multiple AI developer agents from the terminal. Connects to tmux for managing agent workflows.

6840Updated 1mo ago

Intermediate30min to implementautomation

Saves ~240 min per use

Quick InstallView Source

git clone https://github.com/nachoal/ai-fleet.git

Works with:

Claude

Overview

About This Skill

ai-fleet is a project-based fleet management tool that lets you create, orchestrate, and monitor multiple AI developer agents running in parallel from the command line. Each agent operates in an isolated git worktree with its own tmux session, enabling true parallel execution of AI-driven development tasks. The tool supports fanout operations to launch N agents with the same prompt, multi-agent workflows with different tasks, and batch management with real-time CPU/memory monitoring. Teams can initialize fleet management per project with `.aifleet/config.toml`, automatically copy credential files, run custom setup commands, and send interactive prompts to running agents. Operations teams and developers use it to parallelize code fixes, testing, documentation, and refactoring across multiple AI agents simultaneously.

How to Use

Install via pipx, set your ANTHROPIC_API_KEY, run `fleet init` in your git project, then create agents with `fleet create` or launch multiple agents with `fleet fanout` or `fleet multi`. Monitor status with `fleet list`, send prompts with `fleet prompt`, and attach to live output with `fleet attach`.

Use Cases

Launch 5+ agents in parallel to refactor different modules simultaneously

Run independent agents on authentication, testing, and documentation tasks concurrently

Scale code review and bug-fix operations across multiple isolated agent worktrees

Monitor and manage multiple AI development tasks with real-time CPU/RAM metrics

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/nachoal/ai-fleet

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

I need to create a fleet of AI developer agents to work on [PROJECT]. Each agent should focus on a different task: [TASK_1], [TASK_2], and [TASK_3]. Use AI-fleet to spin up 3 agents, assign tasks, and manage their workflows in tmux. Provide the terminal commands and expected outputs.

Example Output

```markdown
# AI Fleet Management for Web Application Development

## Fleet Setup
- **Agent 1**: Focused on backend development (Node.js)
- **Agent 2**: Handling frontend development (React)
- **Agent 3**: Database design and optimization

## Terminal Commands
```bash
# Spin up the fleet
fleet init --agents 3 --tasks "backend, frontend, database"

# Assign tasks to each agent
fleet assign --agent 1 --task backend
fleet assign --agent 2 --task frontend
fleet assign --agent 3 --task database

# Start the workflow in tmux
fleet start --tmux
```

## Expected Outputs
- **Agent 1**: Backend API structure and initial code
- **Agent 2**: React component structure and initial UI
- **Agent 3**: Database schema and optimization recommendations

## Monitoring
Use `fleet status` to check the progress of each agent.
```