MCP Catalogs
Home

apra-fleet

by Apra-Labs·39·Score 45

An MCP server that orchestrates AI coding agents across machines with doer-reviewer loops and multi-machine coordination.

developer-toolsai-llmops-infra
9
Forks
41
Open issues
this month
Last commit
2d ago
Indexed

Overview

Apra Fleet is a comprehensive MCP server that pairs AI coding agents into doer-reviewer loops for higher quality code and coordinates them across machines via SSH. It enables distributed computing with AI agents, allowing developers to run test suites on Linux while developing on macOS, or have multiple agents work on different parts of a project in parallel. The system includes optional Project Manager skills for structured multi-step workflows, with task state persistence across sessions.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

you:Coordinating multiple AI coding agents across different machines for parallel development
you:Implementing doer-reviewer loops for code quality assurance
you:Distributing computationally intensive tasks across machines via AI orchestration
you:What is a 'doer-reviewer loop'?
you:Can I mix different LLM providers with Apra Fleet?
you:Does Apra Fleet require any background services or daemons?

When to choose this

Choose Apra Fleet when you need distributed AI coding across multiple machines with built-in code review through the doer-reviewer pattern.

When NOT to choose this

Avoid if you're looking for a simple local coding tool or if you need more specialized AI features beyond code generation and review.

Tools this server exposes

12 tools extracted from the README
  • register_member

    Register a machine as a fleet member (local or remote via SSH)

  • update_member

    Update a member's registration (rename, change host, folder, auth, git access)

  • remove_member

    Unregister a fleet member

  • list_members

    List all registered fleet members

  • member_detail

    Deep-dive status for a single member

  • fleet_status

    Overview status of all members

  • execute_prompt

    Run an AI prompt on a member (supports session resume)

  • execute_command

    Run a shell command directly on a member (no AI CLI needed)

  • send_files

    Upload local files to a member via SFTP

  • receive_files

    Download files from a member's work folder

  • setup_ssh_key

    Generate an SSH key pair and migrate from password to key auth

  • credential_store_set

    Store a secret credential for use in commands

Comparable tools

cursorgithub-copilottabnine

Installation

Quick Start

**macOS (Apple Silicon)**

curl -fsSL https://github.com/Apra-Labs/apra-fleet/releases/latest/download/apra-fleet-installer-darwin-arm64 -o apra-fleet-installer && chmod +x apra-fleet-installer && ./apra-fleet-installer install

**Linux (x64)**

curl -fsSL https://github.com/Apra-Labs/apra-fleet/releases/latest/download/apra-fleet-installer-linux-x64 -o apra-fleet-installer && chmod +x apra-fleet-installer && ./apra-fleet-installer install

**Windows (x64)**

Invoke-WebRequest -Uri https://github.com/Apra-Labs/apra-fleet/releases/latest/download/apra-fleet-installer-win-x64.exe -OutFile apra-fleet-installer.exe; .\apra-fleet-installer.exe install

Claude Desktop Configuration

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "apra-fleet": {
      "command": "apra-fleet",
      "args": ["mcp"],
      "env": {}
    }
  }
}

FAQ

What is a 'doer-reviewer loop'?
It's a workflow where two AI agents are paired - one writes code while the other reviews it. The Project Manager orchestrates the handoff with structured checkpoints, ensuring every change gets reviewed before you even look at it.
Can I mix different LLM providers with Apra Fleet?
Yes, Fleet members can run different LLM backends. The recommended setup uses Claude for the Project Manager role (orchestrator), any provider for the Doer role, and premium tier models for the Reviewer role to catch subtle issues.
Does Apra Fleet require any background services or daemons?
No, there are no background services or daemons. The fleet server starts on-demand when Claude Code connects and stops when the connection closes.

Compare apra-fleet with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.