Claude Army

Provides an MCP server to orchestrate multiple Claude Code agents across projects for parallel autonomous work.

javascript

1

GitHub Stars

javascript

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "robbhimself-1337-claude-army": {
      "command": "node",
      "args": [
        "/path/to/claude-army/src/index.js"
      ],
      "env": {
        "CLAUDE_BINARY": "/usr/bin/claude-code"
      }
    }
  }
}

ClaudeArmy is an MCP server that lets Claude orchestrate multiple Claude Code agents across different projects. You interact through Claude, directing high-level goals while background agents autonomously execute tasks, enabling parallel work and ongoing progress checks without leaving the chat.

How to use

You use ClaudeArmy by connecting it to your Claude chat client as an MCP server. Once connected, you can instruct Claude to dispatch agents to work on multiple projects, monitor progress, and retrieve outputs. Treat ClaudeArmy as a scalable, autonomous workforce that runs tasks in parallel while you stay in the conversation.

How to install

Prerequisites you need before installing ClaudeArmy:

Node.js 18+ is required to run the MCP server. If you do not have Node.js installed, install it from the official source for your platform.
Git is used to clone the repository. Ensure it is available in your command line environment.
You need a client that can communicate with MCP servers to connect ClaudeArmy.

Install steps you should follow exactly as written here to set up ClaudeArmy locally.

1. Clone this repository
2. npm install
3. Add to your Claude Desktop config (example path shown; adjust to your environment):

```json
"claude-army": {
  "command": "node",
  "args": ["/path/to/claude-army/src/index.js"]
}

Restart Claude Desktop


## Available tools

### dispatch\_task

Deploy a Claude Code agent to a project directory and start work on a task.

### check\_tasks

Monitor status of all running and completed agents.

### get\_task\_output

Retrieve an agent's full output and progress details.

### cancel\_task

Stop a running agent gracefully.

### purge\_tasks

Clean up completed or failed tasks from the registry.