Magic

v1.0.0Coding Agentsstable

Magicrew. The first open-source all-in-one AI productivity platform (Generalist AI Agent + Workflow Engine + IM + Online collaborative office system)

agentagiaigptllm
Share:
4,826
Stars
0
Downloads
0
Weekly
0/5

What is Magic?

Magic is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to magicrew. the first open-source all-in-one ai productivity platform (generalist ai agent + workflow engine + im + online collaborative office system)

Magicrew. The first open-source all-in-one AI productivity platform (Generalist AI Agent + Workflow Engine + IM + Online collaborative office system)

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Magicrew. The first open-source all-in-one AI productivity p

Use Cases

All-in-one AI productivity platform with agent and workflow engine
Collaborative office system powered by AI
dtyq

Maintainer

LicenseNOASSERTION
Languagetypescript
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @mastergo/magic-mcp

Manual Installation

npx -y @mastergo/magic-mcp

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Magic

Magic (Magicrew) is an open-source, all-in-one AI productivity platform that combines a generalist AI agent, a workflow automation engine, instant messaging, and an online collaborative office system in a single self-hosted deployment. It lets teams wrap internal systems like ERP, CRM, and databases as AI-powered digital employees that can execute tasks, generate documents, and gate high-risk actions through approval workflows. The @mastergo/magic-mcp package exposes Magic's agent and workflow capabilities directly to MCP-compatible AI clients.

Prerequisites

  • Docker and Docker Compose installed on macOS or Linux
  • Node.js 18+ and npx for running the MCP package
  • An MCP-compatible client such as Claude Desktop or Cursor
  • curl available for the automated installer script
1

Deploy the Magic platform locally

Run the automated installer script to set up the full Magicrew stack including the cluster, infrastructure, and all services. Docker must be running before executing this command.

curl -fsSL https://getmagicrew.sh | bash
2

Access the Magic web interface

Once installation completes, open your browser to the local web UI to configure agents, workflows, and integrations before connecting via MCP.

open http://localhost:38080
3

Obtain your Magic API credentials

In the Magic web UI, navigate to Settings and create an API key or obtain your workspace credentials. These will be needed when configuring the MCP server.

4

Add the Magic MCP server to your client config

Configure your MCP client (e.g. Claude Desktop) to launch the @mastergo/magic-mcp package using npx. The package connects to your local Magic instance and exposes its tools.

{
  "mcpServers": {
    "magic": {
      "command": "npx",
      "args": ["-y", "@mastergo/magic-mcp"]
    }
  }
}
5

Restart your MCP client

Restart Claude Desktop or your preferred MCP client to load the new Magic server configuration. The server will appear in the tools list once connected.

6

Teardown when done

To remove the local Magicrew deployment and free resources, use the teardown command.

magicrew teardown

Magic Examples

Client configuration

Minimal Claude Desktop configuration for the Magic MCP server using the npm package.

{
  "mcpServers": {
    "magic": {
      "command": "npx",
      "args": ["-y", "@mastergo/magic-mcp"]
    }
  }
}

Prompts to try

Interact with Magic's AI agents and workflow engine through your MCP client.

- "Create a new workflow that sends a Slack message when a CRM deal is closed"
- "Generate a monthly sales report and export it as an Excel file"
- "List all active AI agents in my Magic workspace"
- "Trigger the onboarding workflow for a new employee named Alex"

Troubleshooting Magic

The installer script fails with a Docker connectivity error

Ensure Docker Desktop is running and the Docker daemon is accessible. On Linux, verify your user is in the docker group: sudo usermod -aG docker $USER, then log out and back in.

MCP client cannot connect to the Magic server after config is added

Confirm the Magic platform is running at http://localhost:38080. If the port is in use, check for port conflicts with docker ps and ensure the Magicrew containers are healthy.

npx @mastergo/magic-mcp exits immediately without connecting

Check that Node.js 18+ is installed and npx is available. Run npx -y @mastergo/magic-mcp --help to see if the binary loads correctly, and check the Magic platform logs at localhost:38080.

Frequently Asked Questions about Magic

What is Magic?

Magic is a Model Context Protocol (MCP) server that magicrew. the first open-source all-in-one ai productivity platform (generalist ai agent + workflow engine + im + online collaborative office system) It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Magic?

Install via npm with the command: npx -y @mastergo/magic-mcp. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with Magic?

Magic works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Magic free to use?

Yes, Magic is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "magic": { "command": "npx", "args": ["-y", "@mastergo/magic-mcp"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Magic?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides