CreateOS MCP Server

CreateOS MCP

Deploy, manage, and scale applications directly from your AI assistant.

Introduction

The Model Context Protocol (MCP) is an open standard originated by Anthropic that enables AI assistants — Claude, Cursor, GitHub Copilot, Windsurf, Gemini, and others — to securely connect to external tools, APIs, and services through a unified interface. The latest MCP Authorization and Streamable HTTP specifications are fully implemented.

CreateOS MCP is a production-grade MCP server that exposes 85+ tools for full-stack application deployment and infrastructure management on the CreateOS platform. Connect it once to your AI coding tool, then deploy projects, manage environments, configure domains, run security scans, analyze deployment metrics, and more — all through natural language.

Instead of switching between dashboards, CLIs, and documentation, you stay in your editor and let your AI handle the infrastructure. CreateOS MCP turns prompts like "deploy my app from this GitHub repo" or "scale the staging environment to 3 replicas" into real actions, executed instantly.

Built in Go for performance and reliability, the server supports both Streamable HTTP and stdio transports, with OAuth 2.0 and API key authentication out of the box.

Features

• 🚀 85+ MCP Tools — Full coverage of the CreateOS platform API: projects, deployments, environments, domains, templates, and more
• 🔌 9 Supported Clients — Cursor, VS Code + Copilot, Claude Desktop, Claude Code, Windsurf, Gemini CLI, Gemini Code Assist, Opencode, Zapier, and ElevenLabs
• 🔐 Secure Authentication — API key and OAuth 2.0 with Dynamic Client Registration (RFC 7591), Protected Resource Metadata (RFC 9728)
• ⚡ Dual Transport — Streamable HTTP for remote access, stdio for local/embedded use
• 📦 Deploy Anything — GitHub repos, Docker images, file uploads (zip, base64, multipart), or marketplace templates
• 🌍 Environment Management — Create, configure, and monitor multiple environments per project with independent secrets and resources
• 🔒 Security Scanning — Trigger and review security scans on any deployment
• 📊 Built-in Analytics — Query request metrics, RPM, error paths, success rates, and traffic distribution per environment
• 🏪 Template Marketplace — Browse, purchase, and deploy from a library of production-ready project templates
• 🔑 API Key Management — Create, list, update, and revoke API keys programmatically
• 🌐 Custom Domains — Attach, refresh, and manage custom domains with automatic TLS
• 📋 Build & Runtime Logs — Stream build logs and runtime logs directly in your AI assistant

Prerequisites

1. Create an account at createos.nodeops.network
2. Navigate to Profile Settings and generate your API key

📺 Watch the API key retrieval tutorial

Quick Start

Add the CreateOS MCP server to your AI client using this configuration:

{
  "mcpServers": {
    "createos": {
      "url": "https://api-createos.nodeops.network/mcp",
      "type": "http",
      "headers": {
        "X-Api-Key": "CREATEOS_API_KEY"
      }
    }
  }
}

Replace CREATEOS_API_KEY with your actual key from Profile Settings.

For client-specific setup instructions, see the Integration docs.

Usage Examples

Once connected, interact with CreateOS through natural language in your AI assistant:

Deploying

"Deploy my current project to CreateOS"
"Create a new project called my-api using Node.js 20"
"Deploy the main branch of my GitHub repo naman485/my-app"
"Upload this zip file as a new deployment"

Managing Environments

"List all environments for the my-api project"
"Set the DATABASE_URL environment variable on staging"
"Scale the production environment to 500 millicores and 1024MB memory"
"Show me the runtime logs for staging"

Domains & Networking

"Add a custom domain api.example.com to my project"
"Refresh the SSL certificate for my domain"
"List all domains attac