Skip to main content

​
Overview

The Model Context Protocol (MCP) is an open standard that enables Large Language Models (LLMs) to interact with external systems. RelytONE provides a local MCP server that runs on your machine, allowing AI assistants like Cursor and Claude to manage your RelytONE resources through natural language.
Use with Caution: This MCP server grants LLMs direct access to manipulate your RelytONE resources, including creating, modifying, and deleting projects and databases. We strongly recommend NOT using it in production environments. Only use it in development or testing scenarios where accidental changes won’t impact critical systems.

​
Prerequisites

Before setting up the MCP server, ensure you have:
  1. Node.js version 18.0.0 or higher installed. Check your version with: 
    node --version
  2. RelytONE API Token: Generate an API key from the RelytONE console under Settings → API Keys. You’ll need this token for authentication.

​
Setup

The MCP server runs locally on your machine using npx.
  • Cursor
  • Claude Code

Installation Options

Option 1: One-Click InstallInstall MCP ServerOption 2: Manual ConfigurationNavigate to Preferences → Cursor Settings → MCP → New MCP Server and add:
{
  "mcpServers": {
    "relytone": {
      "command": "npx",
      "args": ["--yes", "@relyt/mcp-server-relytone"],
      "env": {
        "RELYTONE_BEARER_TOKEN": "your-api-token-here"
      }
    }
  }
}

Verify Connection

Once configured successfully, you should see the RelytONE MCP server listed as connected:RelytONE MCP Server Connection Status in Cursor

​
Available Tools

All RelytONE APIs are exposed as MCP tools. The tool names follow the format <method>-<path>, such as:
  • get-organizations - List organizations
  • get-projects - List projects
  • post-projects - Create a project
  • get-projects-slug-databases - List databases in a project
  • post-projects-slug-databases - Create a database
You don’t need to know the exact tool names—simply describe what you want in natural language, and the AI will use the appropriate tools.

​
Usage Examples

Simply describe what you want in natural language. Here are some examples: List projects in an organization: 
Please list all projects in the "my-org" organization
Create a new project: 
Create a project under "my-org" with name "analytics-dev" 
and description "Development environment for analytics"
Create a database: 
Create a database named "customers" in the "analytics-dev" project
Get project details: 
Show me the details and compute resources for project "analytics-dev"
⌘I