5-Minute Quickstart

Get your first MCP Gateway running and connect it to Claude Desktop or your application in just 5 minutes.

Prerequisites

• Noorle Account: Sign up free at noorle.com

Step 1: Create Your Gateway

Using the Console

1. Go to noorle.com
2. Click on MCP Gateways and then click Create
3. Name your MCP Gateway (e.g., my-gateway) and set your desired configurations
4. Activate built-in tools you want your MCP Gateway to have access to and if you have deployed any plugins
5. Copy your new gateway address and pass it to your AI agents

Step 2: Connect Your Application

Option A: Agentic Applications

Connect your Noorle gateway to AI applications that support MCP protocol.

Quick Setup Examples:

Claude Code (docs):

claude mcp add --transport http noorle https://mcp-abcd.noorle.com \
  --header "Authorization: Bearer YOUR_API_KEY"

OpenAI/ChatGPT (docs):

{
  "mcp_servers": {
    "noorle": {
      "url": "https://mcp-abcd.noorle.com",
      "api_key": "YOUR_API_KEY"
    }
  }
}

Learn More:

• Claude Code MCP Guide - Complete Claude integration
• OpenAI MCP Documentation - ChatGPT & GPTs integration
• Connecting Remote MCP Servers - Protocol details
• Cloudflare Remote MCP Guide - Testing remote servers

Option B: Agent Frameworks

Integrate your Noorle gateway with popular agent frameworks.

Quick Setup Examples:

LangChain (docs):

from langchain_mcp_adapters.client import MultiServerMCPClient

# Connect to Noorle gateway
client = MultiServerMCPClient({
    "noorle": {
        "transport": "streamable_http",
        "url": "https://mcp-abcd.noorle.com",
        "headers": {"Authorization": "Bearer YOUR_API_KEY"}
    }
})

tools = await client.get_tools()

CrewAI (docs):

from crewai import Agent, Task, Crew
from crewai.mcp import MCPTool

# Connect to Noorle gateway
mcp_tool = MCPTool(
    server_url="https://mcp-abcd.noorle.com",
    api_key="YOUR_API_KEY"
)

agent = Agent(
    role="AI Assistant",
    tools=[mcp_tool]
)

Learn More:

• LangChain MCP Integration - Complete LangChain guide
• CrewAI MCP Overview - Multi-agent orchestration
• MCP SDK Documentation - Build custom integrations

Step 3: Test Your Connection

Testing Tools

MCP Inspector - Visual debugging and testing tool:

npx @modelcontextprotocol/inspector

Then connect to your MCP Gateway:

• Transport: Streamable HTTP
• URL: https://mcp-abcd.noorle.com
• Headers: Authorization: Bearer YOUR_API_KEY

cURL - Quick command-line test:

curl -X POST https://mcp-abcd.noorle.com \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/list","id":1}'

Testing Resources

• MCP Inspector Guide - Complete debugging tutorial
• Testing Remote MCP Servers - Cloudflare's testing guide
• AWS Gateway Inspector - Enterprise testing

Next Steps

Explore Platform Features

• Built-in Capabilities - Explore all available tools
• Platform Configuration - Configure gateways, teams, and security
• Model Routing - AI model integration

Build & Extend

• Build Plugins - Create custom capabilities
• Quick Start Guide - Build your first plugin
• CLI Reference - Command line interface

Get Support

• Documentation: https://noorle.com/docs/
• Support: [email protected]

---

🎉 Congratulations! You've successfully set up your first MCP Gateway. Your AI agents now have access to powerful capabilities through a secure, unified interface.