Skip to main content
StudioMeyer
Install MCP Tools in 5 Minutes: Step-by-Step Guide
Back to Blog
AI & Automation February 24, 2026 6 min readby Matthias Meyer

Install MCP Tools in 5 Minutes: Step-by-Step Guide

From license to first tool call in under 5 minutes. Tutorial for Claude Desktop, ChatGPT, and other AI clients.

Installing MCP tools sounds like a tech project. It's not. In five minutes, you'll have your first MCP server configured and ready to use AI tools directly in Claude Desktop, ChatGPT, or any other compatible client. This guide walks you through the entire process -- from understanding to a working connection.

What Is MCP?

MCP stands for Model Context Protocol. It's an open standard introduced by Anthropic in 2024 and since adopted by all major AI providers -- OpenAI, Google, Microsoft, and many more.

The idea is simple: instead of every AI tool needing its own integration, there's one unified standard. An MCP server provides tools, an MCP client (such as Claude Desktop) uses them. Communication runs over JSON-RPC 2.0, secured with API keys.

For you as a user, this means: one format, usable everywhere. An MCP tool configured for Claude works with minimal changes in ChatGPT or other clients too.

Step 1: Choose a Product

Before configuring anything, you need an MCP tool. The StudioMeyer Store offers specialized AI tools for different business areas:

  • AI CRM -- Customer management directly from your AI chat
  • AI Shield -- Chatbot protection against injection and abuse
  • SmartBot -- AI chatbot for your website
  • AI Inbox Manager -- Email triage and automation
  • AI Documents -- Process invoices and documents via AI
  • AI Knowledge Vault -- Knowledge base with AI search (Coming Soon)

Each product comes with its own MCP endpoint and specialized tools. Choose what you need and configure only that.

Step 2: Get Your License Key

After purchase, you'll receive a license key in the format:

sm_live_YOUR_KEY

This key authenticates all your MCP requests. Keep it secure -- it's your access to the tools.

Important: The key is used as a Bearer token in your configuration. Don't share it publicly and don't commit it to Git repositories.

Step 3: Configure Your AI Client

Option A: Claude Desktop

Claude Desktop is Anthropic's native client with built-in MCP support. Configuration is done through a JSON file.

Config path:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%/Claude/claude_desktop_config.json

Add the configuration:

{
  "mcpServers": {
    "studiomeyer-crm": {
      "url": "https://mcp.studiomeyer.io/ai-crm/mcp",
      "headers": {
        "Authorization": "Bearer sm_live_YOUR_KEY"
      }
    }
  }
}

Replace sm_live_YOUR_KEY with your actual license key. The server name (here studiomeyer-crm) is freely chosen -- it only serves as identification in your client.

After saving: restart Claude Desktop. The tools will automatically appear in the tool list.

Option B: ChatGPT

ChatGPT has supported MCP since early 2026. Setup works through Custom GPTs or directly in ChatGPT Plus:

  1. Open GPT settings or create a Custom GPT
  2. Add a new "Action"
  3. Enter the URL: https://mcp.studiomeyer.io/ai-crm/mcp
  4. Select "API Key" as authentication
  5. Enter your license key

Available tools are automatically discovered and usable in chat.

Option C: Other Clients

Any MCP-compatible client works with the same endpoint. The gateway URL always follows this pattern:

https://mcp.studiomeyer.io/{product}/mcp

Authentication is always a Bearer token with your license key.

Step 4: Test the Connection

After configuration, test the connection. In Claude Desktop, a successful connection means the server's tools appear in the tool list.

Simple test: Ask your AI client:

"What tools are available from the StudioMeyer MCP server?"

The client should list the available tools. For AI CRM, that includes tools for contact management, lead tracking, and pipeline management.

If the connection doesn't work:

  1. Check your key: Is the API key correct and complete?
  2. Check the URL: Is the endpoint right? No typos?
  3. Restart the client: Some clients only load configuration at startup
  4. Firewall: Make sure outgoing HTTPS connections are allowed

Multiple MCP Servers Simultaneously

You can use multiple MCP servers in parallel. Each gets its own entry in the configuration:

{
  "mcpServers": {
    "studiomeyer-crm": {
      "url": "https://mcp.studiomeyer.io/ai-crm/mcp",
      "headers": {
        "Authorization": "Bearer sm_live_YOUR_CRM_KEY"
      }
    },
    "studiomeyer-inbox": {
      "url": "https://mcp.studiomeyer.io/ai-inbox/mcp",
      "headers": {
        "Authorization": "Bearer sm_live_YOUR_INBOX_KEY"
      }
    }
  }
}

The AI automatically knows which tool belongs to which server and selects the right one for your request.

Security

The connection to the MCP Gateway runs over HTTPS (TLS-encrypted). Your data is not stored or used for training. The API key ensures only you can access your tools.

Best practices:

  • Never put license keys in public repositories
  • Use one key per product
  • If you suspect compromise: renew the key in your dashboard

Next Steps

Installation is the easy part. Real productivity begins when you integrate the tools into your daily workflow:

  1. Start with one tool and learn its available functions
  2. Add more tools as you understand the workflow
  3. Discover all available products in the StudioMeyer Store

Five minutes of setup. Lasting productivity gains. Getting started with professional MCP tools is that simple.

Matthias Meyer

Matthias Meyer

Founder & AI Director

Founder & AI Director at StudioMeyer. Has been building websites and AI systems for 10+ years. Living on Mallorca for 15 years, running an AI-first digital studio with its own agent fleet, 680+ MCP tools and 5 SaaS products for SMBs and agencies across DACH and Spain.

mcptutorialinstallationanleitungai-tools
Install MCP Tools in 5 Minutes: Step-by-Step Guide