---
title: "Configuration Examples"
slug: "aia-mcp-configuration-examples"
description: "Connect your AI agent to Kentik using Claude Desktop or custom MCP-compatible clients. Follow our guide for seamless integration and configuration."
tags: ["Kentik API", "Claude Desktop", "MCP Server"]
status: "update"
updated: 2026-11-10T03:42:57Z
published: 2026-11-10T03:42:57Z
canonical: "kb.kentik.com/aia-mcp-configuration-examples"
---

> ## Documentation Index
> Fetch the complete documentation index at: https://kb.kentik.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Configuration Examples

This article provides configuration examples for connecting your AI agent to the [**Kentik AI Advisor MCP Server**](/v1/docs/ai-advisor-mcp-server) using OAuth or API token authentication.

Choose your preferred environment below to get connected:

## **Authentication Methods**

| Method | Description |
| --- | --- |
| **OAuth** | Redirects you to the Kentik portal to authenticate, so no API token is stored in your agent configuration. You can control access to the MCP server from your identity provider (IdP). Requires a Kentik portal account. |
| **API token** | Uses the API email and token of a single Kentik user. |

> [!NOTE]
> **Note**: Integrations that use an API token are bound to the user who owns that token. For integrations shared across your organization, use OAuth.

## **Server URL**

Use `https://api.kentik.com/mcp`. If your Kentik account is in the EU, use `https://api.kentik.eu/mcp`.

## **OAuth Configuration**

### **Claude Code**

1. In your project directory, run the following command:

```bash
claude mcp add ai-advisor --transport http https://api.kentik.com/mcp
```
2. Launch Claude Code and run `/mcp`. The AI Advisor MCP server appears as configured but unauthenticated.
3. Follow the prompts. A browser tab opens to the Kentik portal login page.
4. Log in. When the connection succeeds, `/mcp` shows a green check next to the server.

### **Claude Desktop**

1. In the prompt box, click **+** and select **Add Connector** → **Add custom connector**.
2. Enter the server URL, then follow the prompts. A browser tab opens to the Kentik portal login page.
3. Authenticate when prompted. The connection is complete when the browser tab confirms authentication.

### **Cursor**

1. In the prompt box, click **+** and select **MCP** → **Manage**, then select **New MCP Server**. The `~/.cursor/mcp.json` file opens.
2. Add the following to the file, then save it:

```json
{
  "mcpServers": {
    "Kentik AI Advisor MCP": {
      "url": "https://api.kentik.com/mcp"
    }
  }
}
```
3. Click **Authenticate** for the new entry and follow the instructions in the browser page until the server shows as connected.

## **API Token Configuration**

The following examples authenticate with the API email and token of a single Kentik user.

### **Claude Desktop**

Claude Desktop supports Model Context Protocol (MCP) servers on both free and paid plans through manual JSON configuration.

#### **Configuration File Location**

- **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
- **Windows**: `%APPDATA%\\Claude\\claude_desktop_config.json`

**Quick Access**: Claude Desktop menu → **Settings** → **Developer** → **Edit Config**

#### **Configuration**

Add the following to your `claude_desktop_config.json`:

```json
{
  "mcpServers": {
    "ai-advisor": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://api.kentik.com/mcp",
        "--header",
        "X-CH-Auth-Email: ${KENTIK_API_EMAIL}",
        "--header",
        "X-CH-Auth-API-Token: ${KENTIK_API_TOKEN}"
      ],
      "env": {
        "KENTIK_API_EMAIL": "your-email-here",
        "KENTIK_API_TOKEN": "your-api-token-here"
      }
    }
  }
}
```

**After Configuration**:

1. Restart Claude Desktop.
2. Open **Settings** to verify the MCP server is connected.
3. Start a conversation with AI Advisor. Remember to mention "Kentik" in your prompts.

### **GitHub Copilot in VS Code**

For [**VS Code with GitHub Copilot**](https://code.visualstudio.com/docs/agent-customization/mcp-servers) or other MCP extensions:

```json
{
  "servers": {
    "ai-advisor": {
      "url": "https://api.kentik.com/mcp",
      "type": "http",
      "headers": {
        "X-CH-Auth-Email": "your-email-here",
        "X-CH-Auth-API-Token": "your-api-token-here"
      }
    }
  },
  "inputs": []
}
```

## **Custom Applications**

For custom integrations, use any MCP-compatible client library. The server supports:

- **Protocol**: MCP over HTTP (Streamable HTTP transport)
- **MCP protocol versions**: 2026-07-28 and 2025-11-25.

> [!NOTE]
> **Note**: Clients running unsupported MCP protocol versions may encounter request timeouts while trying to connect to the AI Advisor MCP Server.
- **Port**: Standard HTTPS (443)
- **Format**: JSON-RPC 2.0 (see [**MCP Payload Reference**](/v1/docs/aia-mcp-payload))
- **Tools**: See [**MCP Tools Reference**](/v1/docs/aia-mcp-tools)

Having an issue? See our [**Troubleshooting**](/v1/docs/aia-mcp-troubleshooting) guide for connecting to the AI Advisor MCP Server.

## Related

- [AI Advisor MCP Server](/ai-advisor-mcp-server.md)