---
title: "AI Advisor MCP Server"
slug: "ai-advisor-mcp-server"
description: "Integrate Kentik’s AI Advisor MCP Server for seamless network data queries, enabling natural language interactions and multi-turn conversations with AI agents."
updated: 2026-03-06T14:19:25Z
published: 2026-03-06T14:20:37Z
canonical: "kb.kentik.com/ai-advisor-mcp-server"
---

> ## 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.

# AI Advisor MCP Server

Kentik’s **AI Advisor MCP Server** enables you to seamlessly integrate [**AI Advisor**](/v1/docs/ai-advisor) into external AI agents.

## **What is MCP?**

The **Model Context Protocol** **(MCP)** is an emerging industry standard for AI agents to access data and interact with external systems. It's supported by major AI platforms including:

- [**Claude Desktop**](https://code.claude.com/docs/en/desktop-quickstart) **(Anthropic)**
- [**GitHub Copilot**](https://docs.github.com/en/copilot/how-tos/provide-context/use-mcp/extend-copilot-chat-with-mcp)
- [**OpenAI Agents**](https://developers.openai.com/api/docs/guides/agents-sdk/)
- **Custom AI agent frameworks**: LangGraph, Google ADK, LlamaIndex, etc.

> [!TIP]
> **TIP**: Already familiar with MCP? Jump straight to [**Getting Started**](/v1/docs/aia-mcp-getting-started) and start setting up `kentik-ai-advisor`.

## **Key Features**

This Kentik-hosted service provides a standardized interface for AI agents to query your network data and maintain conversational context.

- **Natural Language Queries**: Ask questions about your network in conversational language
- **Multi-turn Conversations**: Maintain context across multiple questions
- **Real-time Progress**: See AI Advisor's reasoning process as it works
- **Asynchronous Processing**: Handles long-running queries efficiently
- **Session Management**: Reviews and continues previous conversations

## **Use Cases**

To get the most out of the AI Advisor MCP Server, tailor your prompts to specific operational goals. Here are a few [**NetOps**](https://www.kentik.com/blog/7-real-ways-to-modernize-netops-with-kentik-ai-advisor/)-related examples of how to format your requests to ensure the AI agent correctly routes the query to Kentik:

> [!WARNING]
> **IMPORTANT**: Always explicitly mention "**Kentik**" in your prompts to ensure your agent uses its [**MCP** **tools**](/v1/docs/aia-mcp-tools).

| Operational Goal | Example Prompt |
| --- | --- |
| **Health Monitoring** | "Show me all devices with high CPU usage from **Kentik**." |
| **Troubleshooting** | "Check in **Kentik** which interfaces have errors in the last hour." |
| **Traffic Analysis** | "Ask **Kentik** what the traffic volume is on my external interfaces." |

## **Next Steps**

- Follow the [**Getting Started**](/v1/docs/aia-mcp-getting-started) guide to go over prerequisites.
- Modify one of the [**Configuration Examples**](/v1/docs/aia-mcp-configuration-examples) for your agent/env of choice and connect this MCP server.
- Study the additional background information:
  - [**Kentik AI Advisor**](/v1/docs/ai-advisor)
  - [**Kentik AI Advisor API**](/v1/docs/ai-advisor-api)
  - [**MCP Specification**](https://modelcontextprotocol.io/)