Connecting Power BI to Claude Desktop via the Model Context Protocol (MCP) allows you to manage your data models using natural language. This guide provides a professional walkthrough for configuring this environment.

What is an MCP Server?

The Model Context Protocol (MCP) is an open standard that enables developers to build a secure, two-way bridge between AI models and local or remote data sources. Think of an MCP Server as a universal translator for AI: it takes the “brain” of an LLM (like Claude) and gives it “hands” to interact with specific software like Power BI.  

Prerequisites

Before beginning the configuration, ensure the following software is installed on your workstation:

• Claude Desktop
• Visual Studio Code (VS Code)

Step 1— Install the Power BI MCP Extension

1. Open VS Code
2. Go to the Extensions panel
3. Search for: “Power BI Modeling MCP Server

Step 2: Install the Power BI MCP Extension

To bridge the gap between Power BI and your AI agent, you must first install the necessary extension within VS Code.

1. Launch Visual Studio Code.
2. Navigate to the Extensions panel on the sidebar.
3. In the search bar, enter: Power BI Modeling MCP (Analysis Services).
4. Select the extension and click Install

Step 3 — Locate the MCP Server Installation Path 

After installation, the extension is stored locally on your system.  Go to the following path:  %USERPROFILE%.vscodeextensions  Locate the folder beginning with analysis-services.powerbi-modeling-mcp    Then click the server folder Find the powerbi-modeling -mcp file Copy the full file path of this executable and save it in a text editor for use in the next step.

Step 4: Configure Claude Desktop

You must now inform Claude Desktop where the MCP server is located by editing its configuration file   Open Claude Desktop and navigate to Settings.   Select the Developer tab and click the Edit Config button.   This action opens the configuration folder. Open the claude_desktop_config.json file in a text editor like Notepad. Input the following JSON structure, ensuring you replace the command value with the path you copied in Step 2 { “mcpServers”: { “powerbi-modeling-mcp”: { “command”: “Paste Path Here Replace with “, “args”: [“–start”], “type”: “stdio” } } } Now Paste the Path that you copied earlier inside double quotes

Important: In the JSON file, you must use double backslashes () for the file path to be valid

SAMPLE: { “mcpServers”: { “powerbi-modeling-mcp”: { “command”: “C:UsersTech Arc.vscodeextensionsanalysis-services.powerbi-modeling-mcp-0.1.9-win32-x64serverpowerbi-modeling-mcp.exe”, “args”: [ “–start” ] } } Save the file (CTRL + S) and restart your computer to initialize the server.

Step 5: Verification and Implementation

After your system restarts, you can begin interacting with your Power BI models directly through Claude.  

Establishing a Connection

Open your Power BI file (e.g., “Account Receivables”) and Claude Desktop. Use a prompt such as: “Connect to my opened Power BI file named Account Receivables.”     Claude will confirm the active connection and list the operations it can perform, such as creating measures or running DAX queries.

Example: Automating Model Tasks

You can now automate complex modeling tasks through simple natural language commands:

• Create Tables: “Create a new calculation table called ‘KeyMeasures 2’ as a dedicated measures table.”
• Generate Measures: “In the table ‘KeyMeasures 2’, create financial balance measures for Opening, Ending, Overdue, and Current balances.”

The measures will be generated instantly within your Power BI model, ready for use in your reports.   And it is working correctly.