Power BI MCP Server

Power BI MCP Server

Enables AI assistants to programmatically manage Power BI workspaces, reports, and dashboards while executing DAX queries and triggering dataset refreshes. It supports secure OAuth2 authentication for operations like report exporting, workspace management, and real-time push dataset updates.

Category
访问服务器

README

Power BI MCP Server

A Model Context Protocol (MCP) server that enables AI assistants to interact with Power BI workspaces, datasets, reports, and dashboards programmatically.

🚀 Features

  • Workspace Management: List and manage Power BI workspaces
  • Report Operations: Access, clone, export, and rebind reports
  • Dataset Management: Execute DAX queries, refresh datasets, manage schedules
  • Dashboard Access: List and interact with dashboards
  • Push Datasets: Create and manage push datasets with real-time data
  • Authentication: Secure OAuth2 authentication with Microsoft Entra ID

📋 Prerequisites

  • Node.js (v18 or higher)
  • npm or yarn
  • Power BI Pro or Premium license
  • Microsoft Entra ID app registration
  • MCP-compatible AI assistant (Claude Desktop, etc.)

🔧 Installation

  1. Clone the repository:
git clone https://github.com/michaelmckinleyconsulting/powerbi-mcp-server.git
cd powerbi-mcp-server
  1. Install dependencies:
npm install
  1. Build the project:
npm run build

⚙️ Configuration

1. Microsoft Entra ID Setup

  1. Register an application in Azure Portal
  2. Configure API permissions for Power BI Service
  3. Note your:
    • Client ID
    • Tenant ID
    • Client Secret (if using app-only auth)

2. Environment Variables

Create a .env file in the root directory (see .env.example for template):

PBI_PUBLIC_CLIENT_ID=your_client_id
PBI_TENANT_ID=your_tenant_id  # Optional, defaults to 'common'
PBI_SCOPES=https://analysis.windows.net/powerbi/api/.default  # Optional, or specify custom scopes

Note: This server uses Authorization Code flow with PKCE (no client secret required). Configure your Azure app with redirect URIs:

  • http://localhost
  • http://127.0.0.1

3. MCP Configuration

Add to your MCP client configuration:

Claude Desktop

{
  "mcpServers": {
    "powerbi": {
      "command": "node",
      "args": ["path/to/powerbi-mcp-server/dist/index.js"],
      "env": {
        "PBI_PUBLIC_CLIENT_ID": "your_client_id",
        "PBI_TENANT_ID": "your_tenant_id"
      }
    }
  }
}

VS Code MCP Extension

{
  "mcp.servers": {
    "powerbi": {
      "command": "node",
      "args": ["path/to/powerbi-mcp-server/dist/index.js"],
      "transport": "stdio",
      "env": {
        "PBI_PUBLIC_CLIENT_ID": "your_client_id"
      }
    }
  }
}

📖 Usage

Basic Operations

Once configured, your AI assistant can:

  • List workspaces: "Show me all my Power BI workspaces"
  • Execute DAX: "Run a DAX query to get sales by region"
  • Export reports: "Export the quarterly report to PDF"
  • Refresh datasets: "Refresh the sales dataset"
  • Manage access: "Add user@example.com to the Analytics workspace"

Example Prompts

"List all reports in my Sales workspace"
"Execute this DAX query against the Finance dataset: EVALUATE SUMMARIZE(...)"
"Export the Monthly KPI report to PowerPoint"
"Show me the refresh history for the Customer dataset"
"Create a push dataset for real-time monitoring"

🔐 Authentication

The server supports two authentication methods:

  1. Interactive (Recommended): User signs in via browser
  2. App-Only: Uses client credentials (requires admin consent)

On first run, the server will prompt for authentication via your default browser.

📁 Project Structure

powerbi-mcp-server/
├── src/
│   ├── index.ts           # Main server entry point
│   ├── auth/              # Authentication logic
│   ├── handlers/          # Request handlers
│   ├── types/             # TypeScript definitions
│   └── utils/             # Utility functions
├── dist/                  # Compiled JavaScript (git-ignored)
├── package.json
├── tsconfig.json
└── README.md

🛠️ Development

Running in Development Mode

npm run dev

Running Tests

npm test

Building for Production

npm run build

📊 Supported Power BI Operations

Workspaces (Groups)

  • List workspaces
  • Get workspace users
  • Add/remove users

Reports

  • List reports
  • Get report metadata
  • Clone reports
  • Export to various formats (PDF, PPTX, PNG)
  • Rebind to different datasets

Datasets

  • List datasets
  • Execute DAX queries
  • Trigger refreshes
  • Manage refresh schedules
  • Get refresh history

Dashboards

  • List dashboards
  • Get dashboard tiles

Push Datasets

  • Create push datasets
  • Add rows to tables
  • Clear table data

🔧 Troubleshooting

Common Issues

  1. Authentication fails: Ensure your Azure app registration has proper Power BI permissions
  2. No workspaces found: Verify your account has access to Power BI workspaces
  3. DAX queries fail: Check dataset permissions and query syntax
  4. Export timeouts: Large reports may take time; the server handles async polling

Debug Mode

Enable verbose logging:

DEBUG=powerbi:* node dist/index.js

📄 License

This project is source-available under a Modified MIT License with Commons Clause - see the LICENSE file for details.

Non-Commercial Use: Free for personal, educational, and internal company use.

Commercial Use: Requires a separate commercial license. Contact [michael@mckinley.consulting] for commercial licensing.

This is not OSI-approved "open source" due to the commercial restrictions, but allows full community use for non-commercial purposes.

🤝 Contributing

We welcome contributions! Please note:

  1. By contributing, you agree that your contributions will be licensed under the same license
  2. For substantial changes, please open an issue first to discuss
  3. Ensure all tests pass and add tests for new features

💬 Support

  • Issues: GitHub Issues
  • Commercial Support: [michael@mckinley.consulting]
  • Documentation: Wiki

🙏 Acknowledgments

  • Built on the Model Context Protocol specification
  • Uses Microsoft Power BI REST APIs
  • Inspired by the MCP community

Note: This is not an official Microsoft product. Power BI is a trademark of Microsoft Corporation.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选