Microsoft Fabric MCP Server
Enables AI assistants to interact with Microsoft Fabric and Power BI services through the Model Context Protocol. Users can manage workspaces, execute DAX queries, refresh datasets, and create Fabric notebooks using natural language.
README
Microsoft Fabric MCP Server
A Model Context Protocol (MCP) server that enables AI assistants like Claude to interact with Microsoft Fabric and Power BI services.
Features
-
📊 Power BI Integration
- List and query datasets
- Execute DAX queries
- Refresh datasets
- Manage workspaces
-
🏭 Microsoft Fabric Support
- Create and manage notebooks
- Upload data to warehouses
- Access Fabric workspaces
Installation
Via NPM (Recommended)
npm install -g @strainprint/microsoft-fabric-mcp
From Source
git clone https://github.com/strainprint/microsoft-fabric-mcp.git
cd microsoft-fabric-mcp
npm install
npm run build
Configuration
1. Azure App Registration
Create an Azure App Registration with the following permissions:
- Power BI Service (Delegated or Application):
- Dataset.Read.All
- Workspace.Read.All
- Dataset.Execute.All (for DAX queries)
2. Power BI Tenant Settings
Enable in Power BI Admin Portal → Tenant settings:
- "Allow service principals to use Power BI APIs"
- "Service principals can access read-only admin APIs"
3. Environment Variables
Create a .env file or set environment variables:
# Required
AZURE_TENANT_ID=your-tenant-id
AZURE_CLIENT_ID=your-client-id
AZURE_CLIENT_SECRET=your-client-secret
# Optional
POWERBI_WORKSPACE_ID=workspace-id # If not set, uses workspace-agnostic endpoints
4. Claude Desktop Configuration
Add to your Claude Desktop config (%APPDATA%\Claude\claude_desktop_config.json):
{
"mcpServers": {
"MicrosoftFabric": {
"command": "npx",
"args": [
"-y",
"@strainprint/microsoft-fabric-mcp"
],
"env": {
"AZURE_TENANT_ID": "your-tenant-id",
"AZURE_CLIENT_ID": "your-client-id",
"AZURE_CLIENT_SECRET": "your-client-secret",
"POWERBI_WORKSPACE_ID": "optional-workspace-id"
}
}
}
}
Usage
Once configured, Claude can use commands like:
- "List my Power BI datasets"
- "Execute this DAX query on dataset X"
- "Refresh the sales dataset"
- "Show me all workspaces"
Available Tools
get_workspaces
Get all accessible Power BI/Fabric workspaces
get_powerbi_datasets
List all datasets in the configured workspace
execute_dax_query
Execute DAX queries against a dataset
{
"datasetId": "dataset-guid",
"query": "EVALUATE TOPN(10, 'Sales')"
}
refresh_dataset
Trigger a dataset refresh
{
"datasetId": "dataset-guid"
}
create_notebook
Create a new Fabric notebook
upload_to_datawarehouse
Upload data to a Fabric data warehouse
Troubleshooting
403 Forbidden Errors
-
Check Service Principal Permissions
- Ensure the service principal is added to the Power BI workspace
- Verify API permissions in Azure Portal
-
Power BI Tenant Settings
- Confirm service principals are allowed to use Power BI APIs
- Check if the service principal is in the allowed security group
-
Token Issues
- Verify client secret hasn't expired
- Ensure correct tenant ID
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
MIT © Strainprint Technologies
Credits
Developed by Steve Adams for the MCP community.
Special thanks to Anthropic for the Model Context Protocol specification.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。