Powerbi Modeling MCP server
{
"mcpServers": {
"powerbi-modeling-mcp": {
"command": "npx",
"args": [
"-y",
"@microsoft/powerbi-modeling-mcp@latest",
"--start",
"--skipconfirmation",
"--authmode=interactive"
],
"env": {
"PBI_MODELING_MCP_ACCESS_TOKEN": "<pbi_modeling_mcp_access_token>"
}
}
}
}{
"mcpServers": {
"powerbi-modeling-mcp": {
"command": "npx",
"args": [
"-y",
"@microsoft/powerbi-modeling-mcp@latest",
"--start",
"--skipconfirmation",
"--authmode=interactive"
],
"env": {
"PBI_MODELING_MCP_ACCESS_TOKEN": "<pbi_modeling_mcp_access_token>"
}
}
}
}{
"mcpServers": {
"powerbi-modeling-mcp": {
"command": "npx",
"args": [
"-y",
"@microsoft/powerbi-modeling-mcp@latest",
"--start",
"--skipconfirmation",
"--authmode=interactive"
],
"env": {
"PBI_MODELING_MCP_ACCESS_TOKEN": "<pbi_modeling_mcp_access_token>"
}
}
}
}{
"servers": {
"powerbi-modeling-mcp": {
"command": "npx",
"args": [
"-y",
"@microsoft/powerbi-modeling-mcp@latest",
"--start",
"--skipconfirmation",
"--authmode=interactive"
],
"env": {
"PBI_MODELING_MCP_ACCESS_TOKEN": "${input:pbi_modeling_mcp_access_token}"
}
}
},
"inputs": [
{
"id": "pbi_modeling_mcp_access_token",
"type": "promptString",
"description": "Pbi modeling mcp access token",
"password": true
}
]
}{
"mcpServers": {
"powerbi-modeling-mcp": {
"command": "npx",
"args": [
"-y",
"@microsoft/powerbi-modeling-mcp@latest",
"--start",
"--skipconfirmation",
"--authmode=interactive"
],
"env": {
"PBI_MODELING_MCP_ACCESS_TOKEN": "<pbi_modeling_mcp_access_token>"
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y @microsoft/powerbi-modeling-mcpversion 0.5.0-beta.13
What it does
The Power BI Modeling MCP Server connects AI agents to Power BI semantic models, allowing natural language control over modeling operations like creating tables, columns, measures, and relationships. It is for developers and AI applications that need to automate Power BI model creation and modification at scale.
- Build and modify Power BI semantic models using natural language instructions to an AI assistant.
- Execute bulk modeling operations such as renaming, refactoring, or translating hundreds of model objects simultaneously.
- Query and validate DAX expressions against your model to test measures and troubleshoot calculations.
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.