mcp-powerBI
Enables discovery of Power BI workspaces and semantic models via REST API and Microsoft Modeling MCP, with authentication options including service principal and device login.
README
mcp-powerBI
Claude-compatible MCP server for discovering Fabric/Power BI workspaces and semantic models.
This repo is intentionally based around Microsoft's official powerbi-modeling-mcp:
get_catalog,list_workspaces, andlist_semantic_modelsuse the Power BI REST API for tenant/workspace discovery.list_semantic_models_in_workspace_via_modeling_mcplaunches Microsoft@microsoft/powerbi-modeling-mcpand uses its XMLA/TOM auth path to enumerate semantic models inside a known workspace.
That split is necessary because Microsoft Power BI Modeling MCP can connect and model/query semantic models, but it does not expose a tenant-wide workspace discovery tool.
Tools
auth_statusstart_device_logincomplete_device_loginlist_workspaceslist_semantic_modelsget_cataloglist_semantic_models_in_workspace_via_modeling_mcpget_known_workspace_catalogexecute_dax_query
Install
npm install
npm run setup
npm run build
npm run setup asks for:
- Azure app display name
- Directory tenant ID/domain
- Application client ID
- Client secret value
- Microsoft
powerbi-modeling-mcpcommand and args - Known workspace names
- Default CEO workspace
- Optional default semantic model fallback
It writes a local .env file with mode 0600. The MCP server loads this file automatically on start.
Claude Desktop config
Use the built JS after npm run build.
For this machine, start from docs/claude-desktop-config.example.json. It points the wrapper to the already installed Microsoft native binary:
/Users/ducna/.codex/mcp/powerbi-modeling-mcp/node_modules/@microsoft/powerbi-modeling-mcp-darwin-arm64/dist/powerbi-modeling-mcp
Generic service-principal config:
{
"mcpServers": {
"mcp-powerbi": {
"command": "node",
"args": ["/absolute/path/to/mcp-powerBI/dist/server.js"],
"env": {
"POWERBI_TENANT": "vnu.edu.vn",
"POWERBI_CLIENT_ID": "<app-client-id>",
"POWERBI_CLIENT_SECRET": "<client-secret-value>",
"POWERBI_MODELING_MCP_COMMAND": "/absolute/path/to/powerbi-modeling-mcp",
"POWERBI_MODELING_MCP_ARGS": "--start"
}
}
}
}
For local development:
{
"mcpServers": {
"mcp-powerbi": {
"command": "npx",
"args": ["tsx", "/absolute/path/to/mcp-powerBI/src/server.ts"]
}
}
}
Authentication
The server checks auth in this order:
POWERBI_ACCESS_TOKEN- service principal:
POWERBI_TENANT,POWERBI_CLIENT_ID,POWERBI_CLIENT_SECRET - cached delegated user token from
start_device_login/complete_device_login
For production Claude usage, service principal is the most reliable option.
Power BI tenant/admin requirements:
- Enable Allow service principals to use Power BI APIs.
- Add the service principal to the relevant workspaces, or to an allowed security group.
- App/API permissions should allow workspace and dataset reads. In practice this normally means Power BI REST API application permissions such as
Workspace.Read.AllandDataset.Read.All, with admin consent where required. - Power BI Remote MCP Preview is a separate tenant setting for Microsoft's hosted remote MCP endpoint. It is not the same as allowing service principals to call Power BI APIs.
Usage Examples
Ask Claude:
Use mcp-powerbi to get the full catalog of workspaces and semantic models.
or:
Use mcp-powerbi to list semantic models in workspace test-mcp via Microsoft Modeling MCP.
The second path works when the workspace name is already known and Microsoft powerbi-modeling-mcp can authenticate to XMLA.
If the workspace/model is not provided, Claude should call get_catalog first. If REST authentication is unavailable, Claude should ask the user for the workspace name instead of guessing.
For a CEO workflow, set:
POWERBI_KNOWN_WORKSPACES=test-mcp
POWERBI_DEFAULT_WORKSPACE=test-mcp
# Optional fallback only. Prefer letting Claude choose from workspace schema.
# POWERBI_DEFAULT_SEMANTIC_MODEL=hospital
Then Claude can use get_known_workspace_catalog to list models from configured workspaces without REST auth, choose the relevant semantic model from schema/context, and call execute_dax_query for follow-up business questions. The wrapper keeps the Microsoft Modeling MCP process alive, so repeated questions reuse the same process and should reduce repeated login prompts.
CEO Operating Mode
For the simplest CEO experience:
- Keep Claude Desktop and this MCP server running during the working session.
- Avoid restarting Claude between related questions.
- Configure
POWERBI_DEFAULT_WORKSPACEandPOWERBI_DEFAULT_SEMANTIC_MODEL. - Configure
POWERBI_KNOWN_WORKSPACESandPOWERBI_DEFAULT_WORKSPACE. - Treat
POWERBI_DEFAULT_SEMANTIC_MODELas an optional fallback, not a required CEO input. - Ask business questions in plain language; Claude should generate DAX and call
execute_dax_query.
The first query in a fresh session can still trigger Microsoft authentication. Follow-up queries in the same running MCP session reuse the existing Microsoft Modeling MCP process and connection.
Environment
Copy .env.example for local shell usage:
cp .env.example .env
Then export values before running:
set -a
source .env
set +a
npm run dev
Notes
list_workspacesusesGET https://api.powerbi.com/v1.0/myorg/groups.list_semantic_modelsusesGET /datasetsfor My workspace orGET /groups/{groupId}/datasetsfor a workspace.- The Microsoft Modeling MCP bridge uses
npx -y @microsoft/powerbi-modeling-mcp@latest --startby default. Override withPOWERBI_MODELING_MCP_COMMANDandPOWERBI_MODELING_MCP_ARGSif you have a signed local binary. - Local verification notes are in
docs/verification.md.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.