powerbi-mcp and pbixray-mcp-server

These tools are competitors, as both aim to provide an MCP server for Large Language Models to interact with Power BI datasets, with project A focusing on natural language interaction and project B on providing full Power BI model context from a .pbix file.

powerbi-mcp
51
Established
pbixray-mcp-server
42
Emerging
Maintenance 6/25
Adoption 9/25
Maturity 15/25
Community 21/25
Maintenance 2/25
Adoption 7/25
Maturity 16/25
Community 17/25
Stars: 102
Forks: 44
Downloads:
Commits (30d): 0
Language: Python
License: MIT
Stars: 39
Forks: 11
Downloads:
Commits (30d): 0
Language: Python
License: MIT
No Package No Dependents
Stale 6m No Package No Dependents

About powerbi-mcp

sulaiman013/powerbi-mcp

MCP server for natural language interaction with Power BI datasets

Implements dual connectivity to both Power BI Desktop (via TOM) and Power BI Service (via XMLA endpoints), exposing 34 tools for semantic model querying, DAX execution, and bulk refactoring. Solves the critical challenge of renaming tables without breaking report visuals through PBIP file-based editing—directly manipulating TMDL definitions and JSON report bindings rather than relying on TOM, which can't update the report layer.

About pbixray-mcp-server

jonaolden/pbixray-mcp-server

MCP server to give llms such as Claude, GitHub Copilot etc full PowerBI model context (from input .pbix) through tools based on PBIXRay python package.

Exposes Power BI model internals through 14 configurable MCP tools—including Power Query (M) code, DAX expressions, relationships, and paginated table data—enabling LLMs to analyze and understand complete .pbix structure. Built on the PBIXRay library and deployed via stdio transport, it supports selective tool disabling for security, configurable pagination (default 20 rows per page), and operates seamlessly in WSL environments for Windows integration. The MCP Inspector provides interactive testing during development, with sample PBIX files included for immediate experimentation.

Scores updated daily from GitHub, PyPI, and npm data. How scores work