SkillFoxx

SemanticOps MCP

SemanticOps MCP for Power BI: an agent reads and edits Power BI Desktop models, writes DAX measures and finds bottlenecks.

  • 255 ★ on the repository
  • Last commit
MCP serverOfficial

Broad access

Why this level

  • Reads and changes a Power BI semantic model: tables, measures and relationships.
  • Model data may enter the conversation with the AI assistant provider.
All reasons and checks

maxanatsko/mcp-engine-publicAuthor:maxanatsko

Install

Manual install

Text for your agent

Install the server following the instructions on semanticops.dev, add it to the MCP configuration as `mcp-engine`, open a model in Power BI Desktop and ask the assistant to list tables and measures.

Access and checks

How we review

Broad accessThe tool can change data, run commands or handle secrets.

  • Reads and changes a Power BI semantic model: tables, measures and relationships.
  • Model data may enter the conversation with the AI assistant provider.

This is third-party code. Review the repository files before installing.

What it does

SemanticOps MCP, formerly MCP Engine, connects the AI assistants Claude, Copilot and ChatGPT to Power BI Desktop files. The agent queries the model, creates measures, manages relationships and optimizes performance in plain language. The server runs locally on Windows 10 and 11 or macOS and collects no telemetry, and data processing happens on your machine. The project motto makes three promises: roll back mistakes, test before deploy and control what AI can change. The repository holds skills for areas of model work: DAX debugging and performance, model quality assessment, queries, data readiness for AI and initial setup. The package is distributed as `mcp-engine` on npm under a proprietary license that permits personal and commercial use.

Who it is for. Analysts and BI developers on Power BI Desktop who build semantic models and DAX measures together with an AI assistant.

Good fit when

  • A DAX measure returns a wrong total and the assistant works through it step by step.
  • A report opens slowly and causes and a model optimization plan are needed.
  • Before publishing, a model is checked for quality against a rule set.
  • A model has to be described so that Copilot can answer questions about the data.

Not a fit when

  • The work happens in the Power BI cloud service, while the tool is built for Desktop files.
  • You need open source code, while the license is proprietary.
  • The model contains data that must not be shown to an AI provider.

Example request

Show which measures in the model take the longest, find the cause in the DAX and propose a fix that keeps the result unchanged.

MCP

Transport
stdio
Authentication
not required

Limitations

It works with Power BI Desktop on Windows 10 and 11 or macOS. The license is proprietary, the source code is not published in the repository, and documentation and installation are described on semanticops.dev. The server sends nothing to the developer, but your AI assistant passes data to its own provider such as Anthropic or OpenAI, so consider what model details enter the conversation. The tool changes the model, and rollback of mistakes relies on the product's own features.

How to disable. Remove the mcp-engine entry from the client MCP configuration. Power BI Desktop models stay as they were, and saved changes are rolled back with the product's own features.

FAQ

Which clients does SemanticOps MCP support?

Claude Desktop on Windows and macOS, Claude Code, Visual Studio Code and any MCP client. Installation takes about five minutes, with instructions on semanticops.dev.

Does the server collect telemetry?

No, the developer states that the server runs locally and collects no telemetry. Your assistant may send data to the model provider.

What is MCP Engine?

The former name of SemanticOps MCP. In the MCP registry the server is published as `io.github.maxanatsko/mcp-engine`, and the npm package is named `mcp-engine`.

Which skills does the repository hold?

DAX debugging and performance, model quality, queries, AI readiness, onboarding and bootstrap. They help the agent choose tools and the order of work.

Can it be used in a company?

Yes, the proprietary license permits personal and commercial use. The full text and third-party notices are published on semanticops.dev.

Discussion

No questions or reviews yet.

MCP servers
Official

A local Microsoft MCP server for Power BI semantic models: an agent creates tables, DAX measures and relationships in Desktop, Fabric and PBIP files

MCP serverBroad access1.3K

MCP server and CLI to drive real Microsoft Excel: Power Query, DAX, VBA, PivotTables and charts

MCP serverBroad access887

A CLI and 13 skills for Claude Code: the Power BI data model, DAX measures, row-level security and PBIR report pages straight from chat

CLIBroad access491

dbt MCP

dbt-mcp

Official

The official MCP server from dbt Labs: the Semantic Layer, project metadata, dbt commands and SQL against the warehouse from an agent

MCP serverExtended access621

NVIDIA Auto Ontology semantic layer as an MCP server: SQL gets validated, facts land in a personal knowledge base.

MCP serverExtended access383
Official

Dremio MCP server: an agent browses tables, looks for patterns in the data and analyzes cluster workload.

MCP serverExtended access53

README in short

The repository holds releases, the license, the security policy and a set of skills with reference guides. The server code itself is closed.

README badge

Report a problem