explaingit

tenfingerseddy/semanticus-vscode

Analysis updated 2026-05-18

★ 10C#Audience · developerComplexity · 4/5Setup · moderate

TLDR

A VS Code extension for editing Power BI and Fabric data models where a human and an AI assistant work on the same live model through a shared engine.

Mindmap

mindmap
  root((repo))
    What it does
      Shared model editing
      Human plus AI access
      Undo history
    Tech stack
      C#
      .NET
      TypeScript
    Use cases
      Edit data models
      AI assisted changes
      Best practice checks
    Audience
      BI developers
      Data analysts

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Edit a Power BI or Fabric semantic model from VS Code instead of Power BI Desktop.

USE CASE 2

Let an AI assistant propose changes to a data model while a human reviews them in the same session.

USE CASE 3

Check a model against best practice rules before publishing it.

USE CASE 4

Trace where a piece of model data comes from and what else depends on it.

What is it built with?

C#.NETTypeScriptVS Code

How does it compare?

tenfingerseddy/semanticus-vscodexiu2/stateofdecay2modmanageravbiswas/shadow-rogue-self-play
Stars10109
LanguageC#C#C#
Last pushed—2024-01-20—
Maintenance—Dormant—
Setup difficultymoderateeasy—
Complexity4/52/5—
Audiencedevelopergeneraldeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Remote XMLA and Fabric features need a compatible tenant, capacity, and permissions beyond a local install.

In plain English

Semanticus is a VS Code extension for working on Power BI and Microsoft Fabric semantic models, the data models that sit behind Power BI reports. It is built around the idea that a person and an AI assistant can work on the exact same model at the same time, through two different doors that both lead to one shared engine. A human uses the Model tree, a Properties panel, and a Studio view inside VS Code. An AI assistant, connected separately through a protocol called MCP, can read and change the same model through the same engine session. Whichever side makes a change, it gets broadcast to the other and added to one shared undo history, so nothing gets out of sync. Importantly, Semanticus itself does not run any AI. It has no built in AI model and stores no API keys for one. The AI assistant is a separate tool chosen and connected by the user, Semanticus is only the shared workspace they both act on. The extension supports working offline on model files, along with live connections to remote servers and Fabric for tenants that have the right permissions. Depending on the platform, it can also work with Power BI Desktop directly and preview local calculation changes before applying them. It includes tools for measure and formula authoring, checking a model against best practice rules, tracing where a piece of data comes from and what depends on it, and comparing or copying between models. Any change to a live, connected model runs through a preview first and needs the user to confirm it before it actually happens, and the extension is designed to fail safely rather than damage the source model if something is not supported. It is built with C# and .NET on the engine side and TypeScript for the VS Code interface, and is available as prebuilt packages for Windows, Linux, and Mac, or can be built from source with the .NET 8 SDK and Node.js.

Copy-paste prompts

Prompt 1
Explain how Semanticus keeps a human's changes and an AI assistant's changes from conflicting on the same model.
Prompt 2
Walk me through connecting an MCP based AI assistant to a Semanticus session.
Prompt 3
Summarize what platforms and Fabric features are supported in this release versus planned for later.

Frequently asked questions

What is semanticus-vscode?

A VS Code extension for editing Power BI and Fabric data models where a human and an AI assistant work on the same live model through a shared engine.

What language is semanticus-vscode written in?

Mainly C#. The stack also includes C#, .NET, TypeScript.

How hard is semanticus-vscode to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is semanticus-vscode for?

Mainly developer.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.

Alternatives & similar repos