semantic-model-authoring

Author, inspect, deploy, and optimize Power BI semantic models via MCP, TMDL, and Fabric REST APIs.

Updated Aug 19, 2026
One-click install
npx skills add https://github.com/paulasilvatech/Fabric-Agentic-SDLC --skill semantic-model-authoring-paulasilvatech
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: semantic-model-authoring
Source: https://github.com/paulasilvatech/Fabric-Agentic-SDLC/tree/main/.github/skills/semantic-model-authoring
Command: npx skills add https://github.com/paulasilvatech/Fabric-Agentic-SDLC --skill semantic-model-authoring-paulasilvatech

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Creating and maintaining Power BI semantic models requires choosing the right tool (modeling MCP, TMDL file edits, or Fabric REST APIs) for each environment and permission level, and mistakes like duplicate deployments or stale TMDL edits are common. This Skill routes every authoring, inspection, deployment, refresh, and DAX optimization request through the correct workflow and tool. ## Core Features & Use Cases - Model authoring and metadata discovery: Create Import, DirectQuery, and Direct Lake models from scratch, or list tables, columns, measures, and relationships using MCP TOM inspection or DAX INFO functions. - DAX optimization and best-practice analysis: Diagnose slow measures with trace-based FE/SE analysis and apply a tiered catalog of DAX, query-structure, and model optimization patterns. - Deployment and management: Deploy models to Fabric workspaces, trigger refreshes, configure data sources, parameters, permissions, and bind models to Fabric connections. - Use Case: A user asks to add a year-to-date measure to a Fabric semantic model. The Skill resolves the workspace and model IDs, connects via the modeling MCP, applies the DAX change following modeling guidelines, and validates the result. ## Quick Start Use the semantic-model-authoring skill to list all tables and measures in my Fabric semantic model and then create a new DAX measure for year-to-date sales.

Frequently Asked Questions about semantic-model-authoring

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a Power BI semantic model from scratch?▼

Gather the data source details and business entities first, then choose a storage mode: Direct Lake for OneLake sources, Import by default, DirectQuery only when requested. Build the star schema with partitions, columns, relationships, and measures, then deploy once to Fabric or export to a PBIP project.

How do I list tables, columns, and measures in a semantic model?▼

Use the powerbi-modeling-mcp List and Get operations when connected with Write access, or run DAX INFO.VIEW functions through FabricIQ ExecuteQuery when you only have Read permission. Avoid GetSemanticModelSchema because it can return stale metadata.

Should I use the modeling MCP or edit TMDL files directly?▼

Prefer the powerbi-modeling-mcp server whenever it is registered and connected, since it edits the live model with no TMDL desync risk. Edit TMDL files directly only when MCP is unavailable and you have a PBIP folder or Fabric workspace source.

Why does deploying a semantic model to Fabric fail with a duplicate name error?▼

Deploys are not idempotent, so a retried deploy creates a second model with the same name and subsequent deploys fail. List the workspace's semantic models, delete every duplicate with that name, then deploy once from a clean state.

Can this skill answer questions about the data inside a semantic model?▼

No, it is authoring-scoped and only handles model metadata, structure, and management. For natural-language or data questions against a model, use the FabricIQ skill instead.

How do I optimize a slow DAX measure in Power BI?▼

Establish a baseline with trace capture to measure formula engine versus storage engine time, then apply Tier 1 DAX rewrite patterns from the decision guide. Query structure and model changes require user approval since they can alter results or break downstream reports.