Power BI Modeling

Use this skill when the work is inside a Power BI semantic model rather than a generic SQL schema or spreadsheet.

Activation Conditions

• Designing or cleaning up a star schema
• Creating or reviewing DAX measures
• Configuring relationships and cross-filter direction
• Implementing row-level security
• Auditing model health and performance

Practical Workflow

1. Inspect the current model before changing anything.
2. Classify tables as dimension, fact, bridge, or helper tables.
3. Prefer explicit measures over implicit aggregation.
4. Keep relationships simple and single-direction unless the use case is proven.
5. Hide technical fields from report authors.

MCP Reality

Power BI model tooling is host-specific. If your client exposes a Power BI modeling MCP server, inspect the available operations first and map them to the model areas you need: connections, tables, columns, measures, relationships, DAX queries, and security roles.

For Microsoft documentation, the Microsoft Learn MCP server is a good companion. Prefer:

• microsoft_docs_search_by_product with power-bi
• microsoft_docs_fetch for the final page

References & Resources

Documentation

• STAR-SCHEMA - Dimension and fact modeling guidance
• RELATIONSHIPS - Relationship patterns and cross-filter tradeoffs
• MEASURES-DAX - DAX naming and measure design
• PERFORMANCE - High-impact optimization ideas
• RLS - Row-level security patterns

Scripts

• Power BI Model Audit - Local audit helper for naming, documentation, and modeling smells

Examples

• Model Examples - Example modeling patterns and DAX structure

Cross-Client Portability

This skill is written to stay usable across GitHub Copilot, Claude Code, Codex, and Gemini CLI.

• GitHub Copilot: keep the folder in a Copilot-visible skill or plugin path, or wrap the workflow as project instructions if the host does not support portable skill folders directly.
• Claude Code: keep the folder in a local skills directory or a compatible plugin or marketplace source.
• Codex: install or sync the folder into $CODEX_HOME/skills/<skill-name> and restart Codex after major changes.
• Gemini CLI: this repository generates a project command named /skills:powerbi-modeling from this skill. Rebuild commands with python scripts/export-gemini-skill.py powerbi-modeling and then run /commands reload inside Gemini CLI.

MCP Availability And Fallback

Preferred MCP servers for this skill:

• Power BI MCP (primary)

If MCP is unavailable in the current host:

• Use Power BI Desktop, Tabular Editor, DAX Studio, and exported model metadata when the MCP surface is unavailable.
• Validate measures, relationships, and performance with local model tools before completion.

Related Skills

Skill	Relationship
microsoft-development	Official Microsoft docs for Power BI capabilities and limits
sql-development	Shape the upstream warehouse or SQL source feeding the model
excel-sheet	Excel as a data source or export target