| name | mcp-engine-semantic-authoring |
| description | Author and refine Power BI semantic layer logic. Use when creating or updating measures, calculation groups, named expressions, DAX UDFs, Power Query parameters, model properties, or semantic naming and style patterns, and when deciding between measures and calculated columns. |
PBI Semantic Authoring
Use this skill for semantic-layer work that sits above physical schema shape. Read only the bundled references needed for the current semantic task.
Decide the semantic path
- Decide whether the user needs a measure, calculated column alternative, calculation group, shared expression, model property update, or style cleanup.
- Prefer measures for reusable business logic unless persisted row-level attributes are required.
- Keep semantic edits separate from table or relationship edits unless the task explicitly spans both.
Branch by object type
Measures
Calculation groups and DAX UDFs
- Read calc-groups-guide for reusable transformation patterns such as time intelligence and formatting changes.
- Read dax-udf-guide when the user wants reusable DAX function-style logic.
Named expressions and Power Query parameters
Model properties and style
- Read model-properties-guide when updating descriptions, culture, compatibility, annotations, or implicit-measure behavior.
- Read semantic-layer-style-guide when the task is naming, folders, style consistency, or semantic cleanup across many objects.
Guardrails
- Prefer measure-first reasoning before introducing calculated columns.
- Keep naming and foldering consistent with existing model conventions unless the user requests a deliberate change.
- Validate major semantic edits with targeted queries after authoring.
- Use the best-practices guide when the right home for logic is unclear.
References
