microsoft/skills-for-fabric v0.3.2

v0.3.2

Repository: microsoft/skills-for-fabric

Tag: v0.3.2

Published: 2026-06-11T10:47:20Z

Prerelease: no

Release notes:

[0.3.2] - 2026-06-03

Added

• `semantic-model-authoring` — develop and manage Power BI semantic models across Power BI Desktop, PBIP projects, and the Fabric Service. Covers creating models (Import, DirectQuery, Direct Lake), editing measures/tables/columns/relationships, deploying to Fabric workspaces, refreshing, configuring data sources and permissions, and DAX performance optimization. Ships with 11 reference guides (connection binding, DAX guidelines, DAX performance decision guide, DAX performance patterns, Direct Lake guidelines, modeling guidelines, naming conventions, PBIP, semantic-model AI readiness, semantic-model REST API, TMDL guidelines). Replaces `powerbi-authoring-cli`.
• `semantic-model-consumption` — execute raw DAX queries and inspect metadata of Microsoft Fabric Power BI semantic models via the MCP server ExecuteQuery tool. Use when you already know the DAX (EVALUATE statements) or need to inspect tables, columns, measures, relationships, and hierarchies via INFO functions. Replaces `powerbi-consumption-cli`.
• `fabriciq` — answer business questions by querying Power BI reports and dashboards through the FabricIQ MCP endpoint. Orchestrates artifact discovery, schema inspection, entity-value resolution, DAX generation, and query execution; returns plain-language answers. Use for natural-language questions about Power BI report/dashboard content (use semantic-model-consumption for raw DAX).
• `FabricIQ` agent — answers questions about Power BI artifacts (reports and semantic models) by discovering artifacts, inspecting metadata and schemas, resolving entity values, generating DAX, and executing queries against the Fabric MCP endpoint. Delegates to fabriciq.
• Dedicated `powerbi-authoring` plugin bundle — ships semantic-model-authoring and check-updates with the powerbi-modeling-mcp server (@microsoft/powerbi-modeling-mcp) pre-configured for fine-grained semantic-model modeling operations. Install via /plugin install powerbi-authoring@fabric-collection.
• `dataflows-authoring-cli` reference docs (3 new) — output-destinations.md (Lakehouse/Warehouse/SQL DB output destination patterns including staging behavior, schema mapping, and refresh semantics), connection-management.md (creating, binding, and rotating connection IDs for Dataflows Gen2), and mashup-preview.md (inspecting and validating Power Query M before publishing).
• `spark-operations-cli` automated diagnostic workflow — new references/automated-diagnostic-workflow.md for end-to-end Spark/Livy diagnostics: job triage → executor/driver log mining → Spark Advisor findings → mitigation recommendations.
• `synapse-migration` deep resources (12 new) — capacity sizing, connector refactoring, external Hive Metastore migration, feature parity matrix, lake database migration, library compatibility, migration gotchas, migration orchestrator, migration report, security and governance, Spark item migration, Spark pool migration, and validation/testing.
• `EVENTHOUSE-CONSUMPTION-CORE` common reference — shared Eventhouse/KQL consumption patterns surfaced via the fabric-authoring plugin bundle.

Changed

• `powerbi-authoring-cli` renamed to `semantic-model-authoring` — aligns the skill name with the underlying Microsoft Fabric / Power BI artifact (a *semantic model*) rather than the surface tool. Same coverage of model authoring plus an expanded reference library. Re-invoke as semantic-model-authoring going forward.
• `powerbi-consumption-cli` renamed to `semantic-model-consumption` — same rationale; same DAX query / metadata surface. Re-invoke as semantic-model-consumption going forward.