What problem does it solve? Building and maintaining Power BI / Analysis Services semantic models from the terminal is error-prone: agents pick the wrong tool, break relationships, or ship models with DAX correctness bugs, bloated VertiPaq storage, and missing metadata. This Skill provides a single operating loop that routes every modeling operation to the narrowest capable tool and audits models against best-practice categories. ## Core Features & Use Cases - Tool cascade routing: Drives every change through the te CLI first (measures, relationships, roles/RLS, calculation groups, incremental refresh), falls back to TOM via te script or connect-pbid for unreachable properties, and uses fab + TMDL authoring only for service- and file-shape operations. - Full lifecycle coverage: Design (star schema, storage modes, Direct Lake), build (measures with full metadata, validation and BPA gates), refresh (incremental refresh policy, framing), and review (prioritized audit findings across critical, memory, DAX, and AI-readiness categories). - Deep reference library: Twenty-plus reference documents covering relationships, calculation groups, composite models, Direct Lake guardrails, Copilot grounding, security, and performance, plus scripts for model metadata gathering and AI metadata management. - Use Case: Ask the agent to add a measure and set up RLS on a model; it stages the changes with te add, validates with te validate, gates on te bpa run --fail-on error, and saves only when the DAX and referential-integrity checks pass. ## Quick Start Ask the agent to review the semantic model in your workspace for quality and performance issues and produce prioritized findings.
