Skills
Copilot reads each skill’s description to decide when to load it, then reads the rest of SKILL.md and its references/ only if it needs them.
| Skill | Loads when |
|---|---|
| business-question-evaluation | Answer the five Contoso Pharmacy business questions against the semantic model, record the answers in a run file (evaluation/runs/<label>.json), grade the run, and compare runs before and after model changes. Use when asked to test the model with business questions, create a baseline or after run, grade answers, or explain why a question scored 0 or 1. |
| lab-page-authoring | Write or update a self-paced lab page for the Contoso Pharmacy lab site (Astro Starlight, site/src/content/docs/labs/*.mdx), including the required frontmatter, the ten fixed sections, step anchors that match lab/checks.yaml, site components, and the screenshot checklist. Use when creating a new lab, changing lab steps, fixing a lab-feedback issue, or adding a verify check to a lab. |
| pbir-rebinding | Repair Power BI report visuals stored in PBIR format (visual.json files) after semantic model tables, columns, or measures are renamed, and swap implicit column aggregations for explicit measures. Use after any rename in the TMDL model, when Test-ReportBindings.ps1 reports problems, or when a visual still shows raw names like “Count of fill_id”. |
| pharmacy-model-conventions | House conventions for the Contoso Pharmacy semantic model, covering business names, descriptions, hidden keys, summarization, display folders, required relationships and measures, and the AI-readiness checklist. Use when you assess, review, or change model names, descriptions, measures, relationships, or the date table, or when you need the standard business name for a source column. |
| prep-data-for-ai | Draft the Power BI “Prep data for AI” artifacts for the Contoso Pharmacy model (AI instructions, a simplified AI data schema, synonyms, and verified answer candidates) as reviewable files in fabric/ai-prep/. Use when asked to make the model ready for Power BI Copilot or a Fabric data agent, write AI instructions, add synonyms, or propose verified answers. Drafts only; never writes the Desktop-managed Copilot folder. |
| tmdl-authoring | How to safely edit Power BI semantic model files written in TMDL (Tabular Model Definition Language), including indentation, descriptions, quoting, lineage tags, renames across files, calculated columns, measures, relationships, and the local lint. Use whenever you create or change files under fabric/**/*.SemanticModel/definition/, or when a TMDL lint error needs fixing. |
business-question-evaluation
Section titled “business-question-evaluation”Answer the five Contoso Pharmacy business questions against the semantic model, record the answers in a run file (evaluation/runs/<label>.json), grade the run, and compare runs before and after model changes. Use when asked to test the model with business questions, create a baseline or after run, grade answers, or explain why a question scored 0 or 1.
- File: .github/skills/business-question-evaluation/SKILL.md
- Supporting files: references/run-file.md
lab-page-authoring
Section titled “lab-page-authoring”Write or update a self-paced lab page for the Contoso Pharmacy lab site (Astro Starlight, site/src/content/docs/labs/*.mdx), including the required frontmatter, the ten fixed sections, step anchors that match lab/checks.yaml, site components, and the screenshot checklist. Use when creating a new lab, changing lab steps, fixing a lab-feedback issue, or adding a verify check to a lab.
- File: .github/skills/lab-page-authoring/SKILL.md
- Supporting files: references/components.md
pbir-rebinding
Section titled “pbir-rebinding”Repair Power BI report visuals stored in PBIR format (visual.json files) after semantic model tables, columns, or measures are renamed, and swap implicit column aggregations for explicit measures. Use after any rename in the TMDL model, when Test-ReportBindings.ps1 reports problems, or when a visual still shows raw names like “Count of fill_id”.
pharmacy-model-conventions
Section titled “pharmacy-model-conventions”House conventions for the Contoso Pharmacy semantic model, covering business names, descriptions, hidden keys, summarization, display folders, required relationships and measures, and the AI-readiness checklist. Use when you assess, review, or change model names, descriptions, measures, relationships, or the date table, or when you need the standard business name for a source column.
- File: .github/skills/pharmacy-model-conventions/SKILL.md
- Supporting files: references/ai-readiness-checklist.md, references/data-dictionary.md, references/examples.md
prep-data-for-ai
Section titled “prep-data-for-ai”Draft the Power BI “Prep data for AI” artifacts for the Contoso Pharmacy model (AI instructions, a simplified AI data schema, synonyms, and verified answer candidates) as reviewable files in fabric/ai-prep/. Use when asked to make the model ready for Power BI Copilot or a Fabric data agent, write AI instructions, add synonyms, or propose verified answers. Drafts only; never writes the Desktop-managed Copilot folder.
- File: .github/skills/prep-data-for-ai/SKILL.md
- Supporting files: references/formats.md
tmdl-authoring
Section titled “tmdl-authoring”How to safely edit Power BI semantic model files written in TMDL (Tabular Model Definition Language), including indentation, descriptions, quoting, lineage tags, renames across files, calculated columns, measures, relationships, and the local lint. Use whenever you create or change files under fabric/**/*.SemanticModel/definition/, or when a TMDL lint error needs fixing.
- File: .github/skills/tmdl-authoring/SKILL.md
- Supporting files: references/tmdl-cheatsheet.md
Third-party skill: semantic-model-authoring
Section titled “Third-party skill: semantic-model-authoring”Installed from microsoft/skills-for-fabric by apm install into .agents/skills/ and pinned in apm.lock.yaml. It isn’t committed to this repo. See APM and plugins.