microsoft-fabric-unified-analytics — Claude Code skill for Fabric
microsoft-fabric-unified-analytics walks you through a Fabric analytics stack: OneLake layers, a Dataflow Gen2 cleaning step, PySpark gold tables and Power BI on top.
no license 282 installs ★ 5 updated 2026-10-09
npx skills add reason-machines/data-skills --skill microsoft-fabric-unified-analytics -a claude-codeWhat microsoft-fabric-unified-analytics does
The skill lays out one analytics pipeline on Microsoft Fabric. Raw files land in OneLake as bronze, a Dataflow Gen2 query in Power Query M cleans them into silver, PySpark notebooks in Fabric turn silver into gold KPI tables, and a semantic model feeds Power BI reports above that.
Concrete pieces come with it: the Lakehouse folder layout, an M query that drops duplicate InvoiceNo/StockCode rows, skips records with no CustomerID, adds line_total, splits the date into year and month, and flags returns by the “C” prefix on an invoice number. Another snippet refreshes the dataflow through the Fabric REST API using a bearer token from environment variables.
Capacity is the gate. A workspace sits on Fabric capacity, so a Premium or trial licence comes first, and the skill suggests F64 as a starting size. Publishing reports needs Power BI Pro or Premium. Python 3.8+ only matters when you run notebooks on your own machine.
What it helps with
- split raw CSVs into bronze, silver and gold layers in OneLake
- clean the data in Dataflow Gen2 without writing Python
- compute revenue and average order value in a PySpark notebook
- attach a semantic model to a Power BI report
- trigger a dataflow refresh from a script over the REST API
Example prompts for the agent
- Set up the Fabric Lakehouse with bronze, silver and gold folders and place
online_retail.csvin the raw layer - Write a PySpark notebook that aggregates monthly revenue into the gold layer
- Describe a Dataflow Gen2 query that dedupes on
InvoiceNoandStockCodeand addsline_total
How to install microsoft-fabric-unified-analytics
1Skills CLI (Vercel)
npx skills add reason-machines/data-skills --skill microsoft-fabric-unified-analytics -a claude-code2Manually: put the files in place
~/.claude/skills/microsoft-fabric-unified-analytics/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/microsoft-fabric-unified-analytics/
1Skills CLI (Vercel)
npx skills add reason-machines/data-skills --skill microsoft-fabric-unified-analytics -a codex2Manually: put the files in place
.agents/skills/microsoft-fabric-unified-analytics/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add reason-machines/data-skills --skill microsoft-fabric-unified-analytics -a cursor2Manually: put the files in place
.cursor/skills/microsoft-fabric-unified-analytics/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install reason-machines/data-skills skills/microsoft-fabric-unified-analytics2Manually: put the files in place
.github/skills/microsoft-fabric-unified-analytics/SKILL.md1Skills CLI (Vercel)
npx skills add reason-machines/data-skills --skill microsoft-fabric-unified-analytics -a gemini-cli2Platform CLI
gemini skills install https://github.com/reason-machines/data-skills --path skills/microsoft-fabric-unified-analyticsManually: put the files in place
.devin/skills/microsoft-fabric-unified-analytics/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add reason-machines/data-skills --skill microsoft-fabric-unified-analytics -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about microsoft-fabric-unified-analytics
How do I install microsoft-fabric-unified-analytics in Claude Code?
Run npx skills add reason-machines/data-skills --skill microsoft-fabric-unified-analytics -a claude-code. The skill file ends up at ~/.claude/skills/microsoft-fabric-unified-analytics/SKILL.md.
What does this skill do?
It maps the whole data path in Microsoft Fabric: a Lakehouse on OneLake with bronze, silver and gold layers, cleaning in Dataflow Gen2, KPI work in PySpark notebooks, then a semantic model and Power BI reports.
What do I need before using it?
A Fabric workspace backed by Premium or trial capacity, plus Power BI Pro or Premium to publish reports. Python 3.8+ is needed only for local notebook work, and an Azure Storage Account applies when you ingest from an outside source.