What problem does it solve?
This Skill eliminates the need to manually switch between the Fabric portal and terminal for Data Warehouse authoring tasks, enabling data engineers to execute all T-SQL DDL, DML, ingestion, and ETL operations directly from CLI environments for faster, repeatable, and scriptable workflows.
Core Features & Use Cases
- Full T-SQL Authoring Support: Execute DDL (CREATE/ALTER/DROP tables, schemas), DML (INSERT/UPDATE/DELETE/MERGE), COPY INTO/OPENROWSET data ingestion, transaction management, stored procedure deployment, and time travel data recovery against Fabric Data Warehouses and SQL endpoints.
- Agentic Workflow Guardrails: Built-in schema discovery steps, validation checklists, and troubleshooting guides to prevent common errors like snapshot conflicts, incorrect authentication, or data type mismatches.
- Reusable Script Templates: Pre-built Bash and PowerShell templates for common patterns including incremental upserts with retry logic, schema migrations via CTAS workarounds, and end-to-end ELT pipelines.
- Use Case: A data engineer can use this Skill to deploy a new warehouse table, load 1M rows of sales data from ADLS Gen2, run an incremental upsert of new records, and recover accidentally deleted data via time travel, all without leaving the terminal.
Quick Start
Use the sqldw-authoring-cli skill to create a new dbo.FactSales table in your Fabric Data Warehouse and load 1000 rows of sample sales data from an ADLS Gen2 Parquet file using COPY INTO.