Skip to content

Guides

Task-oriented how-tos. Each assumes you've completed getting started and have a scaffolded project.

ade-ops is usable from a plain CLI, but reaches its potential when paired with an AI coding assistant that drives its skills. Each task below maps to a skill (Claude Code) and its CLI equivalent — pick whichever fits your workflow.

Task Skill CLI
See what would change before pushing /ops-diff python -m core.cli diff --env <env> --scope <scope>
Deploy notebooks to a remote /ops-push, /databricks-deploy python -m core.cli push --env <env> --scope notebooks
Run a notebook or job /databricks-run python -m core.cli databricks-run ...
Query a SQL warehouse /databricks-query python -m core.cli databricks-query ...
Inspect job/notebook lineage /databricks-lineage python -m core.cli databricks-lineage ...
Deploy a notebook / pipeline to Fabric /fabric-notebook-deploy, /fabric-pipeline-deploy python -m core.cli fabric-notebook-deploy ...
Scaffold a Power BI semantic model /powerbi-model-create, /powerbi-directlake-create
Publish a Power BI model /powerbi-publish
Build a Power BI (PBIR) report /pbir-create, /pbir-report
Assess a Databricks → Fabric migration /migration-assess python -m core.cli migration-assess ...
Promote across environments /ops-push --env cert / --env prod python -m core.cli push --env cert

How skills and the CLI relate

Each skill body (under .claude/commands/ in the repo) leads with a runnable python -m core.cli … command, then layers the agentic workflow on top. So a skill doubles as a how-to recipe even if you're not using Claude Code — read the command block and the steps, run them yourself.

Step-by-step long-form guides are landing here as the docs build out. Until then, the skill bodies are the authoritative recipes, and the quickstarts cover end-to-end scenario setup.