What problem does it solve?
Analytics teams need a consistent, safe, and repeatable way to author, test, validate, and troubleshoot dbt projects across dbt Core, Fusion, and dbt Cloud. This Skill centralizes guidance and operational instructions so engineers and AI agents can build models, add unit tests, query the semantic layer, and diagnose job failures without guesswork.
Core Features & Use Cases
- Model authoring & planning: guidance for planning models, choosing materializations, and following DRY best practices.
- Test-driven development: generate and run dbt unit tests, fixture management, and CI-friendly recommendations.
- Semantic layer & metrics: create or modify semantic models, metrics, time spines, and validate via MetricFlow or dbt sl tools.
- CLI & job troubleshooting: format and run dbt CLI commands safely, analyze run_results.json, and investigate dbt Cloud job failures.
- Docs & discovery: fetch dbt docs in LLM-friendly markdown and explore project artifacts for discovery and impact analysis.
Quick Start
Ask the dbt skill to "build and unit-test model orders, list downstream impact, and show the first 10 rows of the compiled model."