dbt — паттерны трансформации аналитических данных

★ 8.5 · data

dbt-transformation-patterns is a Claude Code skill that provides production-ready dbt (data build tool) patterns for analytics engineering, covering model organization, data quality testing, documentation, and incremental processing strategies. It structures projects around a medallion architecture with four layers — sources, staging, intermediate, and marts — enforced through naming conventions using stg_, int_, dim_, and fct_ prefixes. The skill includes ready-to-use dbt_project.yml configuration, a full models/ directory layout with Stripe and Shopify examples, and materialization rules: views for staging, ephemeral for intermediate, and tables for marts. Additional pattern documentation is referenced in references/details.md. Ideal for analytics and data engineers building transformation pipelines, modeling large datasets with incremental runs, or standardizing a dbt project from scratch.