answering-natural-language-questions-with-dbt

What problem does it solve?

This Skill helps data-driven teams answer business questions by orchestrating semantic layer queries, compiled SQL, model discovery, and manifest analysis to obtain reliable answers from dbt projects.

Core Features & Use Cases

• Semantic Layer Query: When a semantic layer is available, query metrics directly and verify dimensions.
• Modified Compiled SQL: If the semantic layer exists but needs minor adjustments, modify the compiled SQL and execute.
• Model Discovery: When no semantic layer but marts/models are accessible, discover models and write SQL against them.
• Manifest/Catalog Analysis: In a full dbt project, analyze manifest/catalog to derive SQL where needed.

Quick Start

Analyze a business question by applying a semantic layer first, then fallback methods as needed.