answering-natural-language-questions-with-dbt

Generates and executes SQL to answer natural-language business questions using dbt semantic layer or compiled SQL.

Updated Mar 22, 2026
One-click install
npx skills add https://github.com/shapris/svarus-darbas --skill answering-natural-language-questions-with-dbt-shapris
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: answering-natural-language-questions-with-dbt
Source: https://github.com/shapris/svarus-darbas/tree/main/.kilocode/skills/dbt/skills/answering-natural-language-questions-with-dbt
Command: npx skills add https://github.com/shapris/svarus-darbas --skill answering-natural-language-questions-with-dbt-shapris

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill turns natural-language business questions into accurate data answers by using dbt's semantic layer, compiled SQL, model discovery, or manifest/catalog analysis so users don't have to write complex SQL or hunt for sources manually.

Core Features & Use Cases

  • Semantic-first querying: Prefer metrics from the semantic layer and query them directly when available.
  • Compiled SQL modification: Fetch compiled metric SQL, apply small customizations (filters, dimensions, case logic), and execute the resolved SQL.
  • Model discovery & manifest analysis: When no semantic layer is present, discover marts and model schemas, write SQL using refs, or inspect target/manifest.json and target/catalog.json to construct queries.
  • Use Cases: Answer questions like "What were total sales last month?", "Show revenue by region", or "List top customers by lifetime revenue" without needing the user to supply SQL.

Quick Start

Show total sales for last quarter using the semantic layer and fall back to compiled SQL or manifest inspection if needed.

Frequently Asked Questions about answering-natural-language-questions-with-dbt

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I query dbt metrics using natural language instead of writing SQL?

To query dbt metrics using natural language, this Skill translates business questions into SQL by leveraging the dbt semantic layer, compiled SQL, or manifest inspection to generate and execute queries automatically.

What is the best way to answer business questions with a dbt semantic layer?

Answering business questions with a dbt semantic layer involves querying defined metrics directly. This Skill prioritizes semantic-first querying, falling back to compiled SQL modification or manifest analysis if metrics are unavailable.

Can I explore dbt model schemas and discover marts without a semantic layer?

Yes, you can explore dbt model schemas and discover marts without a semantic layer. This Skill inspects target/manifest.json and target/catalog.json to construct queries using dbt refs.

Do I need a database connection to execute natural language queries against dbt?

Yes, a database connection is required to execute queries and return results. Alternatively, access to semantic layer APIs or compiled metric SQL can be used to resolve natural language questions.

How to customize compiled metric SQL for specific filters and dimensions in dbt?

To customize compiled metric SQL for filters and dimensions, this Skill fetches compiled metric SQL and applies small customizations like case logic before executing the resolved SQL against the data warehouse.