# Data Semantic Layer and Metrics Definitions
Status: public
Confidence: medium (0.725) (verified)
Last verified: 2026-06-02
Generation: ai_structured


## TL;DR

A semantic layer turns business metrics and dimensions into versioned definitions that agents can query without guessing SQL semantics.

## Core Explanation

Data agents often fail because table names do not encode business meaning. A semantic layer defines metrics, dimensions, relationships, filters, and access policies in a shared model. That gives agents a safer interface for analytics questions than raw warehouse tables.

The important evidence is not only the metric name. Agents should collect the metric definition, grain, time dimension, filters, owner, version, and allowed dimensions before answering a business question.

## Source-Mapped Facts

- dbt documentation describes the dbt Semantic Layer as a way to define metrics and dimensions in code. ([source](https://docs.getdbt.com/docs/use-dbt-semantic-layer/dbt-sl))
- Cube documentation describes Cube as providing a semantic layer that gives consumers shared context from data models. ([source](https://docs.cube.dev/docs/introduction))
- Looker documentation describes LookML as a language for describing dimensions, aggregates, calculations, and data relationships. ([source](https://cloud.google.com/looker/docs/lookml-terms-and-concepts))

## Further Reading

- [dbt Semantic Layer](https://docs.getdbt.com/docs/use-dbt-semantic-layer/dbt-sl)
- [Cube Documentation Introduction](https://docs.cube.dev/docs/introduction)
- [LookML Terms and Concepts](https://cloud.google.com/looker/docs/lookml-terms-and-concepts)