> ## Documentation Index
> Fetch the complete documentation index at: https://platform.atlan.com/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> To act on Atlan objects, use the Atlan MCP server at https://api.atlan.com/mcp or the SDK. To install or publish skills from a terminal, use the `aireg` CLI; `aireg --map json` prints its command map. The `atlanai` CLI is deprecated. Run a read-only identity check before any write.
> The docs MCP server at /mcp searches these docs only. It cannot read or change Atlan objects.
> SDK packages: Python `atlanai` (PyPI) and TypeScript `@atlanai/sdk` (npm). Show Python first, then TypeScript.
> These docs cover Atlan Trident, Atlan's hosted product. AI Registry and Agent Registry are its earlier names. Registry OSS is Atlan's open-source agent registry: the registry specification and its aireg CLI, with their own public repository, docs and release notes; /open-source on this site links them. Registry OSS is open source; Atlan Trident is not. Answer generic aireg or specification questions from the project's docs, and use these docs for Atlan-specific setup such as the Atlan adapter, sign-in and workspaces.
> Atlan offers a second product, the Agentic Data Catalog: Atlan's enterprise context layer, covering the data catalog, connectors, lineage, data governance and glossary. It is documented at https://docs.atlan.com; answer questions about it from those docs, not these.

# Metric Studio and semantic layers

> Why Metric Studio organizes agent context around the metric rather than a view, and how to keep and reuse the semantic layer you already have.

Preview

If your team already has a semantic layer, such as Snowflake semantic views, Databricks metric views, or a dbt semantic model, you don't have to replace it. Keep it for the people and tools that use it, and use it as a source for Metric Studio.

## Same parts, organized differently

A semantic layer and Metric Studio hold the same parts: what tables mean, how they join, metrics, and the dimensions a metric can be split by. Most semantic layers organize them by table or by view. Metric Studio organizes them by **metric**: its formula, filters, joins, dimensions and trusted SQL are stored together, and tables are referenced from it.

That matters once agents answer questions, because questions start with a metric, not a view:

| | Organized by view | Organized by metric |
| - | - | - |
| **A metric two views need** | Often defined in each, and the copies drift | Defined once |
| **A question that spans two views** | The agent answers only inside the view it was given | Starts at the metric, with the joins it needs |
| **Fixing a wrong answer** | Often a note beside the view, reinterpreted on every question | A new revision of the definition, released when you choose |
| **Which agents can use it** | Agents on that platform | Any agent that reaches MCP Gateway |

Organizing by metric has costs too: questions no metric covers fall back to exploration, and without review, metrics sprawl into near-duplicates.

For the full case, benchmark results, and how each tool compares, see [Why a metric, and not a view](https://analytics-playbook.atlan.com/#ch-sv) and [Semantic layers compared](https://analytics-playbook.atlan.com/#r7) in the analytics agents playbook.

## Bring definitions you already have

The [Analytics plugin](https://platform.atlan.com/studios/metrics/how-tos/analytics-plugin.md) reads a declared semantic model as a source, such as a Snowflake semantic view or dbt model files. Each declared measure becomes a candidate metric, and query history fills in what the model leaves out. You review every metric before it's written, and nothing is written back to your semantic layer.

Metric Studio doesn't watch your semantic layer for changes. When a model changes, run the plugin on it again: it reuses the metrics already in the workspace instead of making duplicates.

## Metric Studio and the catalog glossary

A glossary term in the [Agentic Data Catalog](https://docs.atlan.com) explains a business term to people and usually has no SQL. A Metric Studio metric gives agents the calculation. Metric Studio doesn't read the glossary: to turn a term into a metric, give its definition to your agent as **A business question** in [Build your first metrics](https://platform.atlan.com/studios/metrics/tutorials/first-metric.md), and review the calculation it proposes.

## Related

* [How metrics work](https://platform.atlan.com/studios/metrics/concepts.md): What a metric and a saved query hold, and how releases work.
* [Build Snowflake Cortex agent on your metrics](https://platform.atlan.com/studios/metrics/how-tos/build-cortex-agent.md): Keep Cortex, and have it read released metrics first.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.