# dbt MCP
> dbt's MCP server gives AI clients governed access to dbt projects, models, metrics, lineage, jobs, and data queries. The self-hosted and dbt-hosted remote versions expose different tools and require different setup and authentication.
Canonical: https://www.mcpfolio.com/servers/dbt
Last updated: 2026-10-28
## Catalog record

- Publisher: dbt Labs
- Categories: search

## Website

https://docs.getdbt.com/docs/dbt-ai/about-mcp

## Repository

https://github.com/dbt-labs/dbt-mcp

## Transports

stdio (self-hosted); Streamable HTTP (remote)

## Authentication

Remote OAuth (eligible plans and OAuth-capable client), or token headers using a dbt PAT/service token. Self-hosted may use OAuth or local dbt credentials depending on mode.

## Tools

- **execute_sql**: Execute SQL on dbt platform infrastructure with Semantic Layer support.
- **text_to_sql**: Generate SQL from natural language using project context.
- **list_metrics**: Retrieve defined Semantic Layer metrics.
- **query_metrics**: Query metrics with filters and grouping.
- **get_all_models**: Retrieve model names and descriptions.
- **get_all_sources**: List sources and freshness information.
- **get_lineage**: Retrieve a resource lineage graph.
- **get_node_details**: Retrieve details for a dbt resource.
- **get_model_health**: Get run status, test results, and upstream source freshness.
- **run**: Execute dbt models.
- **build**: Execute models, tests, snapshots, and seeds in DAG order.
- **test**: Run dbt tests.
- **list_jobs**: List dbt platform jobs.
- **trigger_job_run**: Trigger a dbt platform job run.
- **search_product_docs**: Search official dbt product documentation.
- **get_product_doc_pages**: Fetch full dbt documentation pages.

## Sources

- [official_tool_reference](https://docs.getdbt.com/docs/dbt-ai/about-mcp)
- [official_documentation_or_source](https://github.com/dbt-labs/dbt-mcp)

## Setup

Use the interactive setup panel on [the detail page](https://www.mcpfolio.com/servers/dbt) to view instructions for each supported client.