Skip to main content

dbt Agents overview BetaEnterpriseEnterprise +

dbt Agents, available on dbt Enterprise-tier plans, are a suite of native AI agents that turn structured dbt context into auditable actions. These agents help you build, manage, and consume governed data at scale by bringing intelligence to every step of the analytics development lifecycle.

info

Some dbt Agents are in beta; others are coming soon. Contact your account manager for early access.

See available agents to find out what's available.

dbt Agents are built on top of dbt's structured context to provide accurate, auditable, and governed results:

  • Semantic Layer — Metrics, dimensions, and business logic
  • Metadata — Lineage, tests, documentation, and ownership
  • Governance — Access policies, data quality rules, and contracts

Having dbt as the standard context layer for agentic analytics means that dbt Agents are built on top of this context to provide accurate results rather than hallucinated or inconsistent answers.

Key benefits​

  • Faster development — Engineers and analysts ship data products faster with AI assistance.
  • Better decisions — Business users get accurate answers grounded in governed data.
  • Auditability — Every agent action includes transparent SQL, lineage, and policies.
  • Scalability — Routine tasks are automated so teams can focus on high-value work.

Available agents​

dbt offers several specialized agents, each designed for specific workflows in the analytics lifecycle to help you scale your data teams across the dbt platform.

The following agents are available. Contact your account manager for early access to agents that are in beta or coming soon.

Analyst agent Beta​

Use Copilot to analyze your data and get contextualized results in real time by asking natural language questions to the Insights Analyst agent.

Chat with your data, get accurate answers powered by the dbt Semantic Layer. That means consistent, explainable results with transparent SQL, lineage, and policies.

The Analyst agent is a beta feature. Enable beta features under Account settings > Personal profile > Experimental features. For more information, see Preview new dbt platform features.

Developer agent BetaEnterpriseEnterprise +​

The Developer agent is the next evolution of Copilot in the Studio IDE, purpose-built to streamline the developer experience. Describe the data product or change you want — the agent writes or refactors models, validates with dbt Fusion engine, and runs against your warehouse with full context. Stay in flow in the Studio IDE.

The agent always has access to the latest dbt-recommended guidance through dbt Agent Skills — curated instructions and scripts managed by dbt Labs, available out of the box with no configuration required.

For setup instructions and available use cases, see Developer agent.

Discovery agent Private beta​

Find the right, approved dataset fast in Catalog. The Discovery agent surfaces definitions, freshness, tests, owners, and lineage right where you work.

To request access to the Discovery agent, contact your account manager.

Observability agent Coming soon​

The Observability agent autonomously and continuously monitors pipelines, flags likely root causes in context, and guides fixes — resulting in faster mean time to resolution, higher reliability, and streamlined ticket queues. No more digging through logs.

dbt MCP server​

Build your own custom agents and copilots with the local or remote dbt MCP server. The Model Context Protocol (MCP) makes dbt's structured context available to any AI system.

Was this page helpful?

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

0
Loading