Your personal dbt agent, available wherever you work.

Install the dbt Wizard CLI

Prefer a fully hosted experience? dbt platform includes Wizard plus Studio, Catalog, and Insights. Book a demo →

$ curl -fsSL https://public.cdn.getdbt.com/dbt-wizard/install/install-wizard.sh | sh

The AI agent built for the way analytics engineers work

Generic coding agents don't understand your data. dbt Wizard is grounded in it: lineage, grain, compiled state, tests, and metric definitions. Available wherever your team or you develop.

Knows your data, not just your code

Natively understands your full dbt project so every change is with full context on what's upstream, downstream, and at stake.

Self-validates before you review

An opinionated workflow makes validation the default, so problems get caught before they ship, not after. Fewer production incidents, less time fixing what AI got wrong.

Governed and auditable by default

Every change is auditable and nothing ships without review. Keep a human in the loop to always know what changed, why, and who approved it.

Always calls the right tool

With high-precision tool orchestration, Wizard knows exactly which dbt-specific call to make and when

No context hunting. Less token waste.

dbt's native metadata engine retrieves exactly what the agent needs before it writes a line, so you get more accurate output while optimizing token spend.

Zero setup. Zero maintenance.

No MCP infrastructure to configure, no DIY system to maintain. Just connect and start shipping.

One agent. Available wherever data work happens.

dbt Wizard works where you do. In the dbt platform or in the terminal, it's grounded in your dbt project from the first prompt.

dbt Wizard in platform

The AI agent layer across the dbt platform. Available as a dedicated workspace (public beta) and embedded in dbt Studio (preview) for inline development. Experience tailored agent capabilities in dbt Canvas and dbt Insights with more coming soon.

dbt Wizard CLI

A terminal-native agent for analytics engineers developing locally, whether you're a dbt platform customer or self-hosted. Everything Wizard knows about your project travels with you to the CLI — so you get accurate, context-aware help wherever you actually build.

Ship your best dbt work without leaving your workflow

Refactor, build, investigate, migrate and ship. All grounded in your dbt project, all in one place.

Refactor

Rename, restructure, and clean up —  confidently

Describe your change and Wizard updates every file that needs to move...and shows its work. Rename a model and the refs follow. More efficient refactoring, without the risk.

Build

Generate a new model without starting from scratch

Describe what you need. Get back a validated model, tests, docs, and semantic definitions. All at once. All grounded in your existing project. All available with dbt Wizard.

Investigate

Find the root cause before it reaches production

Tell Wizard what broke and it traces your lineage, proposes a fix, and validates it before surfacing an edit for review. Investigation through to validation with dbt Wizard. Less time debugging, more reliable pipelines.

More features

One agent. Every part of the job.

From migration to analysis to upgrades, dbt Wizard handles the work that slows teams down.

Move models without breaking dependencies

Describe what you're migrating and Wizard handles every ref, test, and YAML config — with a reviewable diff before anything saves.

Get answers grounded in your dbt data

Ask in plain language, get answers rooted in your defined metrics — traceable back to the models your team already trusts.

Fix Fusion conformance errors, automatically

Wizard reads your project, applies high-confidence fixes, and flags what needs your input. Get to Fusion faster with less manual effort.

  • Before dbt Wizard, our engineers were spending more time correcting AI output than they were writing models. Now the agent actually knows our project. It gets the joins right, it respects our contracts, and it doesn't break things downstream. We've seen a 15-20% reduction in production incidents since we rolled it out.

    Erion Krasniqi Junior Data Scientist at Endress+Hauser InfoServ

  • What really sets the dbt developer agent apart is its precision in identifying and resolving bottlenecks. It doesn't just suggest code; it understands our existing tests and lineage well enough to troubleshoot issues almost instantly. This has significantly reduced our build times and allowed us to scale our dbt project with total confidence.

    Vishal Kaviraj AVP Data Architect

  • We went from about 60 conformance errors to 7, using the Fusion migration agent. That's the difference between too hard and actually doable.

    Michael Fridolfsson Data Architect at Brighte