EN
EnglishEspañolFrançaisPortuguêsDeutsch
Sign in Start free
Pricing
Product How it worksLive demoChrome side panelAI connector (MCP)IntegrationsHubSpotSalesforceSlackJiraTrust CenterSecurityAI and your dataFree SOP generatorAI diagram generatorAI process map generatorMeeting notes to flowchartFlowchart makerVoice to diagram
Solutions Process documentationDiscovery callsWorkshops and live sessionsSOP creationData flow diagramsAutomation designProcess improvementITEngineeringSales and SEOperationsProduct and UXCustomer successFinancial servicesHealthcareConsultingSoftware and SaaSManufacturingPublic sectorView all solutions
Resources Resource centerTutorialsFlow diagram examplesTranscript to flowchartProcess mapping workshopsWho it is forTemplate librarySOP templateProcess map templateFlowchart templateSwimlane process mapWavelength vs LucidchartWavelength vs VisioWavelength vs MiroAll comparisons

Sign in

Integration · Data and lineage

dbt integration

Turn the lineage of a dbt project into a flow you can edit and share. Wavelength draws the sources, the models, and the exposures, with their owners.

Start free Try the live demo

Works with dbt Cloud and with manifest.json from any dbt project, including dbt Core. No AI call.

What you can do

Map the lineage of a model

Pick models and exposures. Turn on "Also include everything downstream" to add what depends on them.

Map a whole project

Pick "Whole project" to draw every source, model, and exposure in one flow.

Show who owns each model

Owners come from model meta.owner and from the exposure owner. They show on the nodes.

See it in action

Watch a short demo of a dbt lineage import.

Wavelength reads the dbt lineage and draws it as a flow.

How it works

  1. Click Automations in the transcript panel and open the dbt tab.
  2. Pick dbt Cloud and click Connect with dbt. Sign in on your access URL and approve read access. A service token also works.
  3. Or pick Upload manifest.json and upload target/manifest.json from your project. Run dbt parse or dbt docs generate first.
  4. For dbt Cloud, pick an environment, or read from a job instead.
  5. Pick models and exposures, or the whole project. Then click Map lineage.

Wavelength reads the dbt Discovery API for the environment or the job you pick. Sources, seeds, and snapshots become data nodes. Models become steps. Exposures become systems.

A flow holds up to 250 nodes. For a bigger project, pick models instead of the whole project.

You can also import from Claude, ChatGPT, or another AI assistant with the Wavelength connector. The tools are list_dbt_jobs and import_dbt_lineage. See Wavelength MCP.

Worked example

An analytics engineer maps a project with 6 models, 3 sources, and 1 exposure.

What lands in Wavelength

  • A lineage flow with 10 nodes.
  • The 3 sources as data nodes and the 6 models as steps.
  • The exposure as a system at the end of the flow.
  • The owners Finance and Sales Ops on the models.

Who uses this

Security and permissions

dbt Cloud sign-in uses OAuth on your account access URL with the scopes account:read, catalog:read, and offline_access. Wavelength reads model lineage from the dbt Discovery API. It does not read your data.

A service token is the other option. Use a token with Read-Only or Metadata Only access. Wavelength stores the token encrypted. Disconnect at any time to delete it.

A manifest.json upload stays in your browser. The import makes no AI call.

The Trust Center and the subprocessor list show how Wavelength handles your data.

Common questions

Does Wavelength work with dbt?

Yes. Wavelength reads the lineage of a dbt project and draws it as an editable flow. Sources, seeds, and snapshots become data nodes, models become steps, and exposures become systems.

Does it work with dbt Core?

Yes. Upload target/manifest.json from any dbt project, including dbt Core. The file stays in your browser.

Which dbt Cloud plans work?

Sign in with dbt works on the Starter, Enterprise, and Enterprise+ plans. A service token is the other option.

Where do the owners come from?

From model meta.owner and from the exposure owner in your project.

How big can the flow be?

Up to 250 nodes. For a bigger project, pick models instead of the whole project.