Power User for dbt

Makes VS Code work seamlessly with dbt™ core, dbt™ cloud, and dbt™ Fusion. Autocomplete, column lineage, AI documentation, query preview, health checks, and more - without leaving the editor.

724,927VS Code + Open VSX installs
5.0★VS Code Marketplace rating
dbt 1.6+Core, Cloud and Fusion supported
What's included

Everything across develop, test, and collaborate

Core features are free with no account. Add a free Altimate AI key to unlock AI-powered features.

Develop

  • ▸Autocomplete model, macro, source names + go to definition
  • ▸Compiled SQL preview as you write
  • ▸Click to run parent / child models and tests
  • ▸Generate and edit dbt models with natural language
  • ▸Visualize compiled SQL as an interactive node graphbeta

Test

  • ▸Column-level and model-level lineage
  • ▸Defer to prod - run only changed models
  • ▸SQL validation without execution
  • ▸Query preview, CTE profiling, and CSV export

Collaborate

  • ▸AI documentation generation for models and columns
  • ▸Documentation editor - write and save to YAML
  • ▸Project governance in IDE, Git, and CI/CD
  • ▸Code collaboration and query bookmarks
Column lineage

Model lineage and column lineage with full dbt entity visibility

View lineage across sources, seeds, models, tests, metrics, and exposures - including model types and documentation status. Click into any model to see its columns, tests, and linkage types. Column lineage traces individual columns through your entire dbt project.

  • ▸Model-level lineage - free, no API key required
  • ▸Column-level lineage - requires Altimate AI key
  • ▸Shows linkage types: unchanged, aliased, transformed
  • ▸Expandable depth - adjust scope without reopening
  • ▸Export lineage for audits and downstream tooling
Lineage - fct_orders.amount_usd
raw_ordersamount
──▶
stg_ordersamount_usd [transformed]
──▶
fct_ordersamount_usd

Downstream

mart_revenue / monthly_total [aggregated]
mart_customer_ltv / lifetime_value [aggregated]
Tests: not_null, uniqueDocumented ✓
Autocomplete + navigation

Auto-complete dbt code and click to go to definitions

Auto-fill model names, macros, sources, and docs as you type. Click on any model name, macro, or source reference to jump directly to its definition - across your entire dbt project, including dbt Mesh cross-project refs.

  • ▸Autocomplete for models, macros, sources, columns, and docs
  • ▸Go-to-definition - click any ref() or source() to navigate
  • ▸Jinja-SQL aware - understands dbt template syntax
  • ▸Works across dbt Core and dbt Cloud projects
  • ▸Supports dbt Mesh and multi-project workspaces
fct_orders.sql
WITH orders AS (
  SELECT * FROM {{ ref('stg_orders') }}
Suggestions
stg_orders model
stg_customers model
stg_payments model
),
⌘ Click to go to definition → staging/stg_orders.sql
Query preview

Preview results, profile CTEs, and export - without leaving the editor

See your query results instantly without leaving VS Code. Debug complex models CTE by CTE, export to CSV, and analyze data inline with graphs and filters - no context switching, no copy-pasting to a SQL client.

  • ▸Preview results for your full model or just a selection
  • ▸Debug step by step - preview individual CTEs separately
  • ▸Export as CSV or analyze inline with graphs and filters
Query Results - fct_orders.sql
ResultsSQLGraph
order_idcustomer_idamount_usdstatus
1001C_421142.50completed
1002C_08988.00completed
1003C_312210.75pending
3 of 2,847 rows · 0.8sExport CSV ↓
Documentation

Generate model and column descriptions, saved directly to YAML

Generate documentation for models and columns automatically, or write descriptions manually in the built-in UI editor. Descriptions are formatted and saved to YAML files. Hover over a column name in schema.yml to get instant AI suggestions for descriptions and tests.

  • ▸AI-generated model and column descriptions
  • ▸Hover shortcut - hover a column name: in YAML for instant AI actions
  • ▸Manual editor - write and save descriptions in the UI
  • ▸Auto-saved and formatted into YAML files
  • ▸Code and documentation collaboration without creating a PR
schema.yml - Documentation Editor
models:
  - name: stg_orders
    columns:
      - name: order_id
(column) order_id
✏ Suggest description🧪 Suggest tests
        tests: [not_null, unique]
Governance

Catch issues before they reach production

Project health check

Scan your entire dbt project for issues - columns not present, models not materialized, missing tests, and other best-practice violations. Available in the IDE, via Python package, and in CI/CD pipelines.

SQL validation without execution

Identify SQL issues - non-existent columns, keyword typos, missing or extra parentheses - before running a query. Every error popup includes a quick-fix shortcut to resolve the issue.

Defer to prod

Build only the models you changed by deferring upstream models to production state. Works with dbt Core via the extension (no dbt Cloud required) using a local or SaaS-stored manifest file.

BigQuery cost estimator

Estimate how much data will be processed by a dbt model in BigQuery - directly inside VS Code, before you run it. Requires dbt version 1.6 or above.

SQL Visualizer

Translate complex SQL queries into interactive node graphs - CTEs, joins, filters, and unions visualised as connected components. Drag-and-drop to restructure and debug query logic. beta

Query history and bookmarks

View the history of all SQL queries and dbt models you ran. Re-run to compare results across code changes. Bookmark and tag useful queries, and share bookmarks across your team.

Security

SOC 2 Type 2 certified

Altimate AI is SOC 2 Type 2 certified. Data is encrypted in transit via TLS. Infrastructure runs on AWS in a private VPC with RBAC access controls. Telemetry can be disabled entirely via VS Code settings.

SOC 2 TYPE 2
Certified
TLS
All data in transit encrypted
AWS VPC
Private cloud infrastructure
MIT
Open source license
From the community

What analytics engineers say

► JAMES HOPGOOD
█
— James HopgoodVS CODE MARKETPLACE · FEB 2025
▼
Common questions

Frequently asked

Join the Community

dbt Power User is built in the open. Get involved.

Get started

Install in VS Code.

Free to install. No account required for core features.

Open source · MITdbt core · cloud · fusionSOC 2 Type 2Free to install