
Loading...
Connect DataChonk to dbt Cloud to trigger runs, monitor status, and close the loop between building models and running them in production.
After pushing your DataChonk-generated code to Git, you can optionally trigger a dbt Cloud run to validate and materialize your models. This creates a seamless build-push-run workflow.
Git repository required
You need two pieces of information to connect DataChonk to dbt Cloud: your Account ID and an API token.
Your Account ID is in the URL when you're logged into dbt Cloud:
https://cloud.getdbt.com/accounts/12345/projects/...
^^^^^
This is your Account IDClick 'Create Service Token' and give it a descriptive name like 'DataChonk Integration'.
Select the following permissions:
Keep your token secure
After pushing code to Git, you'll see the dbt Cloud connection step:
Configure dbt Cloud in your project's datachonk.yaml or use environment variables:
deploy:
dbt_cloud:
account_id: 12345
project_id: 67890
job_id: 11111
# Optional: auto-trigger after push
auto_trigger: trueOr use environment variables:
export DBT_CLOUD_API_TOKEN="dbtc_your_token_here"
export DBT_CLOUD_ACCOUNT_ID="12345"
export DBT_CLOUD_PROJECT_ID="67890"
export DBT_CLOUD_JOB_ID="11111"
datachonk push --trigger-dbt-runOnce connected, DataChonk can trigger dbt Cloud jobs and monitor their status.
After selecting a job, click Trigger Run. DataChonk will:
# Push and trigger run
datachonk push --trigger-dbt-run
# Trigger run without pushing (if code is already in Git)
datachonk dbt-run --job-id 11111
# Trigger run with custom cause message
datachonk push --trigger-dbt-run --cause "New customer dimension model"
# Wait for run to complete
datachonk push --trigger-dbt-run --wait| Status | Meaning | Action |
|---|---|---|
| Queued | Waiting for available runner | Wait for processing |
| Running | Job is executing | Monitor progress |
| Success | All steps completed successfully | Models are live! |
| Error | One or more steps failed | Check logs in dbt Cloud |
| Cancelled | Run was manually cancelled | Re-trigger if needed |
Set up your dbt Cloud jobs to work well with DataChonk's workflow.
Create a dedicated job for validating new models without running the entire project:
commands:
- dbt build --select state:modified+ --defer --state ./prod-manifestMake sure your job has "Triggers: API" enabled in the job settings. This allows DataChonk to start runs programmatically.
Consider separate jobs for development/staging vs production environments. Trigger dev jobs for quick validation, production jobs for final deployment.
If you're integrating programmatically, here are the endpoints DataChonk uses.
| Property | Type | Default | Description |
|---|---|---|---|
| POST /api/dbt-cloud/connect* | JSON | - | Connect to dbt Cloud with API token and account ID |
| GET /api/dbt-cloud/status | - | - | Check connection status and list projects |
| GET /api/dbt-cloud/projects/:id/jobs | - | - | List jobs for a specific project |
| POST /api/dbt-cloud/trigger-run* | JSON | - | Trigger a job run with optional metadata |
| GET /api/dbt-cloud/runs/:id | - | - | Get run status by run ID |