YAML check format

CatalystData's check definitions have a plain YAML shape — the same shape whether a check is built in the UI, embedded in a data contract's checks definition, or drafted in your own repository for review before you add it. Keeping a YAML copy of your check definitions alongside the pipeline or dbt project they cover, even if you then add them through the UI, keeps them reviewed and versioned alongside the code that produces the data.

Format

checks:
  - name: orders_no_null_id
    check_type: null
    table_name: sales.orders
    column_name: order_id

  - name: orders_row_count_sane
    check_type: row_count
    table_name: sales.orders
    parameters:
      min_count: 100
      max_count: 5000000

  - name: orders_updated_recently
    check_type: date_freshness
    table_name: sales.orders
    column_name: updated_at
    parameters:
      max_age_hours: 6
    schedule_cron: "0 * * * *"
    is_active: true

Fields

Field	Required	Description
name	Yes	A unique, human-readable name for this check
check_type	Yes	One of the check types in the check reference
table_name	Yes	The table this check runs against
column_name	No	Required for column-level check types; omit for table-level ones
parameters	No	Check-specific parameters — see each check type's page for what it accepts
schedule_cron	No	A cron expression if this check should run on its own schedule
is_active	No	Defaults to true