@kapa.ai here's an example column_anomalies test: data_tests: - elementary.column_anomalies: arguments: column_anomalies: - average if i don't specify the column_anomalies test type, and only put it like this: data_tests: - elementary.column_anomalies will it work? will it run all the tests such as average, sum, not null, etc. by default?

Yes, it will work. If you don't specify

column_anomalies

, Elementary will automatically run all default monitors applicable to the column's data type. [column_anomalies test] Here's what gets run by default depending on column type: For any column type: •

null_count

•

null_percent

For string columns: •

min_length

,

max_length

,

average_length

•

missing_count

,

missing_percent

For numeric columns: •

min

,

max

,

average

•

zero_count

,

zero_percent

•

standard_deviation

,

variance

Note that

sum

is not included by default — it's an opt-in monitor that must be explicitly specified. [column_anomalies config] So your simplified config:

data_tests:
  - elementary.column_anomalies

is valid and will run all default monitors for that column's type. If you want to limit which monitors run (e.g., only

average

), you should explicitly specify them under

column_anomalies

as in your first example. 👍 0 button 👎 0 button ℹ️ Tip: Mention @elementary.ai in the thread for follow-ups or DM the app for 1-to-1 help.

@kapa.ai is it possible to add a list of columns that use this test instead of adding this test under each column in yml?

Unfortunately, there is no built-in way to apply

elementary.column_anomalies

to a list of specific columns in a single block. The dbt hierarchy requires tests to be defined per column (

model -> column -> test

), so you must repeat the test definition under each column. [Slack discussion] However, here are your options: 1. all_columns_anomalies (closest to DRY) If you want to monitor all columns, use

elementary.all_columns_anomalies

at the model level, with optional exclusions:

models:
  - name: your_table_name
    config:
      elementary:
        timestamp_column: your_timestamp_col
    data_tests:
      - elementary.all_columns_anomalies:
          arguments:
            exclude_regexp: "^(col_to_exclude)$"

This applies the same configuration to all columns, but doesn't allow per-column granularity. [all_columns_anomalies doc] 2. YAML Anchors (reduces repetition) You can use YAML anchors to avoid copy-pasting the test config:

definitions:
  column_test: &column_anomaly_test
    elementary.column_anomalies:
      arguments:
        column_anomalies:
          - average
          - null_count

models:
  - name: your_table_name
    columns:
      - name: col1
        data_tests:
          - *column_anomaly_test
      - name: col2
        data_tests:
          - *column_anomaly_test

3. Elementary Cloud On Elementary Cloud, you can add tests in bulk directly from the UI, which automatically opens a pull request to your repository. [Data Tests Overview] Summary: For a subset of specific columns, YAML anchors are the most practical OSS workaround. For all (or nearly all) columns,

all_columns_anomalies

with exclusions is the cleanest approach. 👍 0 button 👎 0 button ℹ️ Tip: Mention @elementary.ai in the thread for follow-ups or DM the app for 1-to-1 help.