AgentSkillsCN

great-expectations

运用Great Expectations进行数据质量验证——定义期望值、运行验证套件，并审查验证结果。

SKILL.md
--- frontmatter
name: great-expectations
description: >
  Data quality validation with Great Expectations — define expectations,
  run validation suites, and review results.
metadata:
  openclaw:
    requires:
      bins: [clawdata]
    primaryEnv: GE_HOME
    tags: [quality, validation, testing, expectations, data-quality]

Great Expectations

Data quality validation as a standalone skill. Define expectations for your data, run validation suites, and review results.

Commands

TaskCommand
Initialise GE projectclawdata ge init
Create expectation suiteclawdata ge suite create <name>
Run validationclawdata ge validate <suite> --table <table>
List suitesclawdata ge suite list
View resultsclawdata ge results
Generate docsclawdata ge docs

Example Expectations

yaml
# expectations/orders_suite.yml
expectations:
  - expect_column_to_exist:
      column: order_id
  - expect_column_values_to_not_be_null:
      column: order_id
  - expect_column_values_to_be_between:
      column: amount
      min_value: 0
      max_value: 100000
  - expect_table_row_count_to_be_between:
      min_value: 1
      max_value: 1000000

When to use

  • •User needs data validation beyond dbt tests → clawdata ge validate
  • •User wants to profile data quality trends → clawdata ge results
  • •User asks about data quality → run clawdata ge validate then show results

Integration with dbt

Great Expectations can validate the output of dbt models:

  1. •clawdata dbt run — transform data
  2. •clawdata ge validate orders_suite --table fct_orders — validate results
  3. •Report results in CI pipeline