Build a data-quality checklist

Define measurable checks suited to a particular dataset.

What this starting point does

Create a data-quality checklist for the described dataset and its use. This is an editable instruction template, not a claim of tested performance on a particular model.

Make the inputs yours

• Reference material — fill {{MATERIAL}} with your own material, preserving line breaks where useful. This input is required for a filled prompt.

Reusable template

## Task
Create a data-quality checklist for the described dataset and its use.

## Constraints
- Propose measurable checks rather than generic quality claims.
- Mark unconfirmed thresholds as decisions.
- Include completeness, consistency, and uniqueness where relevant.

## Output format
Check, rationale, measurement, proposed threshold, and response to failure

## Reference material
<source_material>
{{MATERIAL}}
</source_material>

A filled example

This example uses fictional demonstration material. It is an example prompt, not a generated answer or image.

## Task
Create a data-quality checklist for the described dataset and its use.

## Constraints
- Propose measurable checks rather than generic quality claims.
- Mark unconfirmed thresholds as decisions.
- Include completeness, consistency, and uniqueness where relevant.

## Output format
Check, rationale, measurement, proposed threshold, and response to failure

## Reference material
<source_material>
An orders export has order_id, customer_id, placed_at, and total. It supports monthly reporting; expected currency and completeness thresholds need confirmation.
</source_material>

Before using it

Review every instruction and replace the example material. Check that its constraints fit your task. The structural Analyzer can flag specific issues, but cannot verify facts or guarantee the quality of a model’s response.