Data Contract Definition AI Prompt

Write a data contract for the dataset: {{dataset_name}} produced by the {{producer_team}} team and consumed by {{consumer_teams}}.

A data contract is a formal agreement between data producers and consumers specifying what data will be delivered, in what format, with what quality guarantees, and on what schedule.

1. Dataset identity:
   - Dataset name and version
   - Producer: team, contact, and escalation path
   - Consumers: teams currently depending on this dataset

2. Schema definition:
   - Table or topic name
   - For each column/field: name, data type, nullable (Y/N), description, example value, PII classification (Y/N)
   - Primary key or unique identifier
   - Partitioning columns

3. Semantics and business rules:
   - Grain: what does one row represent?
   - Business rules: constraints and derived logic (e.g. 'order_total is always the sum of line items')
   - Key relationships to other datasets

4. Quality commitments:
   - Completeness: which columns are guaranteed non-null?
   - Uniqueness: which column combinations are guaranteed unique?
   - Freshness: data will be available by {{sla_time}} on each {{frequency}}
   - Accuracy: key measures are reconciled to source within {{tolerance}}

5. Change management:
   - Breaking change definition: removed column, type change, semantic change
   - Notice period: {{notice_period}} days notice required before a breaking change
   - Deprecation process: how will consumers be notified and given time to migrate?

6. SLA and support:
   - Incident response time: {{response_time}}
   - Scheduled maintenance window: {{maintenance_window}}
   - Where to report issues: {{issue_channel}}

Return: complete data contract document in YAML format.