Data Governance Skill

Implements data governance as a layer inside engineering work -- cataloging, lineage, classification, access control, and compliance. Governance is code, not overhead.

When to Use This Skill

Activate when: Cataloging data assets (dbt docs or external tools), documenting data lineage, classifying sensitive data (PII/PHI), implementing access control patterns (RBAC, row-level security), or applying compliance frameworks (GDPR, HIPAA, SOX, CCPA) to data pipelines.

Don't use for:

• Writing or modifying dbt models → use dbt-transforms
• Pipeline orchestration or scheduling → use data-pipelines
• Data quality testing (freshness, uniqueness, accepted values) → use data-testing
• Data observability and monitoring → use data-observability

Scope Constraints

• Engineering governance only -- does not provide legal counsel or regulatory interpretation.
• Generates governance artifacts (YAML configs, SQL grants, documentation) -- does not execute DDL against production databases.
• Security tier default: Tier 1 (schema/metadata only). User must explicitly elevate to Tier 2 (sample data) or Tier 3 (full data access).