data-analytics
Route durable data modeling, analytics engineering, and data-science analysis; not for model-training systems or generic application architecture.
SKILL.mdSKILLS/data-analytics/SKILL.md
GitHubData And Analytics
Choose one primary mode and read only its workflow. Add a second mode only when the requested deliverable crosses its boundary.
| Mode | Use when | Workflow |
|---|---|---|
data-modeling |
Design entities, schemas, ownership, indexes, retention, migrations, backfills, or consistency. | Data modeling |
analytics-engineering |
Define metric contracts, semantic layers, BI transformations, freshness, or dashboard validation. | Analytics engineering |
data-science-workflows |
Frame analyses or experiments and assess provenance, leakage, uncertainty, and reproducibility. | Data-science workflows |
Do not load sibling workflows or artifacts prospectively. Preserve grain, lineage, provenance, uncertainty, and owner-controlled migration or release decisions.