yigityildiz0 
Data
Start with the question and data grain, then select the narrowest workflow:
data-analysisfor exploration, statistics, trend/metric diagnosis, and quality-aware findings;sql-analyticsfor analytical SQL and query review;data-visualizationfor charts and quantitative figures;data-dashboardfor decision-oriented KPI surfaces;data-contextfor schema, metric, lineage, and semantic documentation.
When the active host provides a managed analytics capability, use it rather than duplicating its connector-specific instructions. Otherwise remain host-neutral, preserve data provenance, surface missingness/uncertainty, and never turn correlation into a causal claim.