explore-data

To perform a dataset overview, the skill lists all tables with row counts, date ranges, and key entity connections. This establishes context and reveals table structures so you can quickly understand what data is available before formal analysis.

Table exploration checks data quality by showing column types, null rates, sample rows, and basic quality indicators. It specifically flags issues like high null rates or low cardinality to guide your data cleaning and validation process.

For deeper exploration, use the column deep-dive guidance to profile columns. It suggests related columns for cross-analysis, supporting deeper investigations by highlighting relationships and profiling needs within your dataset.

Yes, you can prepare for analysis and modeling by producing structured notes and follow-ups. The skill applies standardized discovery steps across overviews and profiling to load workspace context and output structured documentation.

Data discovery works without dependencies to assess dataset suitability. It evaluates tables by showing column lists, types, null rates, and sample rows, providing basic quality indicators to determine if the data fits your needs.