Compare the bound model against another already-processed model directory: added/removed/modified measures, columns, relationships, plus impact analysis (which measures reference each removed/modified measure or column). Use for 'what changed and what might break' after editing a model. Both directories must already be processed by pbi-context.

Impact analysis: which measures reference this column in their DAX expression. One level deep by default. Use for 'what breaks if I remove this column' style questions.

Impact analysis: which other measures reference this one in their DAX expression. One level deep by default. Use for 'what breaks if I change this measure' style questions.

Get one measure's full record (including DAX) from a known table. Cheaper than get_table when you already know the table and measure name.

What a measure references: other measures and columns, parsed from its DAX expression. One level deep by default. Use for 'what does this measure depend on' style questions.

List relationships, optionally filtered to those touching one table.

Get one table's full detail: columns and measures (with DAX). Use list_tables first if you don't know the exact table name.

List tables in the model with hidden/technical flags, column/measure counts, partition counts, and measure categories present. Optionally filter.

Find columns by a case-insensitive name substring across every table — use for 'which table has a column called X' style questions.

Find measures by a case-insensitive name substring across every table — use for 'which tables have a revenue measure' style questions.