Hashing or masking columns before sharing.

Whether keys in one table exist in another.

Testing data against rules you state.

Whether two columns are shaped alike.

Column-level differences between two sources.

Convert data between formats.

How strongly numeric columns move together.

How many rows, exactly, without reading them.

Whether values have shifted between two versions.

Running SQL against a live database connection.

Reading the foreign keys a database declares.

What a file is, before opening it properly.

Values far outside the usual range.

Spotting personal data before sharing a file.

Explaining why a join lost or multiplied rows.

Row-level differences between two tables.

Removing duplicate rows into a new file.

Load more tools for the rest of this conversation. The tools listed below are available but not yet loaded, so their parameters are not visible to you yet. When the job needs one of them, call this with that group's name and it becomes callable straight away - do NOT tell the user a capability is missing without fetching its group first. Ask for the one group you need, not for all of them.

Writing a schema out to reuse or check against.

Filling gaps in a column.

Exact duplicate rows or keys.

Near-duplicates: typos and spelling variants.

Joining on names that nearly, but not exactly, match.

Searching many files at once for a pattern.

Making differing column names line up.

Joining tables on a key, written to a file.

Working with your saved database connections.

Browsing the tables on a database server.

Sheets in a workbook, tables in a database file.

Splitting one file into many by a column.

Turning rows into columns and back.

Per-column statistics, nulls and distributions.

Reading rows out of an open tab or a file.

Reading a text, code or Markdown file's content.

Rolling timestamps up to hours, days, months.

Moving averages and running totals.

Any aggregation, filter or join over open data.

A quick random look at what the data holds.

Column names and types. The usual first step.

Whether the structure changed between versions.

Finding a value somewhere in a table.

Working out which columns two tables share.

Looking at the end of a file without loading it all.

Bulk edits: trim, case, rename, cast.

Stacking tables with the same columns.

Finding which columns could serve as a key.

Checking a file against a schema file.

How often each value occurs in a column.

Writing text content to a file.