Configure a spreadsheet output

Last updated: July 12, 2026

Use a spreadsheet when the result is primarily rows, columns, calculations, classifications, or review queues.

Choose the output type

Open the Output step in the agent builder and choose the artifact the agent should produce every run.

The agent needs to return a table that people will sort, filter, import, or reconcile.
The output should support batch review across many records.
Each finding needs standard fields such as status, confidence, rationale, and source.

Choose Spreadsheet in the output type menu.
List the required columns and define allowed values for status, severity, confidence, and action fields.
Use a JSON schema when another system will ingest the same row structure through the API.

Add a short output instruction to the agent instructions or the relevant skill so the agent knows what good looks like.

Return a spreadsheet-ready table with one row per entity, standard columns, allowed status values, confidence, rationale, and source URLs.

The audience and use case are clear.
The output format is selected before testing.
The agent instructions describe the expected sections, fields, or scenes.
Important claims include evidence or source links.
The output can be reviewed without guessing what each field or section means.
If another system consumes the result, the schema and allowed values are stable.