Skip to main content
Classifies a document or text across one or more dimensions, each with predefined categories. Display name “Document Classification”. On by default for new agents. This output is already presentation-ready; don’t reformat it with another AI step.

​
Authentication and enablement

No integration. Default-on.

​
Inputs

  • file_url or text (one required): PDF, image, or spreadsheet via file_url; raw text via text.
  • dimensions (required): array of {name, categories, allow_multiple?}. allow_multiple: false forces single-select on that dimension; omit or true for multi-select.

​
Output

Per-dimension classifications in a display-ready format.

​
Limits and side effects

  • Spreadsheets are classified from parsed cell text.
  • No external writes.

​
Expected errors

  • Neither file_url nor text.
  • Empty or malformed dimensions.