The unique identifier of the context item.
OptionaldiscardWhether Squid keeps or discards the original file.
Keeping the original file allows reprocessing and the ability for the user to download it later.
Defaults to false.
OptionalextractWhether to extract images from the file, describe them with a vision model, and index those descriptions alongside the text. Defaults to true — a document's charts, diagrams and scanned pages carry content the text layer does not. Note that the text context request has no such option: inline images in a text payload are extracted only for a knowledge-base text context, and default to OFF there.
OptionalextractionThe AI model to use for extraction, if specified.
OptionalimageThe minimum size for extracted images, if applicable.
OptionalmetadataOptional metadata associated with the context item.
OptionaloptionsGeneral options for how to process the file.
OptionalpreferredThe preferred method for extracting data from the document.
Specifies the context type as 'file'.
Request structure for adding file-based context to an AI agent.