Please provide a mechanism that allows third-party applications to generate the PDF, JPEG, and text content that is normally produced by the Representation API.

Currently, Box AI can only be applied to a limited set of file extensions supported by Box services, preventing it from reaching its full potential. If Box AI were able to consume preview images or structured text generated and uploaded by third parties, agents would be able to interpret industry-standard files that Box does not natively support, such as AutoCAD files. This would greatly enhance the strength and versatility of the platform.

Please open a pathway for third parties to convert otherwise unusable binary files into meaningful, AI-readable content.