Hi AIP team,
I’m currently building an AIP agent and noticed that it only supports text and image inputs.
Is there a roadmap or possibility to support additional input types (PDFs, spreadsheets, or video) given that some of the underlying LLMs are already capable of parsing these formats?
Ideally, it would be great if input types could be configurable (allow/deny) at the agent settings level.
Thanks for considering, and appreciate the work on AIP.