It brings the extracted text and an image of each page of the PDF in the context. More details here:

→ View original post on X — @romainhuet