documents

name	documents
description	Documents tools for Dataspheres AI

Documents

Tool reference for this resource group, mirrored by hand from the platform live /api/mcp/schema schema.

Tools

find_pdf_media — Find PDF Pages / Images

Find extracted pages or embedded images from a PDF. Use this AFTER the user has run Extract Pages / Extract Images on a PDF. Pick your render strategy based on user intent:

• Single image: user asks for "the cover", "page 3", "the boat on page 5", a specific named item → return 1 item, render inline as a large image.
• Carousel: user asks for "all the boats", "every page", "a summary of the catalog", comparison requests → return N items, render as a horizontal MiniActivityCard carousel. In your response, reference the image directly by its /api/documents/:id/download URL or use @[doc:id|title|imageUrl] mention so the client renders it inline.

Field	Type	Required	Description
documentId	string	yes	Parent PDF Document id
kind	string	no	Filter: "pages" for page renders, "images" for embedded images, omit for both
pageNumber	number	no	Filter to a specific page number
limit	number	no	Max items to return (default 20)