Under the hoodTier: how much it should matter to you
1Useful: my rating, 1 to 5
1Signal: worth watching, 1 to 5
Read ToolArea: what it touches
Internal ChangesKind: in v2.1.300,
Internal ChangesSection of the release
What
When Claude reads a PDF without naming a page range and the file is too long to read whole, only its first pages are extracted. Two changes go with this:
The read result gains an optional totalPages field giving the number of pages in the document. It is set only in that case, and is marked @internal.
When file read results are combined, totalPages is now carried over along with firstPage. Before, only firstPage and the pages were kept.
Why
Claude can now be told how long a cut-short PDF really is, so it knows there are more pages it could ask for.
