Collected sources and patterns will appear here. Add from search or the patterns library.
Document -> List<ChunkWithCoordinates>
Segment documents while mapping individual chunks back to their exact physical layout coordinates (bounding boxes) on the original page.
Problem it solves
RAG citations lack precise visual page context, making visual verification of retrieved facts difficult for users.
Consumes
Emits
The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.