Enhancement idea: smarter chunking for structured documents

[khaledmahmoudbu]

Im considering a contribution to improve chunking for highly structured documents in the Library ingestion pipeline.
The goal is to preserve document structure more effectively during chunking to improve retrieval quality and citation precision in RAG workflows.

Before I implement it, I wanted to check whether this type of enhancement would be a good fit for LLMWare’s contribution roadmap.