@virajsanghvi1

raglite provides local-first rag caching via doc distillation into structured markdown, indexed in chroma with hybrid vector and keyword search. designed for private knowledge retrieval without replacing model context.

raglite-local-rag-cache distills documents into structured markdown, indexes them with chroma, and retrieves via hybrid search (vector plus keyword). designed for local-first rag workflows on private or non-training data.

raglite-library provides local-first rag indexing via chroma and hybrid search, distilling documents to markdown before embedding for private, auditable retrieval without managed infrastructure.