Imported from sancovp/dark-factory-live (
kbworld/state/libraries/retrieval_augmented_generation_architecture_patt/understand-retrieval_augmented_generation_architecture_patt/understand-vector_database/.claude/skills/0.4.3-understand-vector_database/SKILL.md). Install upstream withnpx skills add sancovp/dark-factory-live --skill 0.4.3-understand-vector_database. Copyright stays with the author.
name: 0.4.3-understand-vector_database description: [0.4.3] Specialized storage system indexing embeddings with efficient similarity search capabilities (Pinecone, Weavia
understand-vector_database
CALL NUMBER: retrieval_augmented_generation_architecture_patt.vector_database
DEFINITION: Specialized storage system indexing embeddings with efficient similarity search capabilities (Pinecone, Weaviate, Chroma, Milvus, pgvector)
Invoke this skill to understand vector_database down to its primitives. The RELATIVE ROOT below is the least-fixed-point closure of everything it bundles from — the full import cone, grouped by the lib each prim comes from. Projected from a prover-typed KB (MAP/SWI-Prolog consistency gate): every reference below resolves.
THE RELATIVE ROOT (the import cone, by lib)
from retrieval_augmented_generation_architecture_patt
- approximate_nearest_neighbor (d1): Vector indexing technique trading exactness for speed: HNSW, IVF, or LSH structures enabling sub-linear retrieval at scale
- incremental_indexing (d1): Index update strategy adding or modifying vectors without rebuilding entire index structure
- index_memory_footprint (d1): Storage requirements for vector index relative to number of embedded documents and embedding dimensionality
- indexing_pipeline (d1): End-to-end process: document ingestion, preprocessing, chunking, embedding generation, and vector store population
- real_time_indexing (d1): Incremental index updates allowing newly ingested documents to be immediately searchable without full recomputation
- vector_similarity_search (d1): Core retrieval operation computing distance or similarity (cosine, dot-product, euclidean) between query and document vectors
Projected from the retrieval augmented generation architecture patterns KB (207 concepts / 225 relations) — consistency-typed by MAP; the facet list after the colon IS the cross-lib dependency web.
(leaf — this is an actual skill.)