RAG
Hybrid search
Vector + BM25 blend with weighted normalization.
import { createHybridRetriever } from '@agentskit/rag'
const retriever = createHybridRetriever({
vector: rag,
lexical: bm25Index,
weights: { vector: 0.7, lexical: 0.3 },
})#When to use
Queries with rare keywords or exact tokens (error codes, SKUs, identifiers) that vector alone misses.