createRAG

import { createRAG } from '@agentskit/rag'
import { openaiEmbedder } from '@agentskit/adapters'
import { fileVectorMemory } from '@agentskit/memory'

const rag = createRAG({
  embed: openaiEmbedder({ apiKey: process.env.OPENAI_API_KEY! }),
  store: fileVectorMemory({ path: '.agentskit/vectors' }),
})

await rag.ingest([{ id: 'doc-1', content: longDoc }])
const hits = await rag.retrieve({
  query: 'How does token budgeting work?',
  messages: [],
})

ingest takes InputDocument[] — the text field is content, not text.

#API

retrieve and search both return RetrievedDocument[]. Use retrieve when you need the contract (runtime, rerank wrappers). Use search when you already have a query string.

#Options

• chunkSize / chunkOverlap / split — passed to chunking. Defaults: chunkSize 512, chunkOverlap 50. split is an optional (text: string) => string[].
• topK / threshold — defaults for search (and therefore retrieve).

#Related

• Rerank · Hybrid · Loaders