ai-chat-toolkit-rag-store-memory

In-memory vector store adapter for ai-chat-toolkit-rag.

Current release: 1.0.0

Install

npm install ai-chat-toolkit-rag-store-memory ai-chat-toolkit-rag@^0.1.1

Usage

const { memoryStore } = require("ai-chat-toolkit-rag-store-memory");
// or: import { memoryStore } from "ai-chat-toolkit-rag-store-memory";

server.use(
  rag({
    sources: [localSource({ path: "./docs" })],
    store: memoryStore(),
    embeddings: {
      provider: "openai",
      apiKey: process.env.OPENAI_API_KEY,
    },
  }),
);

Behavior

• Stores chunk embeddings in process memory (no persistence)
• Searches with cosine similarity
• Returns top matches as RagSearchResult[] (respects search.limit from rag() options)

Peer dependency

Requires ai-chat-toolkit-rag@^0.1.1.

License

MIT