endee-llamaindex

A LlamaIndex vector store integration for Endee — enabling seamless RAG (Retrieval-Augmented Generation) workflows with the Endee vector database.

Installation

npm install endee-llamaindex
# or
pnpm add endee-llamaindex
# or
yarn add endee-llamaindex

Prerequisites

• An Endee account and authentication token
• An existing index created in Endee (or create one using the Endee client)
• Node.js 18+

Quick Start

import { EndeeVectorStore } from "endee-llamaindex";
import { Document, Settings, storageContextFromDefaults, VectorStoreIndex } from "llamaindex";
import { OpenAI, OpenAIEmbedding } from "@llamaindex/openai";
import { Endee } from "endee";

// Configure your embedding model
Settings.embedModel = new OpenAIEmbedding({
  model: "text-embedding-3-small",
  apiKey: process.env.OPENAI_API_KEY,
});

// Configure your LLM
Settings.llm = new OpenAI({
  model: "gpt-4o",
  apiKey: process.env.OPENAI_API_KEY,
});

// Initialize Endee client and get/create an index
const endeeClient = new Endee(process.env.ENDEE_AUTH_TOKEN ?? "");
const index = await endeeClient.getIndex("my-index");
// Or create a new index:
// const index = await endeeClient.createIndex({
//   name: "my-index",
//   dimension: 1536, // Match your embedding model dimension
// });

// Initialize the Endee vector store
const vectorStore = new EndeeVectorStore({
  index: index,
  chunkSize: 100,
});

// Create documents
const document = new Document({
  text: "Your document content here...",
  id_: "doc-1",
});

// Index documents with storage context
const storageContext = await storageContextFromDefaults({
  vectorStore: vectorStore,
});

const vectorStoreIndex = await VectorStoreIndex.fromDocuments([document], { storageContext });

// Query the index
const queryEngine = vectorStoreIndex.asQueryEngine();
const response = await queryEngine.query({
  query: "What is the main topic?",
});

console.log(response.toString());

Configuration

EndeeVectorStore Parameters

| Parameter | Type | Description | | ----------- | -------- | --------------------------------------------------- | | index | Index | Endee index instance (from Endee.getIndex() or Endee.createIndex()) | | chunkSize | number | Batch size for upserting vectors (default: 100) | | textKey | string | Key used to store text in metadata (default: "text") |

Usage Examples

Indexing Documents

import fs from "node:fs/promises";
import { Endee } from "endee";

// Initialize Endee client and get/create an index
const endeeClient = new Endee(process.env.ENDEE_AUTH_TOKEN ?? "");
const index = await endeeClient.getIndex("knowledge-base");

const vectorStore = new EndeeVectorStore({
  index: index,
  chunkSize: 100,
});

const content = await fs.readFile("./documents/article.txt", "utf-8");
const document = new Document({ text: content, id_: "article-1" });

const storageContext = await storageContextFromDefaults({ vectorStore });
const vectorStoreIndex = await VectorStoreIndex.fromDocuments([document], { storageContext });

Querying an Existing Index

import { Endee } from "endee";

// Initialize Endee client and get the index
const endeeClient = new Endee(process.env.ENDEE_AUTH_TOKEN ?? "");
const index = await endeeClient.getIndex("knowledge-base");

const vectorStore = new EndeeVectorStore({
  index: index,
  chunkSize: 100,
});

// Load from existing vector store
const vectorStoreIndex = await VectorStoreIndex.fromVectorStore(vectorStore);

const queryEngine = vectorStoreIndex.asQueryEngine();
const response = await queryEngine.query({
  query: "Summarize the key points",
});

console.log(response.toString());

Filtering in Queries

EndeeVectorStore supports metadata filtering during queries via LlamaIndex's MetadataFilters.
Currently, only the == (equals) and in operators are supported.

import { MetadataFilter } from "llamaindex";
import { Endee } from "endee";

// Initialize Endee client and get the index
const endeeClient = new Endee(process.env.ENDEE_AUTH_TOKEN ?? "");
const index = await endeeClient.getIndex("knowledge-base");

const vectorStore = new EndeeVectorStore({
  index: index,
});

const vectorStoreIndex = await VectorStoreIndex.fromVectorStore(vectorStore);

const filter : MetadataFilter[] = [{key: "type", value: "sports", operator: "=="}];

const queryEngine = vectorStoreIndex.asQueryEngine({
  preFilters: {filters: filter}
});

const response = await queryEngine.query({
  query: "Explain the main concepts",
  filters,
});

console.log(response.toString());

Deleting Documents

import { Endee } from "endee";

// Initialize Endee client and get the index
const endeeClient = new Endee(process.env.ENDEE_AUTH_TOKEN ?? "");
const index = await endeeClient.getIndex("knowledge-base");

const vectorStore = new EndeeVectorStore({
  index: index,
  chunkSize: 100,
});

// Delete by reference document ID
await vectorStore.delete("doc-id-to-remove");

API Reference

EndeeVectorStore

Methods

| Method | Description | | --------------------- | ---------------------------------------- | | add(nodes) | Add embedding nodes to the vector store | | query(query) | Query the vector store for similar nodes | | delete(refDocId) | Delete vectors by reference document ID | | client() | Returns "Endee" string identifier |

Dependencies

• llamaindex - LlamaIndex TypeScript SDK
• endee - Endee vector database client

License

ISC