mongodb-rag

library0.83.0jsnpmunverified

MongoDB RAG v0.83.0 is a JavaScript/TypeScript library that simplifies Retrieval Augmented Generation (RAG) using MongoDB Atlas Vector Search. It provides vector search, batch processing, index management, in-memory caching, and advanced chunking strategies (sliding window, semantic, recursive). Includes a CLI for scaffolding RAG apps and configuration. Released under Apache-2.0, updated regularly with minor releases every few weeks. Key differentiator: tight integration with MongoDB Atlas, built-in chunking, and CLI tooling compared to generic vector search libraries.

ai-ml database

npm install mongodb-rag

github ↗package ↗homepage ↗

INSTALL

IMPORT

SIG · MONGODB-RAG

mongodb-rag

ai-mljavascriptv0.83.0

harness data pending

Install & Compatibility

Where this runs

No compatibility data collected yet for this library.

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

MongoRAG

✓ import { MongoRAG } from 'mongodb-rag'

✗ import MongoRAG from 'mongodb-rag'

MongoRAG is a named export, not default. Use named import syntax.

ChunkingStrategy

✓ import { ChunkingStrategy } from 'mongodb-rag'

✗ const ChunkingStrategy = require('mongodb-rag').ChunkingStrategy

CommonJS require works but TypeScript/ESM prefer named import.

Shows environment setup, MongoRAG initialization with OpenAI embeddings, document ingestion, and vector search query.

import { MongoRAG } from 'mongodb-rag'; import dotenv from 'dotenv'; dotenv.config(); const rag = new MongoRAG({ mongoUrl: process.env.MONGODB_URI ?? 'mongodb://localhost:27017', database: 'my_rag_db', collection: 'documents', embedding: { provider: 'openai', apiKey: process.env.OPENAI_API_KEY ?? '', model: 'text-embedding-3-small', dimensions: 1536 } }); async function main() { // Ingest documents const docs = [ { title: 'MongoDB Atlas', content: 'A fully managed cloud database for modern applications.' }, { title: 'Vector Search', content: 'Enables semantic similarity search on vector embeddings.' } ]; await rag.ingest(docs); // Query const results = await rag.search({ query: 'cloud database', topK: 3 }); console.log(results); } main().catch(console.error);

Debug

Known issues

gotchaEmbedding provider 'openai' requires a valid API key. The library does not validate the key until runtime.

fix

Ensure process.env.OPENAI_API_KEY is set before instantiating MongoRAG.

affects: >=0.0.0

gotchaThe default collection 'documents' must exist or be created with the correct vector index. MongoRAG does not auto-create indexes.

fix

Use npx mongodb-rag create-index to set up the required vector index.

affects: >=0.0.0

deprecatedThe 'provider' field in embedding config is case-sensitive. 'openai' works but 'OpenAI' does not.

fix

Use lowercase provider names: 'openai', 'ollama', 'huggingface', etc.

affects: >=0.0.0

Errors

Common errors & fixes

Error: Cannot find module 'mongodb-rag'

Package not installed or not in node_modules.

fix

Run 'npm install mongodb-rag' in your project directory.

TypeError: MongoRAG is not a constructor

Incorrect import: imported default instead of named export.

fix

Change 'import MongoRAG from...' to 'import { MongoRAG } from...'

Upgrade

Version history

0.83.0latest on npm

Audit

Dependencies

mongodbrequiredRequired to connect to MongoDB Atlas for storing embeddings and documents.

dotenvoptionalCommonly used for loading environment variables; CLI generates .env files.

Agent activity

11 hits · last 30 days

node

Resources

githubhttps://github.com/mongodb-developer/mongodb-rag ↗

packagemongodb-rag ↗

homepagehttps://github.com/mongodb-developer/mongodb-rag#readme ↗