Registry / ai-ml / mongodb-rag-core

mongodb-rag-core

library0.7.1jsnpmunverified

Core library for building RAG (retrieval-augmented generation) chatbots with MongoDB. v0.7.1 — active development with frequent releases. Provides shared components like configs, embedding interfaces, data sources, vector search integration, and LLM abstractions used by the MongoDB Chatbot Framework. Differentiated by deep MongoDB Atlas Vector Search integration, strict TypeScript types, and modular architecture. Designed for Node 18+ and ESM-first.

ai-ml database

npm install mongodb-rag-core

github ↗package ↗homepage ↗

INSTALL

IMPORT

SIG · MONGODB-RAG-CORE

mongodb-rag-core

ai-mljavascriptv0.7.1

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.

AppConfig

✓ import { AppConfig } from 'mongodb-rag-core'

✗ const { AppConfig } = require('mongodb-rag-core')

ESM-only; CJS not supported.

DataStreamer

✓ import { DataStreamer } from 'mongodb-rag-core'

✗ import DataStreamer from 'mongodb-rag-core'

Named export, not default.

MongoDBDataSource

✓ import { MongoDBDataSource } from 'mongodb-rag-core'

✗ import { MongoDataSource } from 'mongodb-rag-core'

Correct name is MongoDBDataSource with DB.

type Embedding

✓ import type { Embedding } from 'mongodb-rag-core'

✗ import { Embedding } from 'mongodb-rag-core'

Use type import for TypeScript type-only exports.

Shows basic setup of AppConfig, MongoDBDataSource, and DataStreamer for querying a vector store.

import { AppConfig, MongoDBDataSource, DataStreamer } from 'mongodb-rag-core'; import { MongoClient } from 'mongodb'; const uri = process.env.MONGO_URI ?? 'mongodb://localhost:27017'; const client = new MongoClient(uri); const config: AppConfig = { dataSources: [ new MongoDBDataSource({ client, databaseName: 'my_db', collectionName: 'documents', embeddingField: 'embedding', }), ], streamer: new DataStreamer(), }; async function run() { await client.connect(); const results = await config.dataSources[0].query('test query'); console.log(results); await client.close(); } run().catch(console.error);

Debug

Known issues

breakingv0.7.0: Breaking changes to data source API. Deprecated old query method signature.

fix

Update query invocations to new signature; see migration guide.

affects: <0.7.0

deprecatedv0.6.0: Deprecated 'MongoDataSource' in favor of 'MongoDBDataSource'.

fix

Rename all imports and usages to MongoDBDataSource.

affects: >=0.6.0

gotchaPackage is ESM-only. Using require() throws MODULE_NOT_FOUND.

fix

Use import syntax and ensure 'type': 'module' in package.json or use dynamic import().

affects: >=0.5.0

gotchaTypeScript types are exported but may require 'skipLibCheck': false in tsconfig to resolve embedded dependencies.

fix

Enable skipLibCheck: false or explicitly install missing @types/* packages.

affects: >=0.4.0

Errors

Common errors & fixes

Cannot find module 'mongodb-rag-core' or its corresponding type declarations.

ESM-only package used with CommonJS require() or missing tsconfig.json settings.

fix

Use import syntax instead of require(). Ensure tsconfig has 'moduleResolution': 'node16' or 'bundler'.

TypeError: (0 , mongodb_rag_core.MongoDBDataSource) is not a constructor

Importing from named export but using wrong symbol name (e.g., MongoDataSource).

fix

Use correct import: import { MongoDBDataSource } from 'mongodb-rag-core'.

Uncaught Error: DataStreamer is not a constructor

Default import when the export is named.

fix

Use named import: import { DataStreamer } from 'mongodb-rag-core'.

Upgrade

Version history

0.7.1latest on npm

Audit

Dependencies

mongodbrequiredMongoDB driver for database operations

mongodb-rag-ingestoptionalShared data ingestion pipeline types and functions

Agent activity

9 hits · last 30 days

node

Resources

githubhttps://github.com/mongodb/chatbot ↗

packagemongodb-rag-core ↗

homepagehttps://mongodb.github.io/chatbot/ ↗