Registry / database / rivet-plugin-mongodb

rivet-plugin-mongodb

library6.4.0jsnpmunverified

A plugin for Rivet (AI graph application) providing 4 nodes to interact with MongoDB: store vectors, perform aggregations, search collections, and execute vector searches ($vectorSearch). Current stable version is 6.4.0, released as part of Rivet's plugin ecosystem. It is a community-maintained plugin that bridges Rivet's graph-based AI workflows with MongoDB's document and vector search capabilities. Key differentiators: native integration with Rivet nodes, support for MongoDB Atlas Vector Search, and simple plugin installation via Rivet's UI. Requires a MongoDB connection string and, for vector search, an appropriate Atlas index. Release cadence is irregular, with last major change being deprecation of MongoDBVectorKNN in favor of MongoDBVectorSearch.

database ai-ml

npm install rivet-plugin-mongodb

github ↗package ↗homepage ↗

INSTALL

IMPORT

SIG · RIVET-PLUGIN-MONGO

rivet-plugin-mongodb

databasejavascriptv6.4.0

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.

MongoDBStore

✓ import { MongoDBStore } from '@rivet-gg/plugin-mongodb'

✗ import MongoDBStore from '@rivet-gg/plugin-mongodb'

Named import, not default import.

MongoDBAggregation

✓ import { MongoDBAggregation } from '@rivet-gg/plugin-mongodb'

✗ const MongoDBAggregation = require('@rivet-gg/plugin-mongodb').MongoDBAggregation

ESM-only package; CommonJS require not supported.

MongoDBCollectionSearch

✓ import { MongoDBCollectionSearch } from '@rivet-gg/plugin-mongodb'

Named import.

MongoDBVectorSearch

✓ import { MongoDBVectorSearch } from '@rivet-gg/plugin-mongodb'

✗ import { MongoDBVectorKNN } from '@rivet-gg/plugin-mongodb'

MongoDBVectorKNN was deprecated in v0.0.3; use MongoDBVectorSearch instead.

Demonstrates how to import and configure the MongoDBVectorSearch and MongoDBStore nodes for use within Rivet graphs.

import { MongoDBVectorSearch, MongoDBStore } from '@rivet-gg/plugin-mongodb'; // Plugin nodes are used within Rivet graphs, not standalone. // Example configuration for MongoDBVectorSearch node: const vectorSearchNode = new MongoDBVectorSearch({ collection: 'documents', vectorField: 'embedding', queryVector: [0.1, 0.2, 0.3], numCandidates: 100, limit: 10, index: 'vector_index', filter: { category: 'science' }, connectionString: process.env.MONGODB_URI ?? '' }); // Example for MongoDBStore node to store a vector: const storeNode = new MongoDBStore({ collection: 'documents', vector: [0.1, 0.2, 0.3], metadata: { title: 'Example', category: 'science' }, connectionString: process.env.MONGODB_URI ?? '' });

Debug

Known issues

breakingMongoDBVectorKNN node removed and replaced with MongoDBVectorSearch in version 0.0.3. Existing graphs using MongoDBVectorKNN will break.

fix

Remove MongoDBVectorKNN node and use MongoDBVectorSearch instead. Create a new Atlas Vector Search index per documentation.

affects: >=0.0.3

gotchaMongoDBVectorSearch requires a pre-created Atlas Vector Search index on the collection. Without it, the node will fail.

fix

Create an index using MongoDB Atlas UI or CLI as described in https://www.mongodb.com/docs/atlas/atlas-vector-search/vector-search-type/

affects: >=0.0.3

gotchaMongoDBCollectionSearch node loads entire collection into memory; may cause performance issues for large collections.

fix

Limit collection size or use MongoDBVectorSearch with filter for large datasets. The plugin roadmap notes this issue.

affects: *

deprecatedMongoDBVectorKNN is deprecated; new projects should use MongoDBVectorSearch.

fix

Replace MongoDBVectorKNN with MongoDBVectorSearch and update index.

affects: >=0.0.3

gotchaPlugin requires a MongoDB connection string set in Rivet settings or passed via environment variable. Missing connection string causes runtime errors.

fix

Ensure MONGODB_URI environment variable is set or connection string is configured in Rivet plugin settings.

affects: *

Errors

Common errors & fixes

MongoDBVectorKNN is not a constructor

Using deprecated MongoDBVectorKNN node in version >=0.0.3.

fix

Replace MongoDBVectorKNN with MongoDBVectorSearch.

MongoError: bad index specification for $vectorSearch

Missing or incorrect Atlas Vector Search index on the collection.

fix

Create a valid vector search index on the collection. See https://www.mongodb.com/docs/atlas/atlas-vector-search/vector-search-type/

MongoError: Authentication failed

Invalid MongoDB connection string or credentials.

fix

Check that the MONGODB_URI is correctly set and includes proper credentials.

Cannot find module '@rivet-gg/plugin-mongodb'

Plugin not installed in Rivet or incorrect import path.

fix

Install plugin via Rivet UI plugins tab or add @rivet-gg/plugin-mongodb to package.json if using custom setup.

Upgrade

Version history

6.4.0latest on npm

Audit

Dependencies

No dependency data recorded yet.

Agent activity

3 hits · last 30 days

node

OpenAI (training)

Resources

githubgit://github.com/a-rothwell/rivet-plugin-mongodb ↗

packagerivet-plugin-mongodb ↗

homepagegit://github.com/a-rothwell/rivet-plugin-mongodb ↗