Skip to content
Nappai Documentation

MongoDB Vector Store

MongoDB Vector Store stores documents in a MongoDB Atlas collection and searches it by meaning, using an Atlas vector search index.

A vector database stores texts so they can be searched by meaning, not just by exact words. That way an agent can find the paragraph of a manual that answers a question even if it uses different words.

When to use it

  • Your data already lives in MongoDB Atlas and you want an agent to search it by meaning.
  • You want to keep a knowledge base’s documents next to the rest of your application data.
  • You need to search a large collection with an index managed by Atlas.

Configuration

The Operation field decides what the component does, and the fields and outputs change with your choice:

  • Add (default): stores in the database the documents it receives in Ingest Data.
  • Search: finds the documents closest to the text in Search Query.
  • Retriever: offers the database as a search engine for another component to use, for example an agent or a question-answering chain.

You need a MongoDB Connection credential with your MongoDB cluster details. See Credentials.

FieldWhat it does
Credential (required)The MongoDB Connection credential with the connection details.
OperationWhat the component does: Add, Search or Retriever.
Ingest Data (required)With Add. The documents to store, usually those coming out of a text splitter or a file loader.
Search QueryWith Search. The text to search for. Required.
Embedding (required)The connector where you plug the component that turns texts into numbers to compare them by meaning, such as OpenAI Embeddings. Always use the same one to store and to search.
Database Name (required)The MongoDB database.
Collection Name (required)The collection that holds the documents.
Vector Search Index Name (required)The name of the vector search index created in Atlas for that collection. You can create it with Search Index.

Advanced options

FieldWhat it does
Number Of ResultsHow many documents each search returns. Default: 4.
Search TypeHow to search: similarity, similarity_score_threshold. Similarity returns the closest; with a threshold, only those above Search Score Threshold.
Score ThresholdMinimum similarity score with the threshold search. Default: 0.7.

Output

The outputs depend on Operation:

  • Results: with Add and Search. With Search, the list of documents found, which you can pass to a prompt or an agent.
  • Vector Store: with Add. The database, to connect it to components that work with a vector database.
  • Retriever: with Retriever. The search engine, to connect it to components that ask for one.

In tool mode, an agent can use the component to search by itself. See Tool Mode.

Example

A course platform keeps its lessons in MongoDB Atlas and wants an assistant that quotes them.

  1. Create the collection’s vector index with Search Index.
  2. Add MongoDB Vector Store with Operation set to Search, the database, the collection and the index name.
  3. Put it in tool mode and connect it to the assistant’s agent.

Good practices

  • This component does not filter by metadata: MongoDB Atlas only filters on the fields you declare as filters in the vector search index. If you need filters, use Qdrant, PGVector or Elasticsearch, which have metadata filters.