Bring Vector Search
to LangChain

Connect your MongoDB Atlas cluster to any AI agent and take full control of your high-performance vector search, embedding storage, and operational data management through natural conversation.

What you can do

• Vector Similarity Search — Execute sophisticated '$vectorSearch' queries against your collections to retrieve semantically relevant matches using raw embedding vectors directly from your agent
• Unified Data Management — Find, insert, and delete standard MongoDB documents using literal MQL (MongoDB Query Language) filters to manage both vector and operational data in a single system
• Search Index Provisioning — Create and configure Atlas Search indices with custom dimensions and mapping definitions to optimize your cluster's similarity calculation infrastructure
• Collection Lifecycle Audit — List all managed data collections and retrieve schema boundaries to understand namespace references and database organization natively
• Real-time Ingestion — Synchronize new JSON records into your collections, allowing for instant searchability and automated vector parsing if Atlas triggers are enabled
• Precision Retrieval — Execute targeted MQL queries to fetch specific data points or metadata chunks, bypassing vector logic for rapid structural verification and auditing

How it works

1. Subscribe to this server
2. Enter your MongoDB Atlas Data API URL and API Key
3. Start optimizing your search infrastructure from Claude, Cursor, or any MCP-compatible client

Who is this for?

• ML Engineers — test vector relevance and verify embedding dimensions through natural conversation without manual SDK scripts
• Backend Developers — manage operational data and vector search results in a single workflow directly from your workspace terminal
• Search Architects — audit search indices and monitor collection organization across multiple Atlas environments efficiently