• PROVISION
  • ADJUST
  • VALIDATE
  • HUMAN

How to configure Atlas search indices with your AI

Connect your AI client to MongoDB Atlas once. From then on, you can manage your vector search infrastructure through natural language.

Ask AI about this page

Short answer

How can I set up Atlas vector search using my AI?

You use the create_index tool to define and deploy vector search indices with specific dimensions. Your AI handles the technical configuration so you don't have to touch the Atlas UI. You'll get immediate index provisioning for your vector data.

The outcomes

What your AI delivers

The AI handles the heavy lifting of index management.

PROVISION

New vector indices

The AI uses create_index to set up new vector search indices. You just tell it the dimensions and field names you need.

ADJUST

Updated dimensions

You can update existing index configurations to match your embedding models. The AI handles the parameter updates for you.

VALIDATE

Index readiness

Your AI can verify if an index is ready for use. This ensures your vector search is active before you run queries.

HUMAN

Final oversight

You stay in control of the final configuration. The AI proposes the index settings, but you decide when to commit the changes.

The workflow

What your AI does when the request arrives.

The AI translates your intent into specific Atlas configuration commands.

  1. Analyze requirements

    Your AI looks at your request to determine the necessary vector dimensions and field mappings for your MongoDB collection.

    create_index

  2. Generate configuration

    The agent builds the specific JSON structure required for a MongoDB Atlas vector index based on your instructions.

    create_index

  3. Execute provisioning

    The AI sends the command to Atlas to begin building the index in your cluster.

    create_index

  4. Confirm readiness

    The AI monitors the process and lets you know once the index is fully built and searchable.

    create_index

Try it

Copy these to start.

Use these prompts to begin managing your indices.

Starting points

Swap out the collection names and dimension numbers to match your specific database setup.

MongoDB Atlas Vector Search Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_rMRfujNKzotBfoPOFvzesaIZTdVkigbmunZrRwce/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — MongoDB Atlas Vector Search capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "mongodb-atlas-vector-search-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_rMRfujNKzotBfoPOFvzesaIZTdVkigbmunZrRwce/mcp"
    }
  }
}
Copy into chat04

  • Create a new vector search index for my 'products' collection with 1536 dimensions.

  • Set up an Atlas search index on the 'embeddings' field in my 'users' collection.

  • Adjust the vector index on my 'documents' collection to use 768 dimensions.

  • Provision a new index for the 'chat_history' collection using the specified vector field.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect MongoDB Atlas Vector Search once, then ask.

Click the link to connect your account. Your credentials stay encrypted and secure, and you can start managing indices immediately.

Connect MongoDB Atlas Vector Search to your AI

FAQ

How the task behaves

  • 01

    Can the AI delete my existing collections?

    No. The AI only has the capability to create indices through the create_index tool.

  • 02

    Does the AI manage my database users?

    No. This connector is strictly for managing vector search indices.

  • 03

    Can I change the embedding model through the AI?

    The AI can change the index dimensions to match a new model, but it does not perform the actual embedding generation.

  • 04

    Will the AI automatically update my code?

    No. The AI only updates the configuration on the MongoDB Atlas side.

  • 05

    Can I create multiple indices at once?

    Yes, you can ask the AI to provision multiple indices by specifying the different collections or fields in your prompt.

  • More questions about MongoDB Atlas Vector Search? The Connector page answers them. See everything the MongoDB Atlas Vector Search Connector can do

Connect MongoDB Atlas Vector Search to Claude, Cursor, ChatGPT & more