- DATA ENTRY
- QUERYING
- CLEANUP
- HUMAN
Short answer
Can I manage MongoDB documents using natural language?
Yes. You can tell your AI to add new records, find specific data, or delete documents in your Atlas collections. You'll see how the agent handles these tasks below.
Where every action lands.
Database outcomes
Your commands translate into direct changes in your MongoDB environment.
DATA ENTRY
New records added
The AI uses the insert tool to write new documents to your collections. You get fresh data without writing code.
QUERYING
Instant data retrieval
The agent uses the find tool to pull specific records. You get the exact information you asked for immediately.
CLEANUP
Targeted record removal
The AI uses the delete tool to strip out unwanted entries. This keeps your collections clean and accurate.
HUMAN
Manual review needed
You must confirm sensitive deletions or large updates. The AI identifies the targets, but you keep the final say.
The workflow
What your AI does when a command arrives.
The AI acts as a bridge between your chat and your database.
Analyze request
Your agent reads your prompt to figure out if you want to add, find, or remove data.
findCreate records
The AI builds a new document based on your description and places it in the collection.
insertSearch data
The agent scans your Atlas collections to locate the specific documents you described.
findRemove records
The AI targets the specific document you want gone and removes it from the database.
delete
Try it
Copy these to start.
Use these lines to test your connection.
Starting points
Replace the collection names and field values with your own data.
MongoDB Atlas Vector Search Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_rMRfujNKzotBfoPOFvzesaIZTdVkigbmunZrRwce/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — MongoDB Atlas Vector Search capabilities are ready to use.
{
"mcpServers": {
"mongodb-atlas-vector-search-mcp": {
"url": "https://edge.vinkius.com/vk_preview_rMRfujNKzotBfoPOFvzesaIZTdVkigbmunZrRwce/mcp"
}
}
}Insert a new user into the 'customers' collection with the name 'Jordan' and status 'active'.
Find all documents in the 'orders' collection where the total is over 100.
Delete the record in the 'temp_data' collection with the ID '98765'.
Add a new product to the 'inventory' collection called 'wireless mouse' with a stock of 50.
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 Atlas account. Your login stays encrypted on our side, and you can find more details on the Connector page.
Connect MongoDB Atlas Vector Search to your AIFAQ
How this task behaves
- 01
Can the AI change my database schema?
No. The AI can only insert, find, and delete documents. It cannot modify indexes or change your schema.
- 02
Will the AI delete my whole collection?
No. The AI only deletes specific documents that match your description. It cannot drop entire collections.
- 03
Do I have to write JSON?
No. You can use plain English to describe what you want to happen.
- 04
Can the AI run complex aggregations?
No. This Connector is limited to the insert, find, and delete tools.
- 05
Can the AI update existing documents?
No. The current tools only support inserting new records, finding them, or deleting them.
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
