home / mcp / mongodb mcp server
A mongo db server for the model context protocol (MCP)
Configuration
View docs{
"mcpServers": {
"rock913-mongo-mcp": {
"command": "npx",
"args": [
"mongo-mcp",
"mongodb://<username>:<password>@<host>:<port>/<database>?authSource=admin"
]
}
}
}You can connect Claude-like MCP clients to a MongoDB database and perform queries, inspect schemas, and manage data using natural language. This MCP server enables you to interact with MongoDB collections directly, making data exploration and manipulation fast and intuitive from your preferred client.
Start by ensuring you have an MCP client (for example Claude Desktop) ready to connect to an MCP server. You will use the two available connection methods to run the MongoDB MCP Server locally or remotely. Once the server is running, you can ask your MCP client questions like: which collections exist, what is the schema of a collection, or find documents matching criteria. You can also create indexes and perform document operations such as insert, update, and delete.
Prerequisites you need before starting:
- Node.js 18 or newer
- npm or npx
- An MCP client (for example Claude Desktop) to connect to the server
- Docker and Docker Compose if you want to run a local sandbox (optional for testing)Configure your MCP client to connect to MongoDB MCP Server using one or more of the following runtime configurations. You can run the server via a CLI command using npx or start a local development instance with Node.
{
"mcpServers": {
"mongodb_cli": {
"command": "npx",
"args": [
"mongo-mcp",
"mongodb://<username>:<password>@<host>:<port>/<database>?authSource=admin"
]
},
"mongodb_local": {
"command": "node",
"args": [
"dist/index.js",
"mongodb://root:example@localhost:27017/test?authSource=admin"
]
}
}
}Prerequisites also mention that you may seed a test sandbox if you need sample data to practice with. This helps you verify prompts and demonstrate capabilities like querying collections, inspecting schemas, and performing document updates in a safe environment.
Query documents with filtering and projection to retrieve data that matches your criteria.
List the available collections in the connected MongoDB database.
Insert a single document into a collection to add new data.
Update a single document in a collection based on a filter.
Delete a single document from a collection based on a filter.
Create an index on a field or fields to improve query performance.
Remove an existing index from a collection.
List all indexes on a given collection.