🗄️ Couchbase MCP Server for LLMs
About
It is a Model Context Protocol (MCP) server that enables LLMs to interact directly with Couchbase databases on Capella clusters. Users can query buckets, perform CRUD operations, execute N1QL queries, and manage data seamlessly through natural language.
Explore
- Execute N1QL queries via natural language
- List, create, and delete buckets, scopes, and collections
- Create, read, update, and delete documents individually or in bulk
- Manage indexes (create primary, create, list, drop)
- All tools support optional collection and scope parameters
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
🗄️ Couchbase MCP Server for LLMsCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
1. Prerequisites
- Node.js 16 or higher
- A running Couchbase instance on Capella
- Claude Desktop application
2. Installation
Couchbase MCP Server can be installed in two ways:
If you prefer to clone and run the project manually:
npm install
json{
"mcpServers": {
"couchbase": {
"command": "npx",
"args": ["-y", "@couchbasedatabase/couchbase-mcp"],
"env": {
"COUCHBASE_URL": "<COUCHBASE CONNECTION STRING>",
"COUCHBASE_BUCKET": "<BUCKET NAME>",
"COUCHBASE_USERNAME": "<COUCHBASE USERNAME>",
"COUCHBASE_PASSWORD": "<COUCHBASE PASSWORD>"
}
}
}
}
json{
"mcpServers": {
"couchbase": {
"command": "node",
"args": ["path/to/MCP-Server-Couchbase/dist/index.js"],
"env": {
"COUCHBASE_URL": "<COUCHBASE CONNECTION STRING>",
"COUCHBASE_BUCKET": "<BUCKET NAME>",
"COUCHBASE_USERNAME": "<COUCHBASE USERNAME>",
"COUCHBASE_PASSWORD": "<COUCHBASE PASSWORD>"
}
}
}
}
```
4. Verify Connection
- Restart Claude Desktop
- The Couchbase MCP server tools should now be available in your conversations
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"\ud83d\uddc4\ufe0f couchbase mcp server for llms": {
"Aniket310101_MCP-Server-Couchbase": {
"command": "npx",
"args": [
"-y",
"@couchbasedatabase/couchbase-mcp"
]
}
}
}
}
McpServers
{
"Aniket310101_MCP-Server-Couchbase": {
"command": "npx",
"args": [
"-y",
"@couchbasedatabase/couchbase-mcp"
]
}
}
A Model Context Protocol (MCP) server that enables LLMs to interact directly with Couchbase databases on Capella clusters. Query buckets, perform CRUD operations, execute N1QL queries, and manage data seamlessly through natural language.
🚀 Quick Start
1. Prerequisites
- Node.js 16 or higher
- A running Couchbase instance on Capella
- Claude Desktop application
2. Installation
Couchbase MCP Server can be installed in two ways:
### Option 1: Using NPX (Recommended)
The quickest way to get started is using NPX:
npx -y @couchbasedatabase/couchbase-mcp
### Option 2: Manual Installation
If you prefer to clone and run the project manually:
# Clone the repository
git clone https://github.com/Aniket310101/MCP-Server-Couchbase.git
cd MCP-Server-Couchbase
# Install dependencies
npm install
# Build the project
npm run build
3. Claude Desktop Integration
Add this configuration to your Claude Desktop config file:
Windows: %APPDATA%/Claude/claude_desktop_config.json
MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
### Option 1: With Package Installation
{
"mcpServers": {
"couchbase": {
"command": "npx",
"args": ["-y", "@couchbasedatabase/couchbase-mcp"],
"env": {
"COUCHBASE_URL": "<COUCHBASE CONNECTION STRING>",
"COUCHBASE_BUCKET": "<BUCKET NAME>",
"COUCHBASE_USERNAME": "<COUCHBASE USERNAME>",
"COUCHBASE_PASSWORD": "<COUCHBASE PASSWORD>"
}
}
}
}
### Option 2: With Manual Installation
{
"mcpServers": {
"couchbase": {
"command": "node",
"args": ["path/to/MCP-Server-Couchbase/dist/index.js"],
"env": {
"COUCHBASE_URL": "<COUCHBASE CONNECTION STRING>",
"COUCHBASE_BUCKET": "<BUCKET NAME>",
"COUCHBASE_USERNAME": "<COUCHBASE USERNAME>",
"COUCHBASE_PASSWORD": "<COUCHBASE PASSWORD>"
}
}
}
}
4. Verify Connection
- Restart Claude Desktop
- The Couchbase MCP server tools should now be available in your conversations
📝 Available Tools
Basic Operations
- query: Execute N1QL queries
- listBuckets: List available buckets
Scope Management
- createScope: Create a new scope in a bucket
- deleteScope: Delete an existing scope
- listScopes: List all scopes in a bucket
Collection Management
- createCollection: Create a new collection in a scope
- dropCollection: Delete a collection from a scope
Document Operations
- createDocument: Create a new document
- getDocument: Retrieve a document by ID
- updateDocument: Update an existing document
- deleteDocument: Delete a document by ID
- bulkCreateDocuments: Create multiple documents at once
Index Management
- createIndex: Create a new index on specified fields
- createPrimaryIndex: Create a primary index on a collection
- listIndexes: List all indexes in a bucket
- dropIndex: Drop an existing index
Each tool supports optional collection and scope parameters for targeting specific data containers.
🔒 Security Considerations
- Always use environment variables for sensitive credentials
- Consider running the server behind a reverse proxy for production use
- Implement appropriate access controls and authentication as needed
📚 Examples
Here are some example interactions with Claude using the MCP server:
1. List all buckets:
Could you show me all available buckets in the database?
2. Create a scope and collection:
Create a new scope called "users" and a collection called "profiles" in it
3. Query documents:
Find all users who signed up in the last 30 days
4. Create a document:
Create a new user document with name "John Doe" and email "[email protected]"
🤝 Contribution
Contributions are welcome! Please feel free to submit a Pull Request.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.



