🗄️ Couchbase MCP Server for LLMs

by MCP-Mirror

267 downloads
Not rated
GitHub

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.

Details

Author
MCP-Mirror
Downloads
267
Categories
Other

- 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:

  1. Download and install Highlight from highlightai.com/download
  2. Navigate to the plugins tab and select "Add Custom Plugin"
  3. Configure the plugin with the settings below
    Plugin Name 🗄️ Couchbase MCP Server for LLMs
    Command (node, npx, python, etc.)

    Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.

  4. Enable "Start Automatically" if you want the plugin to start when Highlight launches

From the repository

Install the server via NPX (npx -y @couchbasedatabase/couchbase-mcp) or clone and build manually. Configure it in the Claude Desktop config file with environment variables for the Couchbase connection string, bucket, username, and password. After restarting Claude Desktop, the server’s tools become available in 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"
        ]
    }
}

🗄️ Couchbase MCP Server for LLMs

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 "john@example.com"

🤝 Contribution

Contributions are welcome! Please feel free to submit a Pull Request.

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.