RavenDB MCP Server

A Model Context Protocol (MCP) server that enables AI assistants to interact with RavenDB databases through a standardized interface.

Overview

This MCP server allows AI assistants to perform common RavenDB operations including:

• Connection management

• Database selection

• Collection listing

• Index management

• Document operations (get, store, delete)

• RQL queries

Related MCP server: MCP Toolkit

Requirements

• Node.js 16+

• RavenDB 7.x

• Authentication using non-secured mode (no authentication)

Installation

# Install globally
npm install -g ravendb-mcp

# Or run directly with npx
npx ravendb-mcp

Configuration

Server Configuration

Configure the server using environment variables or a .env file:

# Authentication Method (Only non-secured mode is supported)
RAVENDB_AUTH_METHOD=none

# Connection
RAVENDB_URL=http://your-ravendb-server:port

# Optional settings
RAVENDB_QUERY_TIMEOUT=30000  # Query timeout in milliseconds (optional)

Cline MCP Configuration

To configure this MCP server for use with Cline AI, add the following to your MCP configuration:

Non-secured Mode Configuration

{
  "mcpServers": {
    "github.com/johnib/ravendb-mcp": {
      "disabled": false,
      "timeout": 60,
      "command": "npx",
      "args": ["-y", "ravendb-mcp"],
      "env": {
        "RAVENDB_AUTH_METHOD": "none",
        "RAVENDB_URL": "http://your-ravendb-server:port"
      },
      "transportType": "stdio"
    }
  }
}

You can customize the environment variables based on your specific RavenDB setup.

Available Tools

Connection Tools

initialize-connection

Establishes a connection to a RavenDB server.

{
  "server_url": "https://your-ravendb-server:port",
  "database": "YourDatabase"
}

select-database

Switches to a specific database context.

{
  "database": "AnotherDatabase"
}

Exploration Tools

show-collections

Lists all collections in the current database.

{}

show-indexes

Lists all indexes in the current database.

{}

Document Operations

get-document

Retrieves a document by ID.

{
  "id": "employees/1"
}

store-document

Creates or updates a document.

{
  "document": {
    "name": "John Doe",
    "email": "john@example.com",
    "department": "Engineering"
  },
  "id": "employees/1"  // Optional, will be generated if not provided
}

delete-document

Deletes a document by ID.

{
  "id": "employees/1"
}

Query Operations

query-documents

Executes RQL queries with results handling.

{
  "query": "from Employees where department = 'Engineering'"
}

Example Usage

Here's a typical workflow for interacting with the RavenDB MCP server through an AI assistant:

1. Connect to the database

   Use the initialize-connection tool to connect to your RavenDB server

2. Explore the database structure

   Use show-collections to see what collections are available

3. Retrieve documents

   Use get-document to fetch specific documents by ID

4. Run queries

   Use query-documents to execute RQL queries

5. Modify data

   Use store-document to create or update documents
   Use delete-document to remove documents

Development

# Install dependencies
npm install

# Build the project
npm run build

# Run in development mode
npm run dev

# Test with the MCP inspector
npm run inspector

License

MIT