Elasticsearch MCP Server

English | 中文

MCP Server for connecting to your Elasticsearch cluster directly from any MCP Client (like Claude Desktop, Cursor).

This server connects agents to your Elasticsearch data using the Model Context Protocol. It allows you to interact with your Elasticsearch indices through natural language conversations.

Demo

Feature Overview

Available Features

• list_indices: List available Elasticsearch indices, support regex
• get_mappings: Get field mappings for a specific Elasticsearch index
• search: Perform an Elasticsearch search with the provided query DSL
• elasticsearch_health: Get Elasticsearch cluster health status, optionally including index-level details
• create_index: Create Elasticsearch index with optional settings and mappings
• create_mapping: Create or update mapping structure for an Elasticsearch index
• bulk: Bulk data into an Elasticsearch index
• reindex: Reindex data from a source index to a target index with optional query and script
• create_index_template: Create or update an index template
• get_index_template: Get information about index templates
• delete_index_template: Delete an index template

How It Works

1. The MCP Client analyzes your request and determines which Elasticsearch operations are needed
2. The MCP server carries out these operations (listing indices, fetching mappings, performing searches)
3. The MCP Client processes the results and presents them in a user-friendly format

Getting Started

Prerequisites

• An Elasticsearch instance
• Elasticsearch authentication credentials (API key or username/password)
• MCP Client (e.g. Claude Desktop, Cursor)

Installation & Setup

Using the Published NPM Package (coming soon)

[!TIP] The easiest way to use Elasticsearch MCP Server is through the published npm package.

1. Configure MCP Client

   • Open your MCP Client. See the list of MCP Clients, here we are configuring Claude Desktop.
   • Go to Settings > Developer > MCP Servers
   • Click Edit Config and add a new MCP Server with the following configuration:

   {
     "mcpServers": {
       "elasticsearch-mcp": {
         "command": "npx",
         "args": [
           "-y",
           "@awesome-ai/elasticsearch-mcp"
         ],
         "env": {
           "HOST": "your-elasticsearch-host",
           "API_KEY": "your-api-key"
         }
       }
     }
   }
   

2. Start a Conversation

   • Open a new conversation in your MCP Client
   • The MCP server should connect automatically
   • You can now ask questions about your Elasticsearch data

Configuration Options

The Elasticsearch MCP Server supports configuration options to connect to your Elasticsearch:

[!NOTE] You must provide either an API key or both username and password for authentication.

Environment Variable	Description	Required
HOST	Your Elasticsearch instance URL	Yes
API_KEY	Elasticsearch API key for authentication	No
USERNAME	Elasticsearch username for basic authentication	No
PASSWORD	Elasticsearch password for basic authentication	No
CA_CERT	Path to custom CA certificate for Elasticsearch SSL/TLS	No

Local Development

[!NOTE] If you want to modify or extend the MCP Server, follow these local development steps.

1. Use the correct Node.js version

   nvm use
   

2. Install Dependencies

   npm install
   

3. Build the Project

   npm run build
   

4. Run locally in Claude Desktop App, (also support Cusor)

   • Open Claude Desktop App
   • Go to Settings > Developer > MCP Servers
   • Click Edit Config and add a new MCP Server with the following configuration:

   {
     "mcpServers": {
       "elasticsearch-mcp": {
         "command": "node",
         "args": [
           "/path/to/your/project/dist/index.js"
         ],
         "env": {
           "HOST": "your-elasticsearch-host",
           "API_KEY": "your-api-key"
         }
       }
     }
   }
   

5. Debugging with MCP Inspector

   HOST=your-elasticsearch-url API_KEY=your-api-key npm run inspector
   

   This will start the MCP Inspector, allowing you to debug and analyze requests. You should see:

   Starting MCP inspector...
   Proxy server listening on port 3000
   
   🔍 MCP Inspector is up and running at http://localhost:5173 🚀
   

Example Queries

[!TIP] Here are some natural language queries you can try with your MCP Client.

• "What indices do I have in my Elasticsearch cluster?"
• "Show me the field mappings for the 'products' index."
• "Find all orders over $500 from last month."
• "Which products received the most 5-star reviews?"
• "What is the health status of my Elasticsearch cluster?"
• "Create a new index called 'users' with 3 shards and 1 replica."
• "Add a keyword type field called 'tags' to the 'products' index."
• "Bulk import these customer records into the 'customers' index."
• "Reindex data from 'old_index' to 'new_index'."
• "Create an index template for logs with pattern 'logs-*'."
• "Show me all my index templates."
• "Delete the 'outdated_template' index template."

If you encounter issues, feel free to open an issue on the GitHub repository.

Inspired by

• mcp-server-elasticsearch.