Overview

The Elasticsearch MCP Server provides a robust bridge for AI models to interact with Elasticsearch and OpenSearch. It enables searching documents, analyzing indices, and managing clusters through a standardized set of tools, making it a crucial component for AI-driven data exploration and management.

Created by:

Developed by cr7258

Key Features

🔍

Document Search & Management

Search, index, get, and delete documents within Elasticsearch indices

📊

Index & Cluster Operations

Manage indices (create, delete, get mappings) and monitor cluster health

🔗

Alias Management

Create, update, and delete index aliases for flexible data routing

🛡️

Security & Control

Option to disable high-risk write operations for enhanced safety

Available Tools

Quick Reference

Tool	Purpose	Category
`general_api_request`	Perform general HTTP API requests	General
`list_indices`	List all available indices	Index
`get_index`	Get information about indices	Index
`create_index`	Create a new index	Index
`delete_index`	Delete an index	Index
`create_data_stream`	Create a new data stream	Data Stream
`get_data_stream`	Get information about data streams	Data Stream
`delete_data_stream`	Delete data streams	Data Stream
`search_documents`	Search for documents	Document
`index_document`	Create or update a document	Document
`get_document`	Get a document by ID	Document
`delete_document`	Delete a document by ID	Document
`delete_by_query`	Delete documents by query	Document
`get_cluster_health`	Get cluster health information	Cluster
`get_cluster_stats`	Get cluster statistics	Cluster
`list_aliases`	List all aliases	Alias
`get_alias`	Get alias information	Alias
`put_alias`	Create or update an alias	Alias
`delete_alias`	Delete an alias	Alias

Detailed Usage

search_documents▶

Search for documents within specified indices using a query DSL.

use_mcp_tool({
  server_name: "elasticsearch",
  tool_name: "search_documents",
  arguments: {
    index: "my_index",
    query: {
      match: {
        message: "hello world"
      }
    }
  }
});

Returns matching documents and their metadata.

index_document▶

Creates or updates a document in the specified index.

use_mcp_tool({
  server_name: "elasticsearch",
  tool_name: "index_document",
  arguments: {
    index: "my_index",
    id: "1",
    document: {
      user: "John Doe",
      message: "Hello from MCP"
    }
  }
});

If ID is provided, updates existing document; otherwise, creates a new one.

get_cluster_health▶

Returns basic information about the health of the Elasticsearch cluster.

use_mcp_tool({
  server_name: "elasticsearch",
  tool_name: "get_cluster_health",
  arguments: {}
});

Provides status (green, yellow, red), node count, and shard information.

list_indices▶

List all available Elasticsearch indices.

use_mcp_tool({
  server_name: "elasticsearch",
  tool_name: "list_indices",
  arguments: {}
});

Returns a list of all indices in the cluster.

Installation

{
  "mcpServers": {
    "elasticsearch": {
      "command": "uvx",
      "args": [
        "elasticsearch-mcp-server"
      ],
      "env": {
        "ELASTICSEARCH_HOSTS": "https://localhost:9200",
        "ELASTICSEARCH_USERNAME": "elastic",
        "ELASTICSEARCH_PASSWORD": "your_password"
      }
    }
  }
}

API Key Authentication:

Alternatively, use ELASTICSEARCH_API_KEY for authentication:

{
  "mcpServers": {
    "elasticsearch": {
      "command": "uvx",
      "args": [
        "elasticsearch-mcp-server"
      ],
      "env": {
        "ELASTICSEARCH_HOSTS": "https://localhost:9200",
        "ELASTICSEARCH_API_KEY": "YOUR_ELASTICSEARCH_API_KEY"
      }
    }
  }
}

Common Use Cases

1. Document Search

Search for specific documents based on keywords or complex queries:

use_mcp_tool({
  server_name: "elasticsearch",
  tool_name: "search_documents",
  arguments: {
    index: "products",
    query: {
      bool: {
        must: { match: { name: "laptop" } },
        filter: { range: { price: { gte: 500, lte: 1500 } } }
      }
    }
  }
});

2. Indexing New Data

Add new data to an Elasticsearch index:

use_mcp_tool({
  server_name: "elasticsearch",
  tool_name: "index_document",
  arguments: {
    index: "logs",
    document: {
      timestamp: new Date().toISOString(),
      level: "info",
      message: "User logged in successfully"
    }
  }
});

3. Monitoring Cluster Health

Periodically check the health of the Elasticsearch cluster:

use_mcp_tool({
  server_name: "elasticsearch",
  tool_name: "get_cluster_health",
  arguments: {}
});

4. Managing Index Aliases

Create an alias for an index to simplify queries or manage reindexing:

use_mcp_tool({
  server_name: "elasticsearch",
  tool_name: "put_alias",
  arguments: {
    index: "products_v1",
    name: "current_products"
  }
});

Connection String Format

The Elasticsearch MCP server primarily uses environment variables for connection details, but the underlying Elasticsearch client supports various connection methods.

Host: ELASTICSEARCH_HOSTS (e.g., https://localhost:9200)
Authentication:
- ELASTICSEARCH_USERNAME and ELASTICSEARCH_PASSWORD for basic auth.
- ELASTICSEARCH_API_KEY for API key authentication.
SSL Verification: ELASTICSEARCH_VERIFY_CERTS (default: false)

Sources

Google Search MCP Server

An advanced Model Context Protocol (MCP) server that provides comprehensive Google search capabilities, webpage content extraction, and AI-powered research synthesis.

S3 MCP Server

S3 MCP servers enable AI models to interact with Amazon S3 object storage, providing capabilities for file operations, metadata management, and versioning in a secure and scalable environment.

Pinecone MCP Server

Integrate Pinecone with your AI assistants using the Model Context Protocol (MCP) for enhanced vector database interactions.

Elasticsearch MCP Server