Elasticsearch MCP Server

Name: Elasticsearch MCP Server
Author: awesimon

@awesimon

Elasticsearch MCP Server について

Elasticsearch MCP server with available features including mappings management, search and indexing, and index management etc.

#elasticsearch #mcp #mcp-server

基本情報

カテゴリ

データベース

ライセンス

MIT

ランタイム

node

トランスポート

stdio

公開者

awesimon

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

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

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Elasticsearch MCP Server?

A Model Context Protocol server that connects MCP Clients—such as Claude Desktop or Cursor—to an Elasticsearch cluster, enabling users to interact with their indices and data through natural language conversations.

How to use Elasticsearch MCP Server?

Configure your MCP Client using npx with the published npm package (@awesome-ai/elasticsearch-mcp), providing your Elasticsearch host and authentication credentials via environment variables (ES_HOST, ES_API_KEY or ES_USERNAME/ES_PASSWORD). After setup, the server connects automatically, and you can query your cluster using everyday language.

Key features of Elasticsearch MCP Server

Check cluster health with optional index-level details
List, create, and reindex Elasticsearch indices
Get and create field mappings for indices
Search Elasticsearch with query DSL
Bulk-import data into an index
Create, list, and delete index templates

Use cases of Elasticsearch MCP Server

Ask for the health status and active node count of your cluster
List all indices or create new ones with custom settings and mappings
Reindex data from one index to another
Retrieve field mappings and add new fields to existing indices
Perform natural language searches on your Elasticsearch data
Manage index templates for automated index configuration

FAQ from Elasticsearch MCP Server

What do I need to run Elasticsearch MCP Server?

You need a running Elasticsearch instance, authentication credentials (an API key or a username/password pair), and an MCP Client such as Claude Desktop or Cursor.

How do I authenticate with my Elasticsearch cluster?

You must provide either an ES_API_KEY or both ES_USERNAME and ES_PASSWORD via environment variables. The configuration accepts legacy variable names as fallbacks.

Can I connect to multiple Elasticsearch nodes?

Yes. Set the ES_HOST environment variable to a comma-separated list of node URLs. The client automatically handles failover and load balancing across the configured nodes.

How do I run the server from source for development?

Clone the repository, install dependencies (npm install), build the project (npm run build), then point your MCP Client configuration to the built dist/index.js file using the node command.

Is there a way to debug or test the server locally?

Yes, use the MCP Inspector. Run ES_HOST=your-url ES_API_KEY=your-key npm run inspector, then access the inspector UI at http://127.0.0.1:6274 to inspect requests and responses.