elasticsearch
Elasticsearch MCP server with available features including mappings management, search and indexing, and index management etc.
claude mcp add --transport stdio awesimon-elasticsearch-mcp npx -y @awesome-ai/elasticsearch-mcp \ --env ES_HOST="your-elasticsearch-host" \ --env ES_API_KEY="your-api-key"
How to use
This MCP server enables natural language access to your Elasticsearch cluster. It exposes a set of MCP operations that map to common Elasticsearch actions such as health checks, index management, mappings, searches, and bulk data operations. You can interact with your Elasticsearch data through an MCP client (for example Claude Desktop or Cursor) by asking natural language questions. The server translates those requests into Elasticsearch API calls and returns structured results that you can present to users in a readable format.
How to install
Prerequisites:
- Node.js and npm installed on your machine
- Access to an Elasticsearch cluster (with suitable credentials)
Step-by-step installation:
- Install the MCP server package via npx (as shown in the example configuration):
# This assumes you have a working MCP client setup and will configure it to point to the MCP package
- Install dependencies and prepare your environment:
# If you are developing locally, ensure you have Node.js v18+ (or the version required by the package)
npm install
- Start using the MCP server from your MCP client:
{
"mcpServers": {
"elasticsearch-mcp": {
"command": "npx",
"args": [
"-y",
"@awesome-ai/elasticsearch-mcp"
],
"env": {
"ES_HOST": "your-elasticsearch-host",
"ES_API_KEY": "your-api-key"
}
}
}
}
- Verify connectivity by opening your MCP client and starting a new conversation with a question about your Elasticsearch data. The MCP client should connect to the server automatically if configured correctly.
Additional notes
Notes and tips:
- You must provide authentication via ES_API_KEY or a combination of ES_USERNAME and ES_PASSWORD as described in the configuration notes.
- The ES_HOST environment variable supports a single URL or a comma-separated list for multiple nodes (with automatic failover).
- You can customize additional TLS options by exposing ES_CA_CERT if your cluster uses a self-signed certificate.
- For multiple Elasticsearch nodes, include all hosts in ES_HOST, e.g. "https://es-node1:9200,https://es-node2:9200".
- When testing locally, you can use the MCP Inspector to debug requests and view the request/response flow.
- The available MCP capabilities cover cluster health, index operations (list, create, reindex), mappings (get, create), searches, bulk writes, and index template management.
Related MCP Servers
context7
Context7 MCP Server -- Up-to-date code documentation for LLMs and AI code editors
obsidian -tools
Add Obsidian integrations like semantic search and custom Templater prompts to Claude or any MCP client.
MiniMax -JS
Official MiniMax Model Context Protocol (MCP) JavaScript implementation that provides seamless integration with MiniMax's powerful AI capabilities including image generation, video generation, text-to-speech, and voice cloning APIs.
mcp-bundler
Is the MCP configuration too complicated? You can easily share your own simplified setup!
akyn-sdk
Turn any data source into an MCP server in 5 minutes. Build AI-agents-ready knowledge bases.
promptboard
The Shared Whiteboard for Your AI Agents via MCP. Paste screenshots, mark them up, and share with AI.