Elasticsearch MCP Server

A Model Context Protocol (MCP) server that exposes Elasticsearch semantic search capabilities as MCP tools. This server is optimized for Elasticsearch Serverless and managed Elastic Cloud clusters.

🚀 Features

• Semantic Search: Use ELSER (sparse) or kNN (dense) directly within Elasticsearch.
• Hybrid Search: Combine keyword (BM25) and semantic scores.
• Automatic Setup: Tools to create indices with the correct mappings and ingest pipelines in one click.
• Serverless Ready: Fully compatible with Elasticsearch Serverless and API Key authentication.

🛠️ Quickstart

1. Configure Environment

Edit the .env file with your connection details:

ELASTICSEARCH_HOSTS=https://your-serverless-endpoint.es.us-east-1.aws.elastic.cloud:443
ELASTICSEARCH_API_KEY=your_api_key

2. Verify Connection

Run the verification suite to ensure your cluster is reachable and tools are registered:

.\venv\Scripts\python.exe verify_server.py

2. Run the Web API (Standard REST)

Start the web server with a simple command:

elasticsearch-api

• API URL: http://localhost:8000
• Interactive Docs: http://localhost:8000/docs

3. Run the MCP Server (For AI Editors)

Start the MCP server with a simple command:

elasticsearch-mcp

🧩 Claude Desktop Integration

Add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "elasticsearch": {
      "command": "C:\\Users\\birar\\Desktop\\elasticsearch-mcp-server\\venv\\Scripts\\python.exe",
      "args": ["-m", "elasticsearch_mcp"],
      "env": {
        "ELASTICSEARCH_HOSTS": "https://your-cluster-url.es.aws.elastic.cloud:443",
        "ELASTICSEARCH_API_KEY": "your_api_key_here"
      }
    }
  }
}

🛠️ Available Tools

• es_ping: Check connectivity.
• es_setup_elser_index: Create a semantic index for ELSER.
• es_setup_dense_index: Create a semantic index for dense vectors (kNN).
• es_index_document: Index data through a pipeline.
• es_semantic_search_elser: Perform sparse semantic search.
• es_semantic_search_knn: Perform dense kNN search.
• es_semantic_search_hybrid: Combined keyword and semantic search.
• es_delete_index: Safely remove indices.

⚠️ Requirements

• Python 3.11+
• Elasticsearch 8.8+ (including Serverless)
• ML Models: ELSER or E5 must be available/deployed in your cluster for semantic search tools to function.