Cr7258 Elasticsearch MCP Server

🚀 Elasticsearch MCP Server

An MCP (Model Context Protocol) server that interacts with Elasticsearch. This server provides functions for searching documents, analyzing indexes, and managing clusters through a series of tools.

https://github.com/user-attachments/assets/f7409e31-fac4-4321-9c94-b0ff2ea7ff15

✨ Features

Index Operations

• list_indices: List all indexes in the Elasticsearch cluster.
• get_mapping: Get the mapping configuration of a specific index.
• get_settings: Get the settings configuration of a specific index.

Document Operations

• search_documents: Search for documents in the index using the Elasticsearch query DSL.

Cluster Operations

• get_cluster_health: Get the health status of the cluster.
• get_cluster_stats: Get the statistical information of the cluster.

📦 Installation

Start the Elasticsearch Cluster

Start the Elasticsearch cluster using Docker Compose:

docker-compose up -d

This will start a 3-node Elasticsearch cluster and Kibana. The default Elasticsearch username is elastic, and the password is test123.

You can access Kibana at https://localhost:5601.

💻 Usage Examples

Use Claude Desktop

1. Add the following configuration to the Claude Desktop configuration file claude_desktop_config.json:

{
  "mcpServers": {
    "elasticsearch": {
      "command": "uv",
      "args": [
        "--directory",
        "path/to/elasticsearch_mcp_server/src",
        "run",
        "server.py"
      ],
      "env": {
        "ELASTIC_HOST": "<your_elastic_url>",
        "ELASTIC_USERNAME": "<your_elastic_username>",
        "ELASTIC_PASSWORD": "<your_elastic_password>>"
      }
    }
  }
}

• On macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• On Windows: %APPDATA%/Claude/claude_desktop_config.json%

3. Restart Claude Desktop to load the new MCP server.

Now you can interact with your Elasticsearch cluster using natural language commands through Claude, for example:

• "List all indexes in the cluster"
• "How old is Bob?"
• "Show the health status of the cluster"