Elasticsearch MCP Server

Enhanced Elasticsearch MCP Server Solution - Security & Threat Analysis Focused

This is a professional security-focused solution maintained by TocharianOU. It enables comprehensive interaction with all Elasticsearch APIs, specifically optimized for security analysis, threat detection, and incident investigation. Features include advanced security monitoring, anomaly detection, threat hunting, root cause analysis, and comprehensive audit capabilities.

Key Security Features:

• Real-time threat detection and security monitoring
• Advanced machine learning for anomaly detection
• Root cause analysis and attack chain tracking
• Security incident investigation and forensics
• Compliance monitoring and audit reporting

Note: This solution requires a valid Elasticsearch license (trial, platinum, or enterprise) and is designed for security professionals, SOC teams, and threat analysts.

Connect to your Elasticsearch data directly from any MCP Client (such as Claude Desktop) using the Model Context Protocol (MCP). Interact with your Elasticsearch security data through natural language queries for advanced threat analysis and incident response.

Prerequisites

• An Elasticsearch instance
• A valid Elasticsearch license (trial, platinum, enterprise) is required.
• Elasticsearch authentication credentials (API key or username/password)
• MCP Client (e.g. Claude Desktop) or HTTP client for remote access

⚠️ This project requires your Elasticsearch cluster to have a valid license. If you do not have a license, you can activate a trial license as shown below.

Multi-Version Elasticsearch Support

Automatically supports Elasticsearch 5.x - 9.x with intelligent version detection:

Key Features:

• Automatic version detection - No manual configuration needed
• Smart client selection - Loads the right client for your ES version
• Adaptive features - Disables unsupported tools (e.g., Data Streams on ES < 7.9, ES|QL on ES < 8.11)
• Version-specific optimizations - Handles API differences transparently

What happens:

Connect → Detect ES version → Load matching client → Register compatible tools

SSL/TLS Connection

To connect to Elasticsearch with a self-signed certificate or in a test environment, you can set the following environment variable:

NODE_TLS_REJECT_UNAUTHORIZED=0

⚠️ This disables Node.js SSL certificate validation. Use only in development or testing environments. For production, always use a trusted CA certificate.

Installation & Setup

1. Start a Conversation
   • Open a new conversation in your MCP Client
   • The MCP server should connect automatically
   • You can now ask questions about your Elasticsearch data

Configuration Options

The Elasticsearch MCP Server supports the following configuration options:

Elasticsearch Configuration

Transport Mode Configuration (NEW in v0.3.0)

Transport Mode Details:

• Stdio mode (default): For Claude Desktop and local MCP clients
• HTTP Streamable mode: Runs as a standalone HTTP server for remote access, API integration, and web applications

Quick Start

Option 1: NPM Installation (Recommended)

1. Install globally via NPM

   npm install -g @tocharianou/elasticsearch-mcp
   

2. Run directly

   npx @tocharianou/elasticsearch-mcp
   

Option 2: GitHub Release (Standalone Package)

1. Download release package

   • Go to GitHub Releases
   • Download the latest .tar.gz file and its checksum files (.sha256 and .sha512)

2. Verify package integrity

   shasum -a 256 -c elasticsearch-mcp-v*.tar.gz.sha256
   # Should output: elasticsearch-mcp-v*.tar.gz: OK
   

3. Extract and use

   mkdir elasticsearch-mcp && cd elasticsearch-mcp
   tar -xzf ../elasticsearch-mcp-v*.tar.gz
   
   # Run with your Elasticsearch credentials
   ES_URL=https://localhost:9200 ES_API_KEY=your-key node dist/index.js
   

Option 3: Source Installation

1. Clone the repository

   git clone https://github.com/TocharianOU/elasticsearch-mcp.git
   cd elasticsearch-mcp
   

2. Install Dependencies

   npm install
   

3. Build the Project

   npm run build
   

4. Configure Claude Desktop App

   • Open Claude Desktop App
   • Go to Settings > Developer > MCP Servers
   • Click Edit Config and add a new MCP Server with the following configuration:

   For NPM Installation:

   {
     "mcpServers": {
       "elasticsearch-mcp-server": {
         "command": "npx",
         "args": [
           "@tocharianou/elasticsearch-mcp"
         ],
         "env": {
           "ES_URL": "your-elasticsearch-url",
           "ES_USERNAME": "elastic",
           "ES_PASSWORD": "your_pass",
           "NODE_TLS_REJECT_UNAUTHORIZED": "0"
         }
       }
     }
   }
   

   For Source Installation:

   {
     "mcpServers": {
       "elasticsearch-mcp-server-local": {
         "command": "node",
         "args": [
           "/path/to/your/elasticsearch-mcp/dist/index.js"
         ],
         "env": {
           "ES_URL": "your-elasticsearch-url",
           "ES_USERNAME": "elastic",
           "ES_PASSWORD": "your_pass",
           "NODE_TLS_REJECT_UNAUTHORIZED": "0"
         }
       }
     }
   }
   

5. Debugging with MCP Inspector

   ES_URL=your-elasticsearch-url ES_USERNAME=elastic ES_PASSWORD=your_pass npm run inspector
   

   This will start the MCP Inspector, allowing you to debug and analyze requests. You should see:

   Starting MCP inspector...
   Proxy server listening on port 3000
   
   MCP Inspector is up and running at http://localhost:5173
   

Method 3: HTTP Streamable Mode (NEW in v0.3.0)

Run the server as a standalone HTTP service for remote access and API integration:

# Start HTTP server (default port 3000)
MCP_TRANSPORT=http \
ES_URL=your-elasticsearch-url \
ES_USERNAME=elastic \
ES_PASSWORD=your_pass \
npx @tocharianou/elasticsearch-mcp

# Or with custom port and host
MCP_TRANSPORT=http \
MCP_HTTP_PORT=9000 \
MCP_HTTP_HOST=0.0.0.0 \
ES_URL=your-elasticsearch-url \
ES_USERNAME=elastic \
ES_PASSWORD=your_pass \
npx @tocharianou/elasticsearch-mcp

HTTP Streamable Mode Features:

• Exposes MCP server at http://host:port/mcp endpoint
• Health check available at http://host:port/health
• Session-based connection management
• Supports both POST (JSON-RPC requests) and GET (SSE streams)
• Compatible with any HTTP client or MCP SDK

Example HTTP client usage:

// Initialize connection
const response = await fetch('http://localhost:3000/mcp', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    jsonrpc: '2.0',
    method: 'initialize',
    params: {
      protocolVersion: '2024-11-05',
      capabilities: {},
      clientInfo: { name: 'my-client', version: '1.0.0' }
    },
    id: 1
  })
});

const sessionId = response.headers.get('mcp-session-id');

// Subsequent requests include session ID
const toolsResponse = await fetch('http://localhost:3000/mcp', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'mcp-session-id': sessionId
  },
  body: JSON.stringify({
    jsonrpc: '2.0',
    method: 'tools/list',
    params: {},
    id: 2
  })
});

// Call a tool (e.g., list_indices)
const indicesResponse = await fetch('http://localhost:3000/mcp', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'mcp-session-id': sessionId
  },
  body: JSON.stringify({
    jsonrpc: '2.0',
    method: 'tools/call',
    params: {
      name: 'list_indices',
      arguments: {}
    },
    id: 3
  })
});

Available Tools

Tools not supported by your cluster version are automatically skipped at startup.

ES|QL Query Tool (esql_query)

ES|QL is Elasticsearch's modern pipe-based query language, ideal for analytics and data exploration without complex JSON DSL.

Example queries:

FROM logs-* | WHERE level == "error" | STATS count = COUNT(*) BY service | SORT count DESC | LIMIT 20
FROM metrics-* | WHERE @timestamp > NOW() - 1 hour | STATS avg_cpu = AVG(cpu.usage) BY host.name
FROM auditbeat-* | WHERE event.action == "user_login" AND event.outcome == "failure" | LIMIT 50

Parameters:

• query — the ES|QL string (required)
• params — positional parameters replacing ? placeholders (optional)
• include_types — include column type info in output (optional, default false)
• break_token_rule — bypass token limit for large results (optional, default false)

Automatically registered only on ES 8.11+ clusters.

Contributing

We welcome contributions from the community! For details on how to contribute, please see Contributing Guidelines.

How It Works

1. The MCP Client analyzes your request and determines which Elasticsearch operations are needed.
2. The MCP server comunicate with ES.
3. The MCP Client processes the results and presents them in a user-friendly format, including highlights, aggregation summaries, and anomaly insights.

Security Analysis Examples

[!TIP] Here are security-focused queries you can try with your MCP Client.

Threat Detection:

• "Analyze brute force attack attempts in the past 24 hours"
• "Detect abnormal login behavior and suspicious IP addresses in the system"
• "Identify potential SQL injection attack patterns and malicious requests"
• "Discover DDoS attack signatures and traffic anomalies in network flows"

Root Cause Analysis:

• "Trace the complete attack chain and impact scope for specific security incidents"
• "Analyze root causes and propagation paths of system failures"
• "Identify data breach sources and involved sensitive information"
• "Investigate user privilege abuse incidents with timeline and operation records"

Threat Intelligence:

• "Create machine learning models to detect zero-day attacks and unknown threats"
• "Establish behavioral baselines and identify activities deviating from normal patterns"
• "Analyze threat levels and attack history of malicious domains and IP addresses"
• "Detect behavioral characteristics and attack patterns of Advanced Persistent Threats (APT)"

Real-time Monitoring:

• "Monitor active threats and ongoing attacks in the current system"
• "Detect abnormal data access patterns and privilege escalation behaviors"
• "Discover suspicious network communications and data exfiltration activities"
• "Identify security causes of abnormal system resource consumption and performance degradation"

Security Best Practices

[!WARNING] Avoid using cluster-admin privileges. Create dedicated API keys with limited scope and apply fine-grained access control at the index level to prevent unauthorized data access.

Package Integrity Verification

When downloading release packages, always verify checksums to ensure integrity:

# Verify SHA256 checksum
shasum -a 256 -c elasticsearch-mcp-vX.Y.Z.tar.gz.sha256

# Verify SHA512 checksum
shasum -a 512 -c elasticsearch-mcp-vX.Y.Z.tar.gz.sha512

This protects against:

• Corrupted downloads
• Tampered packages
• Man-in-the-middle attacks

Elasticsearch Access Control

You can create a dedicated Elasticsearch API key with minimal permissions to control access to your data:

{
  "name": "es-mcp-server-access",
  "role_descriptors": {
    "mcp_server_role": {
      "cluster": [
        "monitor"
      ],
      "indices": [
        {
          "names": [
            "index-1",
            "index-2",
            "index-pattern-*"
          ],
          "privileges": [
            "read",
            "view_index_metadata"
          ]
        }
      ]
    }
  }
}

License

This project is licensed under the Apache License 2.0.

Troubleshooting

• Ensure your MCP configuration is correct.
• Verify that your Elasticsearch URL is accessible from your machine.
• Check that your authentication credentials (API key or username/password) have the necessary permissions.
• If using SSL/TLS with a custom CA, verify that the certificate path is correct and the file is readable.
• Look at the terminal output for error messages.

If you encounter issues, feel free to open an issue on the GitHub repository.

Running with a Trial License

If your Elasticsearch cluster does not have a valid license, you can activate a 30-day trial license with the following command:

curl -X POST -u elastic:your_password \
  -k "https://your-es-host:9200/_license/start_trial?acknowledge=true"

• Replace your_password and your-es-host with your actual credentials and host.
• This will enable all features for 30 days.

Note: This project will not start if your cluster does not have a valid license (trial, platinum, enterprice etc.).