Automation and SOAR

Automation is critical for scaling SOC operations and reducing response times. This layer combines security orchestration (Cortex SOAR) with infrastructure automation (Terraform/PyInfra) to enable rapid, consistent responses to security events and efficient infrastructure management.

Automation reduces manual work, eliminates human error, and enables your team to focus on complex investigations while repetitive tasks are handled automatically.

Architecture Overview

Cortex SOAR

Security orchestration for investigation and response automation

Infrastructure as Code

Terraform and PyInfra for automated infrastructure management

Cortex SOAR Platform

Cortex is the automation engine that powers TheHive, providing analyzers for enrichment and responders for automated actions.

Core Concepts

Analyzers
Responders
Jobs

Observable Enrichment:Analyzers automatically enrich observables with contextual information:

Threat Intelligence: VirusTotal, MISP, OTX
Reputation: AbuseIPDB, URLhaus, PhishTank
Sandboxing: Joe Sandbox, Cuckoo, ANY.RUN
DNS/WHOIS: PassiveTotal, DomainTools
File Analysis: YARA, ClamAV, PEFile
Network: Shodan, Censys, MaxMind GeoIP

Analyzer Configuration

Start with free analyzers (AbuseIPDB, OTX, MISP) and gradually add commercial services (VirusTotal, PassiveTotal) based on investigation needs and budget.

Analyzer Types by Data Type

IP Address Analyzers:

AbuseIPDB_1: IP reputation and reports
GreyNoise: Internet scanner identification
IPVoid: Multi-engine reputation check
MaxMind GeoIP: Geographic location
Shodan: Internet exposure scanning
Tor Project: Tor exit node detection

File Hash Analyzers:

VirusTotal: Multi-AV scanning
MISP: Threat intelligence correlation
MalwareBazaar: Known malware repository
HybridAnalysis: Sandbox analysis

URL/Domain Analyzers:

URLhaus: Malware distribution URLs
PhishTank: Phishing URL database
Google Safe Browsing: URL safety
Censys: Certificate and host analysis

Analyzer Configuration Example

// VirusTotal analyzer configuration
{
  "name": "VirusTotal_GetReport_3_0",
  "version": "3.0",
  "description": "Get VirusTotal report for file, hash, domain, IP, or URL",
  "dataTypeList": ["file", "hash", "domain", "ip", "url"],
  "baseConfig": "VirusTotal",
  "config": {
    "service": "get_report"
  },
  "configurationItems": [
    {
      "name": "key",
      "description": "VirusTotal API key",
      "type": "string",
      "multi": false,
      "required": true
    },
    {
      "name": "polling_interval",
      "description": "Polling interval (seconds)",
      "type": "number",
      "multi": false,
      "required": false,
      "defaultValue": 60
    }
  ]
}

Custom Analyzer Development

Create custom analyzers for internal tools:

#!/usr/bin/env python3
from cortexutils.analyzer import Analyzer

class InternalThreatIntel(Analyzer):
    def __init__(self):
        Analyzer.__init__(self)
        self.service = self.get_param('config.service', None)
        self.api_url = self.get_param('config.url', None)
        self.api_key = self.get_param('config.key', None)
    
    def summary(self, raw):
        taxonomies = []
        level = "info"
        namespace = "InternalTI"
        predicate = "Threat"
        
        if raw.get('is_malicious'):
            level = "malicious"
            value = "Known Threat"
        else:
            level = "safe"
            value = "Clean"
        
        taxonomies.append(self.build_taxonomy(
            level, namespace, predicate, value
        ))
        
        return {"taxonomies": taxonomies}
    
    def run(self):
        data = self.get_data()
        
        # Query internal threat intel database
        result = self.check_internal_ti(data)
        
        self.report(result)
    
    def check_internal_ti(self, observable):
        # Implementation to check internal systems
        return {
            'is_malicious': False,
            'last_seen': None,
            'related_incidents': []
        }

if __name__ == '__main__':
    InternalThreatIntel().run()

Responder Configuration

Responders can make destructive changes (block IPs, quarantine hosts, disable accounts). Always test in a lab environment and implement approval workflows for critical actions.

Network Responders
Endpoint Responders
Notification Responders

Firewall Integration:

#!/usr/bin/env python3
# OPNsense firewall block IP responder
from cortexutils.responder import Responder
import requests

class OPNsenseBlockIP(Responder):
    def __init__(self):
        Responder.__init__(self)
        self.firewall_url = self.get_param('config.url')
        self.api_key = self.get_param('config.api_key')
        self.api_secret = self.get_param('config.api_secret')
    
    def run(self):
        data = self.get_param('data', None, 'Missing data field')
        
        if data.get('dataType') == 'ip':
            ip_address = data.get('data')
            
            # Add IP to blocklist
            result = self.block_ip(ip_address)
            
            if result:
                self.report({'message': f'Blocked {ip_address}'})
            else:
                self.error('Failed to block IP')
        else:
            self.error('Responder works only with IP observables')
    
    def block_ip(self, ip):
        # Implementation to block IP on firewall
        endpoint = f"{self.firewall_url}/api/firewall/alias/addItem"
        
        payload = {
            'alias': 'BlockedIPs',
            'address': ip,
            'description': f'Blocked by Cortex - {self.get_param("title")}'
        }
        
        response = requests.post(
            endpoint,
            json=payload,
            auth=(self.api_key, self.api_secret),
            verify=False
        )
        
        return response.status_code == 200

if __name__ == '__main__':
    OPNsenseBlockIP().run()

Host Quarantine:

# Wazuh active response integration
class WazuhQuarantineHost(Responder):
    def run(self):
        hostname = self.get_param('data.data')
        
        # Get Wazuh agent ID
        agent_id = self.get_agent_id(hostname)
        
        if agent_id:
            # Execute active response
            self.quarantine_agent(agent_id)
            self.report({'message': f'Quarantined {hostname}'})
        else:
            self.error(f'Agent not found: {hostname}')
    
    def quarantine_agent(self, agent_id):
        # Execute Wazuh active response
        # Block network, disable services, etc.
        pass

Slack Integration:

# Send case update to Slack
class SlackNotification(Responder):
    def run(self):
        webhook_url = self.get_param('config.webhook_url')
        
        case_title = self.get_param('data.title')
        case_severity = self.get_param('data.severity')
        
        message = {
            'text': f'Security Incident: {case_title}',
            'attachments': [{
                'color': self.get_color(case_severity),
                'fields': [
                    {'title': 'Severity', 'value': case_severity},
                    {'title': 'Status', 'value': self.get_param('data.status')}
                ]
            }]
        }
        
        requests.post(webhook_url, json=message)
        self.report({'message': 'Notification sent'})

Automation Workflows

Chain multiple analyzers and responders for complete automation:

Automated Triage Workflow

Observable Added → Automatic Enrichment:

IP address added to case
Run AbuseIPDB analyzer (reputation)
Run GreyNoise analyzer (scanner detection)
Run Shodan analyzer (exposure check)
Run MaxMind analyzer (geolocation)
If malicious score > threshold:
- Run firewall block responder
- Send Slack notification
- Create blocking task

Phishing Response Workflow

Email Observable → Automated Response:

URL extracted from phishing email
Run URLhaus analyzer
Run PhishTank analyzer
Run VirusTotal analyzer
If confirmed phishing:
- Report to email gateway
- Block sender domain
- Search for similar emails
- Notify affected users

Malware Detection Workflow

File Hash → Sandbox Analysis:

File hash observable created
Run VirusTotal analyzer
If unknown or suspicious:
- Submit to Joe Sandbox
- Submit to Hybrid Analysis
If confirmed malicious:
- Quarantine affected hosts
- Search for hash in environment
- Update antivirus signatures
- Block at proxy/firewall

Infrastructure as Code (IaC)

Terraform Automation

Terraform manages SOC infrastructure deployment and configuration:

Infrastructure Provisioning
Network Configuration
State Management

# Terraform configuration for SOC infrastructure
terraform {
  required_providers {
    proxmox = {
      source = "telmate/proxmox"
      version = "2.9.14"
    }
  }
}

# Deploy Wazuh server
resource "proxmox_vm_qemu" "wazuh" {
  name        = "wazuh-server"
  target_node = "proxmox-node-01"
  clone       = "ubuntu-22.04-template"
  
  cores   = 4
  memory  = 8192
  scsihw  = "virtio-scsi-pci"
  
  disk {
    size    = "100G"
    type    = "scsi"
    storage = "local-lvm"
  }
  
  network {
    model  = "virtio"
    bridge = "vmbr0"
  }
  
  # Post-deployment configuration
  provisioner "remote-exec" {
    inline = [
      "curl -sO https://packages.wazuh.com/4.x/wazuh-install.sh",
      "bash wazuh-install.sh -a"
    ]
  }
}

# Deploy Elasticsearch cluster
resource "proxmox_vm_qemu" "elasticsearch" {
  count       = 3
  name        = "elasticsearch-${count.index + 1}"
  target_node = "proxmox-node-0${(count.index % 3) + 1}"
  
  cores  = 8
  memory = 32768
  
  # ... additional configuration
}

# Firewall rules for SOC components
resource "opnsense_firewall_rule" "allow_wazuh_agents" {
  description = "Allow Wazuh agents to manager"
  interface   = "LAN"
  protocol    = "tcp"
  source      = "any"
  destination = var.wazuh_server_ip
  port        = "1514,1515"
  action      = "pass"
}

resource "opnsense_firewall_rule" "block_suspicious_ips" {
  description = "Block IPs flagged by threat intel"
  interface   = "WAN"
  protocol    = "any"
  source      = "alias:BlockedIPs"
  destination = "any"
  action      = "block"
  log         = true
}

# Remote state backend for team collaboration
terraform {
  backend "s3" {
    bucket         = "soc-terraform-state"
    key            = "infrastructure/terraform.tfstate"
    region         = "us-east-1"
    encrypt        = true
    dynamodb_table = "terraform-locks"
  }
}

# Outputs for other modules
output "wazuh_server_ip" {
  value       = proxmox_vm_qemu.wazuh.default_ipv4_address
  description = "Wazuh manager IP address"
}

output "elasticsearch_cluster" {
  value = {
    for idx, vm in proxmox_vm_qemu.elasticsearch :
    idx => vm.default_ipv4_address
  }
  description = "Elasticsearch node IPs"
}

PyInfra Automation

PyInfra provides Python-based configuration management:

Use PyInfra for complex configuration tasks that benefit from Python’s flexibility, while Terraform handles infrastructure provisioning.

System Configuration

# deploy.py - Configure Wazuh agents
from pyinfra.operations import apt, files, systemd

# Install Wazuh agent
apt.packages(
    name="Install Wazuh agent",
    packages=['wazuh-agent'],
    update=True,
)

# Configure agent
files.template(
    name="Configure Wazuh agent",
    src="templates/ossec.conf.j2",
    dest="/var/ossec/etc/ossec.conf",
    manager_ip="{{ wazuh_manager }}",
    mode="640",
)

# Enable and start service
systemd.service(
    name="Start Wazuh agent",
    service="wazuh-agent",
    running=True,
    enabled=True,
    restarted=True,
)

Security Hardening

# harden.py - Security baseline
from pyinfra.operations import apt, files, server

# Install security tools
apt.packages(
    name="Install security packages",
    packages=[
        'fail2ban',
        'aide',
        'auditd',
        'ufw',
    ],
)

# Configure firewall
server.shell(
    name="Configure UFW",
    commands=[
        'ufw default deny incoming',
        'ufw default allow outgoing',
        'ufw allow ssh',
        'ufw --force enable',
    ],
)

# Secure SSH configuration
files.line(
    name="Disable SSH password auth",
    path="/etc/ssh/sshd_config",
    line="PasswordAuthentication no",
    replace="PasswordAuthentication yes",
)

Deployment Automation

# update_ids_rules.py - Update IDS rules across all sensors
from pyinfra import host
from pyinfra.operations import files, systemd

# Download latest Suricata rules
files.download(
    name="Download Suricata rules",
    src="https://rules.emergingthreats.net/open/suricata/emerging.rules.tar.gz",
    dest="/tmp/emerging.rules.tar.gz",
)

# Extract and deploy
server.shell(
    name="Deploy rules",
    commands=[
        'tar -xzf /tmp/emerging.rules.tar.gz -C /etc/suricata/rules/',
        'suricata -T -c /etc/suricata/suricata.yaml',  # Test config
    ],
)

# Reload Suricata
systemd.service(
    name="Reload Suricata",
    service="suricata",
    reloaded=True,
)

Integration Patterns

Automated Incident Response

Detection

IDS/SIEM detects threat → Creates alert

Enrichment

Cortex analyzers gather intelligence

Decision

Automated scoring determines severity

Response

Cortex responders execute containment

Continuous Deployment

Best Practices

SOAR Automation

Start conservative: Begin with analyzers only, add responders gradually
Test thoroughly: Validate in lab before production deployment
Implement approvals: Require human approval for destructive actions
Monitor job failures: Alert on analyzer/responder errors
Document workflows: Maintain runbooks for automated procedures
Rate limiting: Prevent API quota exhaustion

Infrastructure as Code

Version control: Store all Terraform/PyInfra code in Git
Code review: Require peer review for infrastructure changes
State management: Use remote backends with locking
Secrets management: Never commit credentials, use vaults
Modular design: Create reusable modules and roles
Testing: Validate configurations before applying

Security

Least privilege: Grant minimum necessary permissions
Audit logging: Log all automated actions
Credential rotation: Regularly update API keys and passwords
Network segmentation: Isolate automation systems
Change management: Follow approval processes for automation changes

Metrics and Monitoring

SOAR Metrics
IaC Metrics

Track automation effectiveness:

Analyzer success/failure rates
Average enrichment time
Responder execution success
API quota utilization
Time saved through automation
Manual intervention frequency

Official Documentation

Cortex Documentation

Official Cortex documentation and analyzer catalog

Terraform Docs

Complete Terraform documentation and provider registry

PyInfra Documentation

PyInfra operations reference and examples

Cortex Analyzers

Open-source analyzer and responder repository

Next Steps

Review Operations Guide for automation best practices
Explore Incident Handling playbooks for automation opportunities
Check Threat Detection strategies for automated enrichment workflows

Overview

Architecture Components

Deployment

Operations

Security

Automation and SOAR

Automation and SOAR

Architecture Overview

Cortex SOAR

Infrastructure as Code

Cortex SOAR Platform

Core Concepts

Analyzer Configuration

Responder Configuration

Automation Workflows

Infrastructure as Code (IaC)

Terraform Automation

PyInfra Automation

Integration Patterns

Automated Incident Response

Detection

Enrichment

Decision

Response

Continuous Deployment

Best Practices

Metrics and Monitoring

Official Documentation

Cortex Documentation

Terraform Docs

PyInfra Documentation

Cortex Analyzers

Next Steps

Build docs developers (and LLMs) love

Overview

Architecture Components

Deployment

Operations

Security

​Automation and SOAR

​Architecture Overview

Cortex SOAR

Infrastructure as Code

​Cortex SOAR Platform

​Core Concepts

​Analyzer Configuration

​Responder Configuration

​Automation Workflows

​Infrastructure as Code (IaC)

​Terraform Automation

​PyInfra Automation

​Integration Patterns

​Automated Incident Response

Detection

Enrichment

Decision

Response

​Continuous Deployment

​Best Practices

​Metrics and Monitoring

​Official Documentation

Cortex Documentation

Terraform Docs

PyInfra Documentation

Cortex Analyzers

​Next Steps

Build docs developers (and LLMs) love

Automation and SOAR

Architecture Overview

Cortex SOAR Platform

Core Concepts

Analyzer Configuration

Responder Configuration

Automation Workflows

Infrastructure as Code (IaC)

Terraform Automation

PyInfra Automation

Integration Patterns

Automated Incident Response

Continuous Deployment

Best Practices

Metrics and Monitoring

Official Documentation

Next Steps