Security Operations

Security Operations

Security Operations & Incident Response

Security Operations Center (SOC)

SOC Structure

Tier 1: Alert Triage (L1 Analyst)
├─ Monitor SIEM dashboards and security alerts
├─ Perform initial triage and classification
├─ Escalate confirmed incidents to Tier 2
├─ Document all activities in ticketing system
└─ Response Time: 15 minutes for critical alerts

Tier 2: Incident Investigation (L2 Analyst)
├─ Deep investigation of escalated incidents
├─ Contain threats and perform forensic analysis
├─ Coordinate with IT teams for remediation
├─ Escalate to Tier 3 for complex threats
└─ Response Time: 1 hour for critical incidents

Tier 3: Threat Hunting & Advanced Analysis (L3 Analyst)
├─ Proactive threat hunting
├─ Advanced forensics and malware analysis
├─ Create custom detection rules
├─ Research emerging threats
└─ Mentor junior analysts

SOC Manager
├─ Oversee SOC operations
├─ Manage team performance and training
├─ Report metrics to leadership
├─ Coordinate with other departments
└─ Budget and resource planning

SIEM Configuration

Log Sources to Ingest:

Critical Priority (Real-time):
├─ Firewall logs (allow/deny, connections)
├─ IDS/IPS alerts
├─ EDR/antivirus alerts
├─ Authentication logs (successes and failures)
├─ VPN connections
├─ Privileged account activity
└─ Database access logs (for sensitive data)

High Priority (Near real-time, <5 min delay):
├─ Web application logs (access, errors)
├─ Cloud infrastructure logs (AWS CloudTrail, Azure Activity)
├─ Email gateway logs (spam, malware detection)
├─ DLP alerts
└─ File integrity monitoring

Medium Priority (15-30 min delay):
├─ Application logs
├─ DNS query logs
├─ Proxy logs
├─ Network flow data (NetFlow/IPFIX)
└─ Patch management logs

Low Priority (Hourly or daily):
├─ Backup logs
├─ System performance metrics
└─ Non-security application logs

Essential SIEM Use Cases:

# 1. Brute Force Attack Detection
use_case: "Detect brute force login attempts"
data_sources:
  - Windows Security Event Logs (Event ID 4625)
  - Linux auth logs
  - VPN logs
  - Application authentication logs

detection_logic: |
  More than 10 failed login attempts
  FROM same source IP
  TO same user account
  WITHIN 5-minute window

actions:
  - Alert: High severity
  - Block source IP (automatic via firewall integration)
  - Notify account owner
  - Create incident ticket

# 2. Unusual Privileged Account Activity
use_case: "Detect anomalous admin account usage"
data_sources:
  - Active Directory logs
  - Unix/Linux sudo logs
  - PAM session logs

detection_logic: |
  Privileged account login
  OUTSIDE business hours (8 AM - 6 PM)
  OR FROM unusual location
  OR ON unusual system

actions:
  - Alert: Critical severity
  - Require MFA step-up authentication
  - Notify security team and account owner
  - Create incident ticket

# 3. Data Exfiltration Detection
use_case: "Detect large outbound data transfers"
data_sources:
  - Firewall logs
  - Proxy logs
  - DLP alerts
  - Cloud storage logs

detection_logic: |
  Outbound data transfer > 1 GB
  TO external destination
  FROM single user/system
  WITHIN 1-hour window

actions:
  - Alert: High severity
  - Block connection if still active
  - Investigate user/system activity
  - Check for data classification violations

# 4. Malware Detection
use_case: "Detect malware execution"
data_sources:
  - EDR alerts
  - Antivirus logs
  - Process execution logs (Sysmon)
  - Network connections

detection_logic: |
  Malware signature match
  OR suspicious process execution (PowerShell obfuscation)
  OR connection to known C2 IP
  OR file hash matches threat intelligence

actions:
  - Alert: Critical severity
  - Isolate endpoint from network (EDR integration)
  - Kill malicious process
  - Collect forensic artifacts
  - Create incident ticket

# 5. Insider Threat - Abnormal File Access
use_case: "Detect abnormal access to sensitive files"
data_sources:
  - File server audit logs
  - Database query logs
  - SharePoint access logs

detection_logic: |
  User accesses > 100 files
  CONTAINING sensitive data (SSN, credit card, PHI)
  WITHIN 1-hour window
  WHERE user has no recent history of accessing these files

actions:
  - Alert: High severity
  - Notify manager and security team
  - Investigate user activity
  - Check for USB device insertion
  - Review data transfer logs

Threat Hunting

Threat Hunting Process:

1. Hypothesis Generation
   ├─ Based on threat intelligence (e.g., "APT group targeting our industry")
   ├─ Based on recent attacks (e.g., "Check for Log4Shell exploitation attempts")
   └─ Based on anomalies (e.g., "Unusual PowerShell activity in environment")

2. Investigation
   ├─ Query SIEM and EDR for indicators
   ├─ Analyze logs for suspicious patterns
   ├─ Review network traffic
   └─ Examine endpoint artifacts

3. Discovery
   ├─ Confirm presence or absence of threat
   ├─ Document findings
   └─ Assess impact if threat found

4. Response
   ├─ Containment and eradication if threat confirmed
   ├─ Create detection rule for future prevention
   └─ Share findings with team

5. Continuous Improvement
   ├─ Update threat intelligence
   ├─ Refine detection rules
   └─ Document lessons learned

Example Threat Hunt: Living Off the Land (LOLBins)

# Hypothesis: Attackers using native Windows tools for malicious activity

# Hunt 1: Suspicious PowerShell execution
# SIEM query to find encoded PowerShell commands
index=windows EventCode=4688
NewProcessName="*powershell.exe"
CommandLine="*-enc*" OR CommandLine="*-encodedcommand*"
| stats count by Computer, User, CommandLine
| where count > 0

# Hunt 2: Unusual certutil usage (commonly used to download malware)
index=windows EventCode=4688
NewProcessName="*certutil.exe"
CommandLine="*-urlcache*" OR CommandLine="*-decode*"
| stats count by Computer, User, CommandLine

# Hunt 3: WMIC used for lateral movement
index=windows EventCode=4688
NewProcessName="*wmic.exe"
CommandLine="*/node:*"
| stats count by Computer, User, CommandLine

# Hunt 4: BITSAdmin downloading files
index=windows EventCode=4688
NewProcessName="*bitsadmin.exe"
CommandLine="*/transfer*"
| stats count by Computer, User, CommandLine

# Hunt 5: Malicious use of regsvr32
index=windows EventCode=4688
NewProcessName="*regsvr32.exe"
CommandLine="*/s*" AND CommandLine="*http*"
| stats count by Computer, User, CommandLine

---

Incident Response

Incident Response Lifecycle

1. Preparation
   ↓
2. Detection & Analysis
   ↓
3. Containment
   ↓
4. Eradication
   ↓
5. Recovery
   ↓
6. Post-Incident Activity

Incident Response Plan Structure

# Incident Response Plan

## 1. Preparation

### Incident Response Team (CIRT)

| Role | Name | Contact | Responsibilities |
|------|------|---------|-----------------|
| Incident Commander | Jane Doe | +1-555-0101 | Overall coordination |
| Security Lead | John Smith | +1-555-0102 | Technical investigation |
| IT Lead | Bob Johnson | +1-555-0103 | System remediation |
| Communications | Alice Brown | +1-555-0104 | Internal/external comms |
| Legal | Carol White | +1-555-0105 | Legal implications |
| HR | Dave Lee | +1-555-0106 | Insider threats |

### Tools and Resources

- SIEM: Splunk (https://siem.company.com)
- EDR: CrowdStrike (https://falcon.crowdstrike.com)
- Ticketing: Jira Service Desk (security-incidents project)
- Communication: Dedicated Slack channel #incident-response
- Forensic workstation: Located in SOC, admin laptop with encrypted drive
- Incident response jump bag: USB with tools, clean OS images, cables

### Contact Information

- Internal IT Help Desk: x5000
- Security Team On-Call: +1-555-SECURITY
- Legal Department: legal@company.com
- PR/Communications: pr@company.com
- Cyber Insurance: PolicyCo, +1-800-CYBER-INSURE, Policy #12345
- External IR Firm: MandiantFire, +1-888-RESPOND
- Law Enforcement: FBI Cyber Division, Agent Smith, +1-202-555-CYBER

## 2. Detection & Analysis

### Incident Classification

| Severity | Definition | Examples | Response Time |
|----------|-----------|----------|---------------|
| P0 - Critical | Active breach, data exfiltration, ransomware | Active data theft, ransomware encryption, complete system compromise | Immediate (24/7) |
| P1 - High | Confirmed malware, unauthorized access | Malware on critical system, confirmed intrusion | 1 hour |
| P2 - Medium | Suspicious activity requiring investigation | Potential malware, failed intrusion attempt | 4 hours |
| P3 - Low | Policy violation, informational | Minor policy violation, phishing email (no click) | 24 hours |

### Initial Assessment Questions

When receiving an incident report, gather:

1. **What happened?** (Description of the incident)
2. **When did it occur?** (Date and time)
3. **Who discovered it?** (Reporter name and contact)
4. **Which systems are affected?** (Hostnames, IP addresses)
5. **What is the current status?** (Ongoing, contained, resolved)
6. **What data is at risk?** (PII, PHI, financial, IP)
7. **Has law enforcement been notified?** (Yes/No)
8. **What immediate actions were taken?** (System isolated, account disabled)

### Incident Documentation

Create ticket in Jira with:
- Incident ID (auto-generated)
- Classification (P0-P3)
- Timeline of events
- Systems affected
- Actions taken
- Evidence collected
- Next steps

## 3. Containment

### Short-term Containment (Stop the Bleeding)

**Network-Based Containment**:

Isolate compromised system (via firewall)

Block inbound and outbound traffic except to/from SOC analyst workstation

iptables -A INPUT -s <SOC_IP> -j ACCEPT iptables -A OUTPUT -d <SOC_IP> -j ACCEPT iptables -P INPUT DROP iptables -P OUTPUT DROP

Block malicious IP at perimeter firewall

(Use firewall management interface or API)

Disable compromised user account

Active Directory

Disable-ADAccount -Identity compromised_user

Linux

usermod -L compromised_user

**Endpoint-Based Containment**:

Using EDR (CrowdStrike example via API)

curl -X POST "https://api.crowdstrike.com/devices/v2/actions/contain" \ -H "Authorization: Bearer $FALCON_TOKEN" \ -d '{"ids": ["device_id_here"], "action_parameters": []}'

Kill malicious process

Windows

taskkill /F /IM malware.exe /T

Linux

pkill -9 -f malware

**Account-Based Containment**:

Reset password and revoke sessions

Azure AD

Revoke-AzureADUserAllRefreshToken -ObjectId user@company.com

Force password change on next login

Set-ADUser -Identity compromised_user -ChangePasswordAtLogon $true

Disable API keys/tokens

(Application-specific, e.g., AWS IAM)

aws iam delete-access-key --user-name compromised_user --access-key-id AKIA...

### Long-term Containment

- Patch vulnerable systems
- Implement compensating controls (WAF rules, additional monitoring)
- Rebuild compromised systems from clean images
- Update detection rules to prevent recurrence

## 4. Eradication

### Malware Removal

Identify all affected systems

Search EDR for same malware hash or C2 communications

Remove malware from each system

Preferred: Reimage from clean backup

If reimaging not possible:

- Use EDR to quarantine/delete malicious files

- Remove persistence mechanisms (registry, scheduled tasks, services)

- Clear cached credentials

Windows: Remove scheduled task

schtasks /Delete /TN "Malicious Task" /F

Windows: Remove registry persistence

reg delete "HKCU\Software\Microsoft\Windows\CurrentVersion\Run" /v "Malware" /f

Linux: Remove cron job

crontab -e # Remove malicious line

### Vulnerability Remediation

Patch vulnerable systems

Ubuntu/Debian

apt-get update && apt-get upgrade -y

RHEL/CentOS

yum update -y

Windows (via PowerShell)

Install-WindowsUpdate -AcceptAll -AutoReboot

Application-specific (e.g., Log4Shell)

Upgrade Log4j to 2.17.1+

Set JVM option: -Dlog4j2.formatMsgNoLookups=true

### Credential Reset

Force password reset for all potentially compromised accounts

Generate list of users who logged in during breach window

Bulk password reset (Azure AD example)

$users = Get-AzureADUser -All $true | Where-Object {$_.UserPrincipalName -like "*@company.com"} foreach ($user in $users) { Set-AzureADUser -ObjectId $user.ObjectId -PasswordPolicies "DisablePasswordExpiration" -Password (ConvertTo-SecureString "TempPassword123!" -AsPlainText -Force) Set-AzureADUser -ObjectId $user.ObjectId -ForceChangePasswordNextLogin $true }

Rotate service account credentials

Rotate API keys and tokens

Rotate database passwords

Rotate encryption keys (where feasible)

## 5. Recovery

### System Restoration

Restore from clean backup (verified malware-free)

Validate backup integrity

sha256sum backup.tar.gz

Compare to known good hash

Restore data

tar -xzf backup.tar.gz -C /restore/location

Rebuild from golden image (preferred for compromised systems)

Deploy fresh OS from trusted source

Apply all patches

Reinstall applications

Restore data only (not configurations that may contain backdoors)

### Validation Testing

Before returning systems to production:

□ Antivirus/EDR scan shows clean □ No outbound connections to malicious IPs □ File integrity monitoring shows no unexpected changes □ No unauthorized user accounts or scheduled tasks □ All patches applied □ Passwords rotated □ System logs reviewed for anomalies □ Functionality testing completed □ Monitoring/alerting configured

### Return to Normal Operations

- Gradual restoration (not all systems at once)
- Enhanced monitoring for 30 days post-incident
- User communication about password resets
- Document all changes made

## 6. Post-Incident Activity

### Post-Incident Review (PIR)

**Within 5 business days of incident closure, conduct PIR meeting**

Attendees: CIRT members, affected business units, leadership

Agenda:
1. Incident timeline review
2. What went well?
3. What could be improved?
4. Root cause analysis
5. Action items for improvement

### Incident Report Template

Incident Report: [Incident ID]

Executive Summary

Brief non-technical summary of what happened, impact, and resolution.

Incident Details

• Incident ID: INC-2025-001
• Severity: P1 (High)
• Detected: 2025-01-15 14:23 UTC
• Contained: 2025-01-15 16:45 UTC
• Resolved: 2025-01-17 10:00 UTC
• Total Duration: 44 hours

Timeline

Root Cause

Apache web server running vulnerable version of Log4j (2.14.1) was exploited via crafted HTTP User-Agent header. Patch released on 2021-12-10 was not applied due to lack of automated patch management.

Impact Assessment

• Systems Affected: 3 web servers (web-01, web-02, web-03)
• Data Compromised: None confirmed
• Downtime: 12 hours (web services offline during remediation)
• Financial Impact: Estimated $50,000 (12 hours downtime × $4,200/hour)
• Customers Affected: None (no data breach)
• Regulatory: No breach notification required

Actions Taken

1. Isolated affected servers from network 2. Captured forensic images 3. Analyzed malware (CobaltStrike beacon) 4. Identified 3 vulnerable systems 5. Patched Log4j on all servers 6. Rebuilt servers from clean images 7. Deployed WAF rule to block exploit attempts 8. Reset all service account passwords

Lessons Learned

What Went Well

• SIEM alert fired immediately
• CIRT responded within 15 minutes
• Isolation prevented lateral movement
• Backups were recent and intact

What Could Be Improved

• Patch management process too slow (vulnerability was 1 month old)
• No vulnerability scanning for application dependencies
• Incident response playbook for ransomware needed update

Recommendations

1. Implement automated patch management (Priority: High, Owner: IT, Due: 2025-02-15)

• Deploy Patch Manager to automate patching
• SLA: Critical patches within 7 days

2. Add SCA scanning to CI/CD (Priority: High, Owner: AppSec, Due: 2025-02-28)

• Use Snyk or similar to scan for vulnerable dependencies
• Block deployments with critical vulnerabilities

3. Update incident response playbooks (Priority: Medium, Owner: Security, Due: 2025-03-15)

• Add ransomware playbook
• Add supply chain attack playbook
• Conduct tabletop exercise

4. Enhance vulnerability scanning (Priority: Medium, Owner: Security, Due: 2025-03-01)

• Configure Nessus to scan for application vulnerabilities
• Integrate with vulnerability management workflow

Regulatory Reporting

• GDPR Breach Notification: Not required (no personal data compromised)
• State Breach Laws: Not required
• Cyber Insurance: Notified on 2025-01-15, claim filed

Sign-off

• Incident Commander: Jane Doe, 2025-01-18
• CISO: John Smith, 2025-01-18

---

## Security Metrics & KPIs

### SOC Metrics

Detection Metrics:

• name: "Mean Time to Detect (MTTD)"

formula: "Time from incident occurrence to detection" target: "<1 hour for critical incidents" measurement: "Automated via SIEM"

• name: "Alert Volume"

formula: "Total alerts per day/week/month" target: "Trending down (improving signal-to-noise)" measurement: "SIEM dashboard"

• name: "False Positive Rate"

formula: "(False positives / Total alerts) × 100" target: "<20%" measurement: "Track in ticketing system"

Response Metrics:

• name: "Mean Time to Respond (MTTR)"

formula: "Time from detection to initial response" target: "P0: <15 min, P1: <1 hour, P2: <4 hours" measurement: "Ticket timestamps"

• name: "Mean Time to Contain (MTTC)"

formula: "Time from detection to containment" target: "P0: <1 hour, P1: <4 hours" measurement: "Ticket timestamps"

• name: "Mean Time to Resolve (MTTR)"

formula: "Time from detection to incident closure" target: "P0: <72 hours, P1: <7 days" measurement: "Ticket timestamps"

Effectiveness Metrics:

• name: "Incidents by Severity"

formula: "Count of P0/P1/P2/P3 incidents per month" target: "Zero P0, minimal P1 incidents" measurement: "Ticket reports"

• name: "Repeat Incidents"

formula: "(Repeat incidents / Total incidents) × 100" target: "<5%" measurement: "Track root causes in tickets"

• name: "SLA Compliance"

formula: "(Incidents meeting SLA / Total incidents) × 100" target: ">95%" measurement: "Automated via ticketing system"

Team Metrics:

• name: "Analyst Utilization"

formula: "Hours on incidents / Total work hours" target: "60-80% (balance investigation and improvement)" measurement: "Time tracking"

• name: "Training Completion"

formula: "(Analysts completing training / Total analysts) × 100" target: "100% quarterly" measurement: "LMS reporting"

• name: "Threat Hunts Conducted"

formula: "Number of proactive threat hunts per month" target: "Minimum 4 hunts per month" measurement: "Track in project board"

### Sample SOC Dashboard

┌─────────────────────────────────────────────────────────────┐ │ Security Operations Dashboard - January 2025 │ ├─────────────────────────────────────────────────────────────┤ │ Detection & Response Times │ │ ┌─────────────┬─────────────┬─────────────┬─────────────┐ │ │ │ MTTD │ MTTR │ MTTC │ MTT-Resolve │ │ │ │ 42 min │ 18 min │ 2.1 hours │ 15 hours │ │ │ │ ↓ 12% MoM │ ↓ 5% MoM │ → 0% MoM │ ↓ 20% MoM │ │ │ └─────────────┴─────────────┴─────────────┴─────────────┘ │ │ │ │ Incident Volume (January 2025) │ │ ┌─────────────┬─────────────┬─────────────┬─────────────┐ │ │ │ Critical │ High │ Medium │ Low │ │ │ │ 2 (P0) │ 12 (P1) │ 45 (P2) │ 87 (P3) │ │ │ └─────────────┴─────────────┴─────────────┴─────────────┘ │ │ │ │ Alert Metrics │ │ Total Alerts: 15,234 │ False Positives: 2,841 (18.6%) │ │ True Positives: 146 │ Under Investigation: 23 │ │ │ │ Top Attack Vectors │ │ 1. Phishing (62 incidents) │ │ 2. Brute force login (31 incidents) │ │ 3. Malware (18 incidents) │ │ 4. Vulnerability exploitation (12 incidents) │ │ 5. Insider threat (2 incidents) │ │ │ │ Threat Hunting │ │ Hunts Conducted: 5 │ Threats Found: 1 │ New Rules: 3 │ │ │ │ SLA Compliance: 97.3% (Target: >95%) │ └─────────────────────────────────────────────────────────────┘

---

## Tabletop Exercises

### Ransomware Tabletop Exercise

**Objective**: Test incident response procedures for ransomware attack

**Duration**: 90 minutes

**Participants**:
- Incident Commander
- Security team
- IT team
- Legal
- Communications
- Executive sponsor

**Scenario**:

Wednesday, 9:00 AM: The IT help desk receives multiple calls from users reporting that files are encrypted and they see a ransom note demanding $500,000 in Bitcoin.

Initial Investigation:

• 50+ workstations affected across 3 departments
• Ransomware appears to be spreading
• Backup server also appears to be encrypted
• Ransom note gives 48-hour deadline

Your Task: Respond to this incident using your incident response plan.

**Exercise Flow**:

Inject 1 (0:00): Initial notification

• Question: Who do you notify first?
• Question: What immediate containment actions do you take?
• Expected: Activate CIRT, isolate network segments, disable VPN

Inject 2 (0:15): Scope determination

• Update: EDR shows 127 affected systems, ransomware is REvil variant
• Question: How do you determine the full scope?
• Question: Do you pay the ransom?
• Expected: Query EDR for all infected systems, check backups, legal consult

Inject 3 (0:30): Backup assessment

• Update: Backup server encrypted, but offline backups from 48 hours ago exist
• Question: How do you verify backup integrity?
• Question: What is your recovery strategy?
• Expected: Test restore from offline backups, prioritize critical systems

Inject 4 (0:45): External communication

• Update: News outlet calls asking about a data breach
• Question: What do you tell them?
• Question: Do you need to notify regulators?
• Expected: Escalate to communications team, assess breach notification requirements

Inject 5 (1:00): Recovery decisions

• Update: Backups verified clean, recovery will take 3-5 days
• Question: What is your recovery priority order?
• Question: How do you prevent reinfection?
• Expected: Restore critical systems first, implement enhanced monitoring, patch vulnerabilities

Inject 6 (1:15): Post-incident

• Question: What improvements are needed to prevent recurrence?
• Question: What metrics will you track?
• Expected: Better backup strategy, EDR on all systems, security awareness training

**Facilitator Notes**:
- Pause after each inject for discussion
- Ask probing questions to test knowledge
- Document gaps in procedures or knowledge
- Create action items for improvements

**After Action Report**:

Ransomware Tabletop Exercise - After Action Report

Strengths

• Team quickly activated CIRT
• Network isolation performed correctly
• Good understanding of legal/regulatory requirements
• Clear communication during exercise

Areas for Improvement

1. Backup Strategy

• Current: Backups stored online, vulnerable to ransomware
• Recommendation: Implement 3-2-1 backup strategy (3 copies, 2 media types, 1 offsite)
• Owner: IT Manager
• Due: 2025-03-01

2. Incident Response Playbook

• Current: Generic IR plan, lacks ransomware-specific procedures
• Recommendation: Create ransomware playbook with decision trees
• Owner: Security Manager
• Due: 2025-02-15

3. Executive Decision Making

• Current: Unclear who has authority to approve ransom payment
• Recommendation: Define authority matrix in IR plan
• Owner: CISO
• Due: 2025-02-01

4. Communication Templates

• Current: No pre-approved external communication templates
• Recommendation: Create templates for customers, media, regulators
• Owner: Communications Director
• Due: 2025-02-15

Next Steps

• Schedule follow-up tabletop in 6 months
• Conduct technical drill (actual backup restoration test)
• Update incident response plan based on lessons learned

---

## Playbooks

### Phishing Email Response Playbook

Phishing Email Response Playbook

Trigger

User reports suspicious email via "Report Phishing" button or to security@company.com

Severity Classification

• P1 (High): User clicked link or entered credentials
• P2 (Medium): User opened attachment
• P3 (Low): User did not interact with email

Response Steps

Step 1: Triage (Within 15 minutes)

□ Review reported email □ Classify severity (P1/P2/P3) □ Check email gateway logs for delivery count

   # Exchange example
   Get-MessageTrace -SenderAddress "phishing@evil.com" -StartDate (Get-Date).AddHours(-24)

Step 2: Analysis

□ Analyze email headers (sender IP, SPF/DKIM/DMARC results) □ Check URLs in email (use URL sandbox like urlscan.io) □ Analyze attachments (upload to VirusTotal, do not execute) □ Search threat intelligence for IOCs (AlienVault OTX, VirusTotal)

Step 3: Containment

□ Delete email from all mailboxes

   # Office 365 example
   $SearchName = "Phishing Campaign - evil.com"
   New-ComplianceSearch -Name $SearchName -ExchangeLocation All -ContentMatchQuery '(subject:"Invoice 12345") AND (from:phishing@evil.com)'
   Start-ComplianceSearch -Identity $SearchName
   # After search completes:
   New-ComplianceSearchAction -SearchName $SearchName -Purge -PurgeType HardDelete

□ Block sender domain/IP at email gateway □ Block malicious URLs at web proxy/firewall

Step 4: Credential Reset (If P1 - User Clicked/Entered Credentials)

□ Force password reset for affected user

   Set-ADUser -Identity affected_user -ChangePasswordAtLogon $true

□ Revoke active sessions

   Revoke-AzureADUserAllRefreshToken -ObjectId user@company.com

□ Monitor account for suspicious activity (24-48 hours) □ Enable additional logging on account

Step 5: Malware Scan (If P2 - User Opened Attachment)

□ Isolate endpoint (if not already quarantined by EDR) □ Run full antivirus/EDR scan □ Check for IOCs from attachment analysis □ If malware found, follow Malware Incident Playbook

Step 6: User Communication

□ Notify affected users that email was malicious □ Thank users who reported (reinforce positive behavior) □ If credentials reset, provide instructions □ For P3 (no interaction), no additional user action needed

Step 7: Documentation

□ Update incident ticket with:

• Email headers and content
• Number of recipients
• Number who clicked/opened
• IOCs identified
• Actions taken

□ Add IOCs to threat intelligence platform □ Create detection rules for similar emails

Step 8: Prevention

□ Update email gateway rules to block similar emails □ Add sender domain to blacklist □ If widespread campaign, send security awareness notice □ Consider additional user training if many users clicked

This comprehensive security operations and incident response guide provides SOC teams with the structure, processes, and playbooks needed to detect, respond to, and recover from security incidents effectively.


## Usage

Invoke this skill with:

$security-operations [arguments]

Or let Codex auto-select based on your prompt.