Post-Exploitation
This is one of the most important phases of professional penetration testing because getting initial access is not the end of the assessment.
The real questions are:
What can I do with this access? How far can the compromise go? What sensitive assets are reachable? How can I prove the impact safely?
Everything below should be performed only on systems you own or are explicitly authorized to test.
1. What Is Post-Exploitation?
Post-exploitation is the phase after you have obtained authorized access to a system.
Example:
Recon ↓ Scanning ↓ Vulnerability Assessment ↓ Exploitation ↓ Initial Access ↓ POST-EXPLOITATION ↓ Privilege Escalation ↓ Credential Discovery ↓ Lateral Movement ↓ Data/Asset Discovery ↓ Impact Validation ↓ Cleanup ↓ Report
You already learned Privilege Escalation separately.
Post-exploitation is broader.
2. Initial Access ≠ Full Compromise
Suppose you obtain a shell:
www-data@server:~$
You have access.
But you don't yet know:
- Who am I?
- What permissions do I have?
- What operating system is this?
- What network am I connected to?
- What other machines can I reach?
- What credentials are exposed?
- What applications are installed?
- What sensitive data exists?
- Is this system connected to a domain?
- Can this machine provide access to another system?
So the next phase is situational awareness.
3. The Post-Exploitation Mindset
Think:
ACCESS ↓ UNDERSTAND ↓ ENUMERATE ↓ MAP ↓ VALIDATE ↓ ESCALATE ↓ PIVOT ↓ PROVE IMPACT ↓ CLEAN UP
The key principle:
Don't immediately start running random commands.
First understand the environment.
4. Step 1 — Identify Yourself
Linux
whoami
id
groups
Example:
uid=33(www-data) gid=33(www-data) groups=33(www-data)
You now know:
User: www-data Privilege: Low
Windows
whoami
whoami /all
whoami /groups
This can reveal:
- Username
- Domain
- Groups
- Privileges
- Security identifiers
5. Step 2 — Identify the Operating System
Linux
uname -a
cat /etc/os-release
hostname
Windows
systeminfo
hostname
You want to determine:
OS Version Architecture Hostname Patch level Domain membership
6. Step 3 — Understand the Network
This is extremely important.
Linux
ip addr
ip route
ss -tulpn
Windows
ipconfig /all
route print
netstat -ano
You may discover:
Interface ↓ IP address ↓ Subnet ↓ Gateway ↓ DNS ↓ Internal networks ↓ Other reachable systems
7. Example Network Discovery
Imagine you obtain:
IP: 10.10.10.25 Subnet: 10.10.10.0/24 Gateway: 10.10.10.1 DNS: 10.10.10.10
You now know this machine belongs to:
10.10.10.0/24
This can become important during authorized lateral-movement testing.
8. Step 4 — Process Enumeration
Find running processes.
Linux
ps aux
or:
ps -ef
Windows
tasklist
PowerShell:
Get-Process
You are looking for:
- Security software
- Databases
- Web servers
- Backup software
- Custom applications
- Administrative tools
- Scripts
- Services running as privileged users
9. Step 5 — Service Enumeration
Linux
systemctl list-units --type=service
Windows
sc query
PowerShell:
Get-Service
Ask:
Which services are running?
Then:
Which services run with high privileges?
Then:
Can the current user influence them?
This connects directly to Privilege Escalation.
10. Step 6 — Find Installed Software
Debian/Ubuntu
dpkg -l
RPM-based systems
rpm -qa
Windows
Get-ItemProperty HKLM:\Software\Microsoft\Windows\CurrentVersion\Uninstall\*
Why is this useful?
Because applications may have:
- Vulnerable versions
- Weak configurations
- Exposed credentials
- Interesting configuration files
11. Step 7 — Search for Sensitive Information
This is a major post-exploitation activity.
You may encounter:
Passwords API keys Tokens SSH keys Database credentials Configuration files Cloud credentials Application secrets Backup credentials Environment variables
Common Linux locations worth reviewing within scope:
/etc /opt /var/www /home /usr/local
Also examine application configuration files.
For example:
.env config.php settings.py application.yml database.yml
But remember:
Finding a credential does not automatically mean you should use it.
First determine whether the credential is within the authorized scope and whether using it is necessary to demonstrate impact.
12. Environment Variables
Linux:
env
or:
printenv
Windows:
set
PowerShell:
Get-ChildItem Env:
Sometimes applications store secrets in environment variables.
Example:
DATABASE_HOST=... DATABASE_USER=... DATABASE_PASSWORD=... API_KEY=...
This is a common security issue in poorly configured applications.
13. Shell History
Linux shells may store command history.
For example:
history
Depending on the shell:
cat ~/.bash_history
Potentially sensitive information can sometimes appear there.
However, professional testing should avoid unnecessarily exposing or copying secrets.
14. SSH Keys
Linux systems may contain SSH configuration under:
~/.ssh/
Potential files include:
authorized_keys known_hosts config id_rsa id_ed25519
Conceptually:
Private Key ↓ Authentication ↓ Remote System
A discovered private key can potentially represent another access path.
Again:
Don't automatically use discovered credentials or keys. Validate whether doing so is necessary and authorized.
15. Credential Discovery
Credentials can appear in:
Configuration files Environment variables Scripts Logs History Backups Source code Password managers Application databases Memory Credential stores
The professional approach is:
Find ↓ Identify owner ↓ Determine scope ↓ Determine privilege ↓ Assess exposure ↓ Safely validate ↓ Document
16. Credential Reuse
One of the most important concepts.
Suppose you discover:
User: alice Password: ********
You shouldn't immediately assume:
“This gives me access everywhere.”
Instead determine:
Credential ↓ Where is it valid? ↓ Local? ↓ Application? ↓ Database? ↓ SSH? ↓ Domain?
Credential reuse can turn one compromised system into multiple compromised systems.
17. Lateral Movement
Lateral movement means moving from one compromised system/account to another system within the authorized environment.
Example:
Internet ↓ Web Server ↓ Application Server ↓ Internal Server ↓ Domain Controller
Or:
CLIENT01 ↓ SERVER01 ↓ DATABASE01
The goal in a penetration test is to determine whether segmentation and access controls actually prevent this.
18. Common Lateral Movement Paths
Depending on the environment:
Windows
- SMB
- RDP
- WinRM
- Remote services
- Domain credentials
Linux
- SSH
- Application credentials
- Shared keys
- Administrative interfaces
Enterprise environments
- Active Directory
- VPN
- Internal web applications
- Databases
- Management systems
19. Pivoting
Pivoting is different from simple lateral movement.
Imagine:
Kali │ │ Internet-facing access ▼ Compromised Server │ │ Internal network ▼ Internal Network │ ├── Server A ├── Server B └── Database
Your compromised server becomes a pivot point.
Your testing machine may not directly reach the internal systems, but the compromised host can.
20. Pivoting Concept
Without pivot:
Kali ─────────X──────── Internal Server
With pivot:
Kali │ ▼ Compromised Host │ ▼ Internal Server
This is important in real penetration tests because external access does not necessarily expose the entire internal network.
21. Data Discovery
Post-exploitation often involves determining:
What valuable assets could an attacker access?
Examples:
- Customer records
- Source code
- Database information
- Internal documents
- Credentials
- Financial records
- Intellectual property
But a professional pentester should not download huge amounts of sensitive information simply to prove access.
Instead:
Collect the minimum evidence necessary.
For example:
Found: Customer database Proof: Database name + table names + one synthetic/test record Impact: Sensitive customer data is accessible
That's usually better than copying the entire database.
22. Data Staging
Attackers may prepare data for exfiltration.
Conceptually:
Sensitive Data ↓ Collection ↓ Staging Area ↓ Exfiltration
In a penetration test, actual exfiltration should be tightly controlled and explicitly authorized.
Often you can demonstrate the issue using:
- File names
- Metadata
- Small synthetic samples
- Hashes
- Test records
rather than real sensitive data.
23. Persistence
Persistence means maintaining access after the original access mechanism disappears.
Examples conceptually include:
Scheduled tasks Services Startup mechanisms SSH keys Web shells Accounts Registry mechanisms
However, persistence can create significant operational risk.
Therefore professional pentesting generally follows:
Only establish persistence when explicitly authorized.
And if required:
Create ↓ Document ↓ Test ↓ Remove ↓ Verify removal
24. Backdoors vs Professional Testing
A real attacker may attempt to leave hidden access.
A professional pentester should never casually leave:
Backdoor Hidden account Persistent malware Unauthorized scheduled task Unknown service
Instead, persistence testing should be:
- Explicitly approved
- Minimal
- Documented
- Reversible
- Removed after testing
25. Defense Evasion
Attackers may try to avoid detection.
Examples conceptually:
Log manipulation Process hiding Security-tool evasion Timestamp manipulation Obfuscated payloads
For professional testing, this should only be performed when the engagement explicitly includes adversary simulation/red-team objectives.
Normal vulnerability assessment usually doesn't require destructive or stealthy evasion.
26. Command & Control
You may hear:
C2 — Command and Control
Conceptually:
Attacker Infrastructure │ │ C2 channel ▼ Compromised Host │ ▼ Commands / Results
Common legitimate red-team/C2 frameworks exist for adversary simulation.
But the important conceptual understanding is:
A C2 channel allows an operator to manage compromised systems remotely.
For normal pentesting, you don't need to build a sophisticated C2 infrastructure just to prove a vulnerability.
27. Post-Exploitation in Active Directory
This connects directly to the previous topic.
Example:
Compromised Workstation ↓ Identify Domain ↓ Identify Current User ↓ Find Domain Groups ↓ Find Accessible Systems ↓ Identify Credentials/Tokens ↓ Lateral Movement ↓ Privilege Escalation ↓ High-Privilege Account
Tools such as BloodHound can help visualize relationships.
28. Windows Token Concept
Windows uses security tokens to represent:
- User identity
- Groups
- Privileges
- Security context
Conceptually:
User ↓ Security Token ↓ Access Check ↓ Resource
This is why:
whoami /all
is so useful.
It gives insight into the security context of your session.
29. Linux Capabilities
Linux capabilities divide certain privileged operations into more granular permissions.
For example:
getcap -r / 2>/dev/null
You might discover a binary with an interesting capability.
The security question becomes:
Can this capability allow the current user to perform an operation they normally shouldn't?
This connects post-exploitation with privilege escalation.
30. Containers
Modern environments frequently use:
- Docker
- Kubernetes
- Containers
After gaining access to a container, ask:
Am I inside a container? ↓ What privileges does it have? ↓ What mounts exist? ↓ What network access exists? ↓ Are secrets exposed? ↓ Is the container isolated properly?
Useful initial checks:
cat /proc/1/cgroup
ls /
mount
Again, only perform escape testing in an authorized lab.
31. Cloud Post-Exploitation
Cloud environments introduce another dimension.
Potential assets include:
IAM identities Access keys Metadata services Storage buckets Secrets Roles Service accounts Cloud APIs
The attack chain can become:
Compromised Application ↓ Cloud Credential ↓ IAM Permissions ↓ Cloud Resources ↓ Sensitive Data
This is why least privilege is extremely important.
32. Evidence Collection
A professional pentester needs evidence.
Capture:
Timestamp Host Username Finding Command/result Screenshot Relevant logs Affected asset Impact
Example:
Finding: Sensitive API key exposed in application configuration. Host: 10.10.10.20 User: www-data Evidence: Configuration file contains an active test credential. Impact: Application-level credential exposure. Recommendation: Move secret to secure secret management and rotate credential.
33. Cleanup
This is one of the most important professional responsibilities.
After testing:
Remove test accounts Remove temporary files Remove payloads Remove test services Remove persistence Restore configuration Remove temporary keys Close tunnels Stop test processes
Then verify:
Environment ↓ Before testing ↓ During testing ↓ After cleanup ↓ No unintended changes
34. Cleanup Checklist
Before closing the engagement:
- Remove test accounts
- Remove temporary credentials
- Remove test files
- Remove shells
- Remove persistence
- Remove scheduled tasks
- Remove temporary firewall changes
- Close tunnels
- Stop listeners
- Restore configurations
- Verify system health
35. Complete Post-Exploitation Workflow
Your professional workflow should look like:
INITIAL ACCESS │ ▼ ┌─────────────────┐ │ Identify User │ └────────┬────────┘ ▼ ┌─────────────────┐ │ Identify System │ └────────┬────────┘ ▼ ┌─────────────────┐ │ Network Mapping │ └────────┬────────┘ ▼ ┌─────────────────┐ │ Process/Service │ │ Enumeration │ └────────┬────────┘ ▼ ┌─────────────────┐ │ Credential / │ │ Secret Discovery│ └────────┬────────┘ ▼ ┌─────────────────┐ │ Privilege │ │ Escalation │ └────────┬────────┘ ▼ ┌─────────────────┐ │ Lateral Movement│ └────────┬────────┘ ▼ ┌─────────────────┐ │ Data / Asset │ │ Discovery │ └────────┬────────┘ ▼ ┌─────────────────┐ │ Impact Proof │ └────────┬────────┘ ▼ ┌─────────────────┐ │ Cleanup │ └────────┬────────┘ ▼ REPORT
36. Practical Lab 🔬
Build a small lab:
Kali │ ▼ WEB-SERVER 10.10.10.20 │ ┌───────┴────────┐ │ │ DATABASE INTERNAL SERVER SERVER
Start with an intentionally vulnerable application.
Your exercise:
Phase 1 — Initial access
Obtain authorized lab access.
Phase 2 — Identify
whoami id hostname uname -a
Phase 3 — Network
ip addr ip route ss -tulpn
Phase 4 — Processes
ps aux
Phase 5 — Services
systemctl list-units --type=service
Phase 6 — Secrets
Inspect the application's test configuration.
Phase 7 — Privilege
Look for an authorized privilege-escalation path.
Phase 8 — Internal access
Determine whether the compromised server can reach another lab system.
Phase 9 — Impact
Find a controlled test asset.
Phase 10 — Cleanup
Remove everything you changed.
37. Example Attack Chain
Imagine your lab produces:
Public Web Application ↓ Web Vulnerability ↓ www-data ↓ Application Configuration ↓ Database Credential ↓ Database Access ↓ Sensitive Test Data ↓ Privilege Escalation ↓ Root
That is much more valuable than simply saying:
“I found SQL injection.”
You demonstrated an attack chain and business impact.
38. Post-Exploitation vs Privilege Escalation
Don't confuse them.
| Privilege Escalation | Post-Exploitation |
|---|---|
| Increase privileges | Understand and leverage access |
| Low → high privilege | Broad phase after access |
| SUID, sudo, services | Enumeration, credentials, pivoting, impact |
| Linux/Windows | Linux/Windows/AD/cloud |
| One part of attack | Larger lifecycle |
Privilege escalation is often inside post-exploitation.
39. Professional Mindset 🧠
When you get a shell, don't think:
“I got shell!”
Think:
Who am I? ↓ What can I access? ↓ What can I modify? ↓ What networks can I reach? ↓ What secrets exist? ↓ What identities exist? ↓ What other systems are reachable? ↓ What is the highest-impact path? ↓ What is the minimum evidence needed? ↓ How do I clean up?