Digbit – a Bit-shifted automatic domain generation for BitSquatting [python]

BitSquatting is not new but it’s relatively new. This type of attack relies on a random bit that switches from 0 to 1 or from 1 to 0 inside a RAM. Because of this even though you try to access domain like cnn.com, in fact your computer is requesting ann.com. It’s rare but it happens. It can be caused by a cosmic radiation or overheated memory modules. If You would like to learn more I can recommend Artem’s page.

To make it easier to find a domain that is in a single bit-shift distance (as in Hamming distance) I’ve created a script that is generating all the possibilities.

For example lets search for domains (bit-wise) close to cnn.com. The script output will be:

snn.com knn.com gnn.com ann.com bnn.com c.n.com cfn.com cjn.com cln.com con.com cnf.com cnj.com cnl.com cno.com cnn.som cnn.kom cnn.gom cnn.aom cnn.bom cnn.cgm cnn.ckm cnn.cmm cnn.cnm cnn.coe cnn.coi cnn.coo cnn.col

To make it easier to check if particular domain is already registered or not, I’ve made a wrapper script that will execute the python script and for each generated domain it will execute command:

> nslookup domain | grep "NXDOMAIN"

The wrapper script is executed with a single argument that is a domain name. Sample output for twitter.com:

Twitter bitsquatting

Some of those reported as available are obviously false-positive since TLDs like kom don’t really exist. I did not removed them because new TLDs are added from time to time and you might as well have a custom domain setup within your LAN.

The wrapper code:


for i in $( ./digbit.py $1);
        dotcheck=$( echo "$i" | grep "\." | wc -l)
        echo -n "$i ";
        check=$(nslookup $i| grep "NXDOMAIN" | wc -l);
        if [[ $check -ne 0 && dotcheck -ne 0 ]];
                echo " < available";
                echo " ";

The Digbit script code:

#!/usr/bin/env python3
import binascii
import re
import os
import sys

def new_text2bin(mtext):
        result = [ bin(ord(ch))[2:].zfill(8) for ch in mtext ]
        return ''.join(result)

def switch_bit(bit, my_list ):
        result_list = my_list[:]
        if result_list[bit] == '1':
                result_list[bit] = '0'
                result_list[bit] = '1'
        return result_list

def generate_similar( domain ):
        domains = []
        domain_list = list(domain)
        for i in range(len(domain)):
                print(switch_bit(i, domain_list))
        return domains

def binstr_to_ascii( binstr ):
        binstr = -len(binstr) % 8 * '0' + binstr
        string_blocks = (binstr[i:i+8] for i in range(0, len(binstr), 8))
        string = ''.join(chr(int(char, 2)) for char in string_blocks)
        return string

if len(sys.argv) != 2:
        sys.exit("Not enough args")

        domain_list = list(new_text2bin(domain))

        for i in range(len(domain_list)):
                new_d = (''.join(switch_bit(i, domain_list)))

                new_d_str = binstr_to_ascii(new_d)
                corect_domain = re.match("^(((([a-z0-9]+){1,63}\.)|(([a-z0-9]+(\-)+[a-z0-9]+){1,63}\.))+){1,255}$", new_d_str + "." )

                if ( corect_domain is not None and len(corect_domain.string) > 1 and (corect_domain.string).index(".") < len(corect_domain.string)-1 ):
                        print(new_d_str, end="\n")

        print(" ")



Mèdved – web-based DNS zone transfer automation

It’s been a while since my last post, so today i have something bigger and – most probably – more usefull than usually. [Download link here]

I present to you Mèdved (bear in serbian language). It is a part of the suite of tools i’m creating thus the main directory is named carnivores.
This is a web-based tool designed to automate the search for domain transfers. It has an intuitive interface and few helpful shortcuts. As an input it expects a domain or list of domains. Ideally the list should be comma-separated but it will handle space or CR-LF separated lists as well. Aside from normal results it gives you a log of performed searches and all successful transfers are archived.


  • Linux + Apache2
  • path to medved.php: /var/www/carnivores/medved/medved.php (simply extract the archive to /var/www/)

There are some requirements for directory structure and permissions so i’ll show you how the tree should look like:

directory tree required

Below is the first page with help toggled:

dns zone transfer Medved tool

It has been implemented with responsive design so you can use it on your smartphone/tablet although the interface becomes slightly denser:

dns zone automation Medved responsive/mobile

You can supply the list as

domain.com, domain2.com, domain3.com.

If you have a list looking like this:

domain.com domain2.com domain3.com

or like this:


you can paste it as well, just use the button Spaces to commas, before clicking Analyze button, and the list will be corrected to expected form. If you have an URL list instead of domains – use the button Sanitize URLs and it should strip all the unnecessary stuff from the URL.

This and more about the available functions and shortcuts is described in the help.

Lets see how it works with example:

dns zone automation tool in action

As you can see the transfer for Microsoft is actively refused. All tested NS servers have separate tabs. The warning sign shown for other domain (which i removed from the picture) indicates that server do not accept TCP connections. The OK sign for one of the servers indicates successful transfer.

It is common to frequently test the same domain after some time, to see if new records have been added or if the server configuration has been corrected. That is why all the successful transfers are saved in the archive. An archive is a simple list of available transfer results.

dns zone transfer archive

You can filter the results to show the particular domain only, by clicking on a domain name. The list shows the date of the transfer, records discovered and the link to review the transfer data. If more than one server responded with transfer data for particular domain, the number of records shown will be the sum from all the servers.
This might give you a false idea that you might get 1000 records in the transfer when in fact you received 500 records but from two servers.

If you need unique records just save the file and use the command

cat records.txt | sort -u

I’m not going to post any code in here as it would be really tedious work. Instead you can download all of it using the link below.

Download from here or medved [change to 7z].

As usual i’m not responsible for how you use this tool. This is presented only as a proof of concept. You can use it but you cannot distribute it without my knowledge and explicit consent.

I’ve used a code snippet from stevenlevithan.com for URL parsing and shortcut.js file from openjs.com for creating keyboard shortcuts.


DNS zone transfer script

Notice. I’ve made a web-based version of this script that has more functions and an archive for successful transfers. 

Script automating discovery of name servers allowing zone transfers.

Nothing fancy. Just to make it easier.

The output:

Zone transfer discoveredIf you use the command presented on the bottom of the image above you will get results like this:

Successful zone transfer for example domain




for dnsserver in $(host -t ns "$domains" | cut -d " " -f 4);
        # VARIABLES
        str_len=$(echo "$dnsserver" | tr -d " " | wc -c)
        str_len=$(echo "$str_len-2"| bc )
        dns_server=$(echo "$dnsserver" | cut -b "-$str_len")
        zone_data=$(dig axfr "$1" "@$dns_server")

        check=$(echo "$zone_data" | grep "Transfer failed" | wc -l)

        if [[ $check -ne 0 ]];
                echo -e " Transfer \033[31mFAILURE\033[00m at $dns_server"
                echo -e " Transfer \033[32mSUCCESS\033[00m at $dns_server"



echo ""
echo " Use command: dig axfr $1 @$server"

# echo "$data"