cracke-dit ("Cracked It") makes it easier to perform regular password audits against Active Directory environments.
Ensuring your users have strong passwords throughout the organisation is still your best line of defence against common attacks. Many organisations over estimate just how secure their users' passwords are. "London123", "Winter2017", "Passw0rd" - all complex passwords, according to the default Group Policy rules, and probably your users.
By performing regular audits, you can identify users with weak passwords and take action inline with your policies and procedures.
Python 2.7+ and pip are required. Then just:
git clone https://github.com/eth0izzle/cracke-dit.git
pip install virtualenv && virtualenv .virtualenv && source .virtualenv/bin/activate
pip install -r requirements.txt
python cracke-dit.py --help(and see Usage)
The first step in your password cracking adventure is to extract a copy of the Active Directory database, ntds.dit, which contains the password hashes. Depending on your persuasion you have a few options - choose one:
If you are on the same network as a domain controller you can extract the hashes directly.
python cracke-dit.py --username administrator --password passw0rd --target 192.168.1.1(where 192.168.1.1 is a DC).
ntdsutil "ac i ntds" "ifm" "create full c:\temp" q q.
c:\temp\registry\SYSTEMto your system with cracke-dit.
(not required if you followed Step 1a and remotely extracted with cracke-dit)
All password hashes are protected with 3 layers of encryption. Thankfully everything we need to decrypt is within the SYSTEM hive. The next step is to extract the hashes and usernames in to a separate file for cracking:
python cracke-dit.py --system SYSTEM --ntds ntds.dit(optionally with
--no-historyflag if you don't care about historic passwords)
cracke-dit doesn't actually crack passwords, you will need to use your favourite password cracker for that. cracke-dit just needs a
.pot file (hash:password) for processing. I'm partial to hashcat so:
hashcat -m 1000 --potfile-path <domain>.pot --username <domain>.hashes.ntlm /usr/share/Wordlists/rockyou.txtwhich will be pretty quick.
hashcat -m 1000 --potfile-path <domain>.pot --username <domain>.hashes.ntlm /usr/share/Wordlists/rockyou.txt -r hob064.rule
Now we have cracked a bunch of hashes, let's load them in to cracke-dit!
python cracke-dit.py --pot <domain>.pot --domain <domain>(where domain is FQDN, i.e. acme.local). Optionally pass in
--only-enabled- hopefully they are self explanatory.
Using the ntds.dit and SYSTEM in
./samples we get the following output:
Users highlighted in green are enabled, red are disabled, and gray is an historic password.
Password scores are based on Dropbox's zxcvbn:
|0||Too guessable: risky password.||< 10^3|
|1||Very guessable: protection from throttled online attacks.||< 10^6|
|2||Somewhat guessable: protection from unthrottled online attacks.||< 10^8|
|3||Safely unguessable: moderate protection from offline slow-hash scenario.||< 10^10|
|4||Very unguessable: strong protection from offline slow-hash scenario.||=> 10^10|
By default passwords are checked against Have I Been Pwned to see if they appear in breach lists. Passwords are never transmitted as we use the k-Anonymity model.
Results can be processed by different output modules via the
The default output module and shown in the demo above. Shows interesting stats, top 10 passwords by reuse, the top 5 worst passwords and if any passwords use month or day names.
E-mails the top 25 passwords (by reuse).
Spits out a wordcloud of all passwords, colored by password score.
Introduce internal training on what a secure password is, why they are important and embed it in to your induction programme.
Consider rolling out a password manager and adequate training for all of your users. Stronger, longer and more unique passwords is better for everyone.
Gradually increase your password minimum length requirement to 12 characters. Generally longer passwords are more secure.
Phase out forcing your users to "reset password every X days". There is research to suggest that this doesn't help create strong passwords, but in fact has the opposite effect. Don't forget the actual cost of this.
Carry out a password audit quarterly. You can very easily cron the steps above and run with the
Do not name and shame people with poor password hygiene - educate them! Get HR buy-in and introduce a "3 strike system" that will carry a formal warning.
Check out the issue tracker and see what takes your fancy.
git checkout -b my-new-feature
git commit -am 'Add some feature'
git push origin my-new-feature
Huge thanks CoreSecurity's Impacket!