Skip to content

A set of prioritized Hashcat .hcmask files intelligently developed from terabytes of password breach datasets and organized by run time.

Notifications You must be signed in to change notification settings

itdojo/Extreme_Breach_Masks

 
 

Repository files navigation

Extreme_Breach_Masks

A set of prioritized Hashcat masks intelligently developed from terabytes of password breach datasets and organized by run time.

Goal

To improve the efficiency of password cracking using Hashcat mask attacks by prioritizing masks with the highest password cracking probability in the shortest possible time using high volumes of password breach data.

Background

Inspired by the work of golem445 who compiled a set of password hashcat password masks using real-world data. I took this a step further by building a set of prioritized Hashcat masks using an enormous password breach dataset that I have been personally compiling and curating.

Methodology

  1. Compiled every available password breach dataset that I could find -- terabytes of data! Wordlists include everything readily google-able and torrent-able. Noteable inclusions are: crackstation.net, seclists, rockyou, COMB, breach-parse... and many, many more.
  2. Combined the wordlists in a way that they were generally sorted by password usage commonality.
  3. Deduplicated the wordlist without re-sorting (important to retain the commonality order) using this tool: https://github.com/nil0x42/duplicut
  4. Ran the wordlist through the statsgen.py tool to convert the wordlist into a counted set of password masks: https://github.com/iphelix/pack
python statsgen.py breach_wordlist.txt -o masks.statsgen
  1. Ran the resulting statsgen.py output through maskgen.py to generate .hcmask files that are efficently ordered and seperated by run time. The run time duration assumes a hashing speed of 56,636,300,000 keys per second. This was determiend based on the performance of 1x Nvidia GTX1080Ti cracking NTLM hashes in Hashcat. Example command below:
python maskgen.py --optindex -o ./1-hour_8.hcmask --minlength=8 --maxlength=8 --pps 56636300000 --targettime 3600 masks.statsgen
  1. Repeated step #5 with various execution times to generate files optimized for various run times.

Usage

The .hcmask files above describe passwords of differing character lengths, each sorted by efficiency, and formatted for use by the Hashcat password cracking tool. Depending on your situation, you might want to focus on passwords of a specific length only vs the entire set. You should select the hcmask file optimized for your desired time frame. The statsgen file is included if you want to re-sort and generate your own hcmask files; however, I had to pair it down to only 8-14 characters and 7zip it because the full version was too large for github. Recognize that this type of brute force mask attack can take a long time and should be performed last after you have exhausted more targeted methods. My recommended password cracking attack order is below:

  1. Backup/Clear your hashcat potfile for the new set of hashes because mixing previous results gets confusing and if the potfile gets too large it can slow down the cracking process.
  2. Basic dictionary attack with your favorite wordlist... ie rockyou.txt
hashcat.exe -d <include_gpu_numbers> -m 1000 -w 4 -a 0 --session <name_your_session> <ntlm_hashes.txt> <rockyou.txt> -O
  1. Brute force all permutations 1-7 character length passwords... this does not take long given the minimal keyspace of this group.
hashcat.exe --increment --increment-min=1 -d <include_gpu_numbers> -m 1000 -w 4 -a 3 --session <name_your_session> <ntlm_hashes.txt> ?a?a?a?a?a?a?a -O
  1. Targeted dictionary attack... create a custom lowercase wordlist using CeWL and add local sports teams, city names, mascots, etc and apply the best64.rule
hashcat.exe -d <include_gpu_numbers> -m 1000 -w 4 -a 0 --session <name_your_session> <ntlm_hashes.txt> <custom_wordlist.txt> -r best64.rule -O
  1. Analyze the set of cracked passwords for potential patterns, run targeted attacks which reflect those patterns.
  2. BIG dictionary attack... run the passwords through the largest wordlist you have.
  3. Analyze any newly cracked passwords for potential patterns, run targeted attacks which reflect those patterns.
  4. Use this repository of work and run the "duration_characters.hcmask" according to your needs.

Example Hashcat Command for Using the .hcmask to Crack NTLM Hashes

hashcat.exe -d <include_gpu_numbers> -m 1000 -w 4 -a 3 --session <name_your_session> <ntlm_hashes.txt> 1-day_8-14.hcmask -O

Extra flags that might be necessary to make Hashcat work on your system. NOTE: These are dangerous, and should only be used very rarely!

--force | Force Hashcat to stop bothering you and just crack the damn passwords.
--hwmon-disable | Hashcat will throttle based on temp readings... ignore this and go for broke!

About

A set of prioritized Hashcat .hcmask files intelligently developed from terabytes of password breach datasets and organized by run time.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published