Special thanks to DavidAngelos:
- In-order joins (
-i/--inorder): join keywords only in the original order (e.g.foo,bar,baz→foo, bar, baz, foobar, foobaz, barbaz, foobarbaz). - All-order combinations (
-c/--combinations): generate every ordering of each subset (e.g.foo,bar,baz→foo, bar, baz, foobar, foobaz, barfoo, …, bazbarfoo). - Custom separator (
--sep <string>): when joining words, insert this string between tokens (defaults to no separator). - Max combine size (
--max-combine <N>): limit how many raw keywords get joined together (default: 2). - Min/Max length filtering of final words (
--minlen/--maxlen <N>): filter the final wordlist only with word with the desired length.
Psudohash is a password list generator for orchestrating brute force attacks and cracking hashes. It imitates certain password creation patterns commonly used by humans, like substituting a word's letters with symbols or numbers (leet), using char-case variations, adding a common padding before or after the main passphrase and more. It is keyword-based and highly customizable. 🎥 -> Video Presentation
System administrators and other employees often use a mutated version of the Company's name to set passwords (e.g. Am@z0n_2022). This is commonly the case for network devices (Wi-Fi access points, switches, routers, etc), application or even domain accounts. With the most basic options, psudohash can generate a wordlist with all possible mutations of one or multiple keywords, based on common character substitution patterns (customizable), case variations, strings commonly used as padding and more. Take a look at the following example:
The script implements the following character substitution schema. You can add/modify character substitution mappings by editing the transformations list in psudohash.py and following the data structure presented below (default):
transformations = [
{'a' : ['@', '4']},
{'b' : '8'},
{'e' : '3'},
{'g' : ['9', '6']},
{'i' : ['1', '!']},
{'o' : '0'},
{'s' : ['$', '5']},
{'t' : '7'}
]
When setting passwords, I believe it's pretty standard to add a sequence of characters before and/or after the main passphrase to make it "stronger". For example, one may set a password "dragon" and add a value like "!!!" or "!@#" at the end, resulting in "dragon!!!", "dragon!@#", etc. Psudohash reads such values from common_padding_values.txt and uses them to mutate the provided keywords by appending them before (-cpb) or after (-cpa) each generated keyword variation. You can modify it as you see fit.
When appending a year value to a mutated keyword, psudohash will do so by utilizing various seperators. by default, it will use the following seperators which you can modify by editing the year_seperators list:
year_seperators = ['', '_', '-', '@']
For example, if the given keyword is "amazon" and option -y 2023 was used, the output will include "amazon2023", "amazon_2023", "amazon-2023", "amazon@2023", "amazon23", "amazon_23", "amazon-23", "amazon@23".
Install Python 3.x and tqdm first:
pip3 install tqdmThen clone the repo and make the script executable:
git clone https://github.com/t3l3machus/psudohash.git
cd ./psudohash
chmod +x psudohash.py
./psudohash.py [-h] -w WORDS [-i] [-c] [--sep SEP] [--max-combine N] [-an LEVEL] [-nl LIMIT] [-y YEARS] [-ap VALUES] [-cpb] [-cpa] [-cpo] [-o FILENAME] [-q]
The help dialog [ -h, --help ] includes usage details and examples.
-
-w, --words <kw1,kw2,…>
Comma‐separated raw keywords (required). -
-i, --inorder
Join up to--max-combinekeywords in the given order (e.g.foo,bar,baz→foo, bar, baz, foobar, foobaz, barbaz, foobarbaz). -
-c, --combinations
Generate every permutation of each subset (up to--max-combine) (e.g.foo,bar,baz→foo, bar, baz, foobar, foobaz, barfoo, …). -
--max-combine <N>(default: 2)
Maximum number of raw keywords to join into one base string. -
--sep <string>
When joining words (-ior-c), place this string between tokens. Defaults to an empty string. -
--minlen <N>
Discard any final password shorter than N characters. -
--maxlen <N>
Discard any final password longer than N characters. -
-an, --append-numbering <LEVEL>
Append numbered suffixes (zero‐padded to<LEVEL>digits) to each word mutation. -
-nl, --numbering-limit <LIMIT>
Maximum number to count up to when appending numbers (default: 50). -
-y, --years <years>
Append one or more years to each mutation (e.g.1990-2000, or2022,2023). -
-ap, --append-padding <vals>
Append custom padding values (comma‐separated). Must be used with-cpbor-cpa. -
-cpb, --common-paddings-before
Prepend values fromcommon_padding_values.txtbefore each mutation. -
-cpa, --common-paddings-after
Append values fromcommon_padding_values.txtafter each mutation. -
-cpo, --custom-paddings-only
Use only user‐provided paddings (no defaults). Must be used with-ap. -
-o, --output <file>
Write the results to<file>(default:output.txt). -
-q, --quiet
Suppress the ASCII art banner on startup.
-
No multi‐word (singletons only)
./psudohash.py -w foo,bar,baz -cpa # → foo, bar, baz -
In‐order joins (-i, up to 2 words by default)
./psudohash.py -w foo,bar,baz -i # → foo, bar, baz, foobar, foobaz, barbaz -
All‐order combinations (-c, up to 2 words by default)
./psudohash.py -w foo,bar,baz -c # → foo, bar, baz, foobar, foobaz, barfoo, barbaz, bazfoo, bazbar -
Change separator between joined words
./psudohash.py -w foo,bar,baz -i --sep "_" # → foo, bar, baz, foo_bar, foo_baz, bar_baz
-
Length Filtering (
--minlen/--maxlen)./psudohash.py -w apple,banana -i --minlen 10 # Warning: exact size cannot be determined because of length filters. # Example final outputs might include “applebanana” (11 chars), “bananaapple” (11 chars).
-
Combine up to 3 words (instead of default 2)
- Combining options
--yearsand--append-numberingwith a--numbering-limit≥ last two digits of any year input, will most likely produce duplicate words because of the mutation patterns implemented by the tool. - If you add custom padding values and/or modify the predefined common padding values in the source code, in combination with multiple optional parameters, there is a small chance of duplicate words occurring. psudohash includes word filtering controls but for speed's sake, those are limited.
- When using
--minlenor--maxlen, the script cannot pre-calculate the exact word count; you’ll see a “exact size cannot be determined” warning and the size without this filter will be calculated, the final size will be smaller.
When it comes to people, i think we all have (more or less) set passwords using a mutation of one or more words that mean something to us e.g., our name or wife/kid/pet/band names, sticking the year we were born at the end or maybe a super secure padding like "!@#". Well, guess what?
I'm gathering information regarding commonly used password creation patterns to enhance the tool's capabilities.