Tools for encryption, decryption and cracking from several cryptographic systems.
🇪🇸 Obtén información sobre la criptografía clásica y este proyecto en este artículo.
- Install Python 3.6 if you do not have it yet.
- Ensure that you have installed python3.6 in /usr/bin otherwise you will need to make a symbolic link to its real location:
sudo ln -s $(which python3.6) /usr/bin/python3.6
- Install the following Python dependencies:
pip3 install numpy
pip3 install pyenchant
- Clone this repository:
git clone https://github.com/Carleslc/CryptTools.git - Give execution permissions to all tools:
cd CryptTools/tools
chmod +x *
These tools are designed to be useful in the field of computer security and their use is restricted to personal use or under consent. I am not responsible for any illicit use that may occur.
--> What is the Caesar Cipher? <--
./caesar.py --help
usage: caesar.py [-h] [-t TEXT] [-k KEY] [-l LANG] [-V] [-A] [-D]
[-T THRESHOLD] [--beep]
optional arguments:
-h, --help show this help message and exit
-t TEXT, --text TEXT text to read from. If not specified the program will
read from standard input
-k KEY, --key KEY key used to encrypt. If no key is provided the program
will try to crack and decrypt using the specified
language
-l LANG, --lang LANG available languages: ['de_DE', 'en_AU', 'en_GB',
'en_US', 'fr_FR'] (default: en_US). Only useful if no
key is provided
-V, --verbose show extra information
-A, --all show decrypted text for each tested key
-D, --debug show information about text validation
-T THRESHOLD, --threshold THRESHOLD
valid word count percentage to mark the whole text as
valid language (default: 50)
--beep plays a beep sound when program finishes. May require
SOX to be installed
./caesar.py -t "This is the Caesar tool from CryptTools!" -k 5 > test
Ymnx nx ymj Hfjxfw yttq kwtr HwduyYttqx!
-t argument is not mandatory, so if you need to encrypt a long text you can skip it, execute ./caesar.py -k 5 > test and then paste your text. When completed press Return and then finish the input with Ctrl+D so the program will read it.
./caesar.py -t "This is the Caesar tool from CryptTools!" -k 5 -VA
Original text most frequent character: t
Ymnx nx ymj Hfjxfw yttq kwtr HwduyYttqx!
Encrypted text most frequent character: y
To decrypt you will need to know the shift needed to reallocate each character to its correct character in the alphabet. In the example above the encrypted shift is 5 so the shift needed to decrypt is 26 - 5 = 21, where 26 is the size of the alphabet.
./caesar.py -k 21 < test
This is the Caesar tool from CryptTools!
This method cracks the message with bruteforce and then checks every result validating the language to guess which result is the original text. By default language is English, for other languages read below.
./caesar.py < test
This is the Caesar tool from CryptTools!
Read from a file and print cracked keys:
./caesar.py -V < test
Most frequent character: y
Testing 't' (ROT-21) SUCCESS
Decrypted with ROT-21. Original encryption key: 5
This is the Caesar tool from CryptTools!
ROT-X means that each character in the text is shifted X positions in the alphabet.
To decrypt a message without knowing the key you need to know the original text language. It is provided with --lang option. By default it is en_US (American English).
For example, in Deutsch language:
./caesar.py -t "Dies ist das Scytale-Tool von CryptTools!" -k 6
Joky oyz jgy Yiezgrk-Zuur but IxevzZuury!
If language is not provided it will try English and it will fail:
./caesar.py -V -t "Joky oyz jgy Yiezgrk-Zuur but IxevzZuury!"
Most frequent character: y
Testing 'z' (ROT-1)
Sorry. None of decrypted results seems to be written in language en_US...
Then, providing the correct language:
./caesar.py -V -t "Joky oyz jgy Yiezgrk-Zuur but IxevzZuury!" --lang "de_DE"
Most frequent character: y
Testing 's' (ROT-20) SUCCESS
Decrypted with ROT-20. Original encryption key: 6
Dies ist das Scytale-Tool von CryptTools!
You can check all available codes with ./caesar.py --help. You can even install more languages, for that take a look here.
If original text language is unknown you still can generate all possible transformations, then you will need to check them manually in order to know which is the correct:
./caesar.py -V -A -t "Joky oyz jgy Yiezgrk-Zuur but IxevzZuury!"
Most frequent character: y
Testing 'e' (ROT-6) Testing decrypted text:
"Puqe uef pme Eokfmxq-Faax haz OdkbfFaaxe!"
Testing 't' (ROT-21) Testing decrypted text:
"Ejft jtu ebt Tdzubmf-Uppm wpo DszquUppmt!"
...
Testing 's' (ROT-20) Testing decrypted text:
"Dies ist das Scytale-Tool von CryptTools!"
Testing 'h' (ROT-9) Testing decrypted text:
"Sxth xhi sph Hrnipat-Idda kdc RgneiIddah!"
...
Testing 'q' (ROT-18) Testing decrypted text:
"Bgcq gqr byq Qawryjc-Rmmj tml ApwnrRmmjq!"
Testing 'z' (ROT-1) Testing decrypted text:
"Kplz pza khz Zjfahsl-Avvs cvu JyfwaAvvsz!"
Sorry. None of decrypted results seems to be written in language en_US...
Testing order is frequency order (most common letter is tested first).
In addition, with the extra option -D you can check the language validation process:
./caesar.py -VAD < test
Most frequent character: y
Testing 'e' (ROT-6) Testing decrypted text:
"Estd td esp Nlpdlc ezzw qczx NcjaeEzzwd!"
estd: False | Progress 0% | Max 88%
td: False | Progress 0% | Max 75%
esp: True | Progress 12% | Max 75%
nlpdlc: False | Progress 12% | Max 62%
ezzw: False | Progress 12% | Max 50%
qczx: False | Progress 12% | Max 38%
Testing 't' (ROT-21) Testing decrypted text:
"This is the Caesar tool from CryptTools!"
this: True | Progress 12% | Max 100%
is: True | Progress 25% | Max 100%
the: True | Progress 38% | Max 100%
caesar: False | Progress 38% | Max 88%
tool: True | Progress 50% | SUCCESS
Decrypted with ROT-21. Original encryption key: 5
This is the Caesar tool from CryptTools!
-A and -D options may be too verbose, avoid using them for long texts.
You can also set the permissiveness of the language validation process with the --threshold -T option. By default it is set to 50 (half of the text words must be written in the specified language in order to accept it as the original text). Values must be between 1 and 100. Values below 20% are not recommended (an encrypted text may be accepted as decrypted). Higher values indicate toughness, but 100% it is neither recommended (in the text may be non-english nouns and other original but non-english words).
As you can see, it is too easy to crack this classical cryptographic system so it is not recommended to use it in production software.
./scytale.py --help
usage: scytale.py [-h] [-t TEXT] [-k KEY] [-l LANG] [-V] [-A] [-D]
[-T THRESHOLD] [--beep]
optional arguments:
-h, --help show this help message and exit
-t TEXT, --text TEXT text to read from. If not specified the program will
read from standard input
-k KEY, --key KEY key used to encrypt. If no key is provided the program
will try to crack and decrypt using the specified
language
-l LANG, --lang LANG available languages: ['de_DE', 'en_AU', 'en_GB',
'en_US', 'fr_FR'] (default: en_US). Only useful if no
key is provided
-V, --verbose show extra information
-A, --all show decrypted text for each tested key
-D, --debug show information about text validation
-T THRESHOLD, --threshold THRESHOLD
valid word count percentage to mark the whole text as
valid language (default: 50)
--beep plays a beep sound when program finishes. May require
SOX to be installed
./scytale.py -t "This is the Scytale tool from CryptTools!" -k 5 > test
Theoohe moi t lsSoCs cor!iyly st p aft tlrT
-t argument is not mandatory, so if you need to encrypt a long text you can skip it, execute ./scytale.py -k 5 > test and then paste your text. When completed press Return and then finish the input with Ctrl+D so the program will read it.
./scytale.py -t "This is the Scytale tool from CryptTools!" -k 5 -VA
Testing matrix: 5x9
Text to cipher: "This is the Scytale tool from CryptTools!" (41)
[['T' 'h' 'i' 's' ' ' 'i' 's' ' ' 't']
['h' 'e' ' ' 'S' 'c' 'y' 't' 'a' 'l']
['e' ' ' 't' 'o' 'o' 'l' ' ' 'f' 'r']
['o' 'm' ' ' 'C' 'r' 'y' 'p' 't' 'T']
['o' 'o' 'l' 's' '!' ' ' ' ' ' ' ' ']]
Result size: 44
Theoohe moi t lsSoCs cor!iyly st p aft tlrT
That matrix represents the Scytale. To encrypt the text it is written by rows (5) and read by columns (9).
To decrypt you will need to know the columns of the underlying matrix. In the example above they are 9.
If you only have the key 5 (rows) then you can calculate the columns with the text size:
Columns = Size/Rows. For example, the encrypted text above has size 44, so 44/5 = 8.8, rounding we have that columns are 9.
./scytale.py -k 9 < test
This is the Scytale tool from CryptTools!
This method cracks the message with bruteforce and then checks every result validating the language to guess which result is the original text. By default language is English, for other languages read below.
./scytale.py < test
This is the Scytale tool from CryptTools!
Read from a file and print cracked keys:
./scytale.py -V < test
Text to crack: "Theoohe moi t lsSoCs cor!iyly st p aft tlrT" (44)
Testing matrix: 9x5 SUCCESS
This is the Scytale tool from CryptTools!
To decrypt a message without knowing the key you need to know the original text language. It is provided with --lang option. By default it is en_US (American English).
For example, in Deutsch language:
./scytale.py -t "Dies ist das Scytale-Tool von CryptTools!" -k 6
DtcTnTi yo oedtoCosaalrl sl ysi evp!sS-ot
If language is not provided it will try English and it will fail:
./scytale.py -V -t "DtcTnTi yo oedtoCosaalrl sl ysi evp!sS-ot"
Text to crack: "DtcTnTi yo oedtoCosaalrl sl ysi evp!sS-ot" (41)
Testing matrix: 2x40 Sorry. None of decrypted results seems to be written in language en_US...
Then, providing the correct language:
./scytale.py -V -t "DtcTnTi yo oedtoCosaalrl sl ysi evp!sS-ot" --lang "de_DE"
Text to crack: "DtcTnTi yo oedtoCosaalrl sl ysi evp!sS-ot" (41)
Testing matrix: 7x6 SUCCESS
Dies ist das Scytale-Tool von CryptTools!
You can check all available codes with ./scytale.py --help. You can even install more languages, for that take a look here.
If original text language is unknown you still can generate all possible transformations, then you will need to check them manually in order to know which is the correct:
./scytale.py -V -A -t "DtcTnTi yo oedtoCosaalrl sl ysi evp!sS-ot"
Text to crack: "Ddlooiaenoes- ls TCs Sor!icoy sylp tt t avT" (44)
...
Testing matrix: 2x21 Testing decrypted text:
"DltrclT nsTli yysoi oeevdpt!osCSo-soata"
Testing matrix: 21x2 Testing decrypted text:
"Dcniy etCsar lyieps-ttTT oodooalls s v!So"
Testing matrix: 3x14 Testing decrypted text:
"DtytoscCiTo nseTaviap l!yrsolS -osoeltd"
Testing matrix: 14x3 Testing decrypted text:
"DTioeosl ivsotn dCarsy pStcTyotoallse!-"
...
Testing matrix: 9x5 Testing decrypted text:
"DT oasi!ttioCll s c eor eS Tydslyv- nota spo"
Testing matrix: 6x7 Testing decrypted text:
"D tly!tyorsscoCliST o -nosseoTealvtida p"
Testing matrix: 7x6 Testing decrypted text:
"Dies ist das Scytale-Tool von CryptTools!"
Testing matrix: 8x6 Testing decrypted text:
"Dies is t das S cytale- Tool vo n Crypt Tools!"
...
Testing matrix: 2x40 Testing decrypted text:
"Dtt c T n T i y o o e d t o C o s a a l r l s l y s i e v p ! s S - o"
Sorry. None of decrypted results seems to be written in language en_US...
In addition, with the extra option -D you can check the language validation process:
./scytale.py -VAD < test
Text to crack: "Theoohe moi t lsSoCs cor!iyly st p aft tlrT" (44)
Testing matrix: 1x44 Testing decrypted text:
"Theoohe moi t lsSoCs cor!iyly st p aft tlrT"
theoohe: False | Progress 0% | Max 90%
moi: False | Progress 0% | Max 80%
lssocs: False | Progress 0% | Max 60%
cor: False | Progress 0% | Max 50%
iyly: False | Progress 0% | Max 40%
...
Testing matrix: 2x22 Testing decrypted text:
"Tohre!oioyhley msoti pt lasfSto Ctsl rcT"
tohre: False | Progress 0% | Max 86%
oioyhley: False | Progress 0% | Max 71%
msoti: False | Progress 0% | Max 57%
pt: True | Progress 14% | Max 57%
lasfsto: False | Progress 14% | Max 43%
Testing matrix: 22x2 Testing decrypted text:
"TeoemitlSC o!yys attrhoh o soscril tp f lT"
teoemitlsc: False | Progress 0% | Max 89%
yys: False | Progress 0% | Max 67%
attrhoh: False | Progress 0% | Max 56%
...
Testing matrix: 3x15 Testing decrypted text:
"TsshSteo oCpos h eca ofmrto! iit yltlr yTl"
tsshsteo: False | Progress 0% | Max 88%
ocpos: False | Progress 0% | Max 75%
eca: False | Progress 0% | Max 50%
ofmrto: False | Progress 0% | Max 38%
Testing matrix: 15x3 Testing decrypted text:
"ToeotsCc!lspa rho i Ssoiyt ftTehm lo ry tl"
toeotscc: False | Progress 0% | Max 89%
lspa: False | Progress 0% | Max 78%
rho: True | Progress 11% | Max 78%
ssoiyt: False | Progress 11% | Max 56%
fttehm: False | Progress 11% | Max 44%
...
Testing matrix: 5x9 Testing decrypted text:
"ToClahisyfe totcs o otthlr les!pr Si Tmoy"
toclahisyfe: False | Progress 0% | Max 88%
totcs: False | Progress 0% | Max 75%
otthlr: False | Progress 0% | Max 50%
les: False | Progress 0% | Max 38%
Testing matrix: 9x5 Testing decrypted text:
"This is the Scytale tool from CryptTools!"
this: True | Progress 12% | Max 100%
is: True | Progress 25% | Max 100%
the: True | Progress 38% | Max 100%
scytale: False | Progress 38% | Max 88%
tool: True | Progress 50% |
SUCCESS
This is the Scytale tool from CryptTools!
-A and -D options may be too verbose, avoid using them for long texts.
You can also set the permissiveness of the language validation process with the --threshold -T option. By default it is set to 50 (half of the text words must be written in the specified language in order to accept it as the original text). Values must be between 1 and 100. Values below 20% are not recommended (an encrypted text may be accepted as decrypted). Higher values indicate toughness, but 100% it is neither recommended (in the text may be non-english nouns and other original but non-english words).
As you can see, it is too easy to crack this classical cryptographic system so it is not recommended to use it in production software.
--> What is the Vigenère Cipher? <--
./vigenere.py --help
usage: vigenere.py [-h] [-t TEXT] [-k KEY] [--decrypt] [--exhaustive] [-V]
[-A] [-D] [-T THRESHOLD] [--beep]
optional arguments:
-h, --help show this help message and exit
-t TEXT, --text TEXT text to read from. If not specified the program will
read from standard input
-k KEY, --key KEY key used to encrypt or decrypt. If no key is provided
the program will try to crack and decrypt the text
--decrypt use the key to decrypt the text. If no key is provided
this argument is redundant
--exhaustive tests all possible keys. If not provided this program
only will test keys below length 100 while cracking
-V, --verbose show extra information
-A, --all show decrypted text for each tested key
-D, --debug show information about text validation
-T THRESHOLD, --threshold THRESHOLD
valid word count percentage to mark the whole text as
valid language (default: 50)
--beep plays a beep sound when program finishes. May require
SOX to be installed
./vigenere.py -t "This is the Vigenere tool from CryptTools!" -k "CRYPT" > test
Vygh bu kft Okxccxtv rdhn wpdf EiwemVfmal!
-t argument is not mandatory, so if you need to encrypt a long text you can skip it, execute ./vigenere.py "CRYPT" > test and then paste your text. When completed press Return and then finish the input with Ctrl+D so the program will read it.
./vigenere.py -t "This is the Vigenere tool from CryptTools!" -k "CRYPT" -V
Key "CRYPT" shifts: [2, 17, 24, 15, 19]
Vygh bu kft Okxccxtv rdhn wpdf EiwemVfmal!
To decrypt you will need to know the encryption key. In the example above it is CRYPT.
You need to provide the argument --decrypt to use the key to decrypt.
./vigenere.py -k "CRYPT" --decrypt < test
This is the Vigenere tool from CryptTools!
This method cracks the message testing with statistical methods and bruteforce and then checks every result validating the language to guess which result is the original text. In comparison to previous methods this will only work if the original text language is English because of the statistics methods it is using. In addition, in relation to this it is worth noting that long keys may contain some wrong letters. Given this limitation the maximum key length allowed by default is 100. Nevertheless you can force the program to test all possible key lengths using the argument --exhaustive. Also take in account that for small texts this program may fail, as we can see in the example below.
./vigenere.py < test
Sorry. None of decrypted results seems to be written in language en_US...
If you want to try more keys execute this program again with the option --exhaustive. However, it is worth noting that the longer the key is the more errors can have the cracked key. In addition, this program may have difficulties to crack keys on smaller texts in comparison with the key length.
Providing longer texts may be more successful:
test:
This method cracks the message testing with statistical methods and bruteforce and then checks every result validating the language to guess which result is the original text. In comparison to previous methods this will only work if the original text language is English because of the statistics methods it is using. In addition, in relation to this it is worth noting that long keys may contain some wrong letters. Given this limitation the maximum key length allowed by default is 100. Nevertheless you can force the program to test all possible key lengths using the argument --exhaustive. Also take in account that for small texts this program may fail, as we can see in the example below.
./vigenere.py -k "CRYPT" < test > test_encrypted
Vygh fgkfdw eiyrdu kft fgjqpzg kchmkee lbvy qitvzqiberj bxvymsl ceb qkwkcuhttc pgf kftg eycrdu vttka ichnnk tpekuyibpx rwx nrlvncxc ih ilchl yygra tvqjev zq iag fpxzkeya mgor. Xg efkettzqdg vf ngxxzmjl ovrwhfj rwbu ngae qejn pqii xy vyc dkkxgctn kcmm nrlvncxc xl Geeabuy ztvclqt hh kft lvrrxlvzah fgkfdwu zr xl wjgcz. Ke yswkkgdg, ke pteckgdg vf rwbu zr xl yfpia pfrxgi kfpm nflv dgpq bta tmcmczl hhov ughpx jtmvvph. Zkmcc mjzq abozrpmkfl iag dymbolk zxa ccczvy yaeqncs ua ucutwcr xl 100. Pvttkvycaxuj wdn erl uhttc iag gpdztrk ih vvqi tnc ndluzzax mvw axpxrwl wjgcz vyc pkilktgv --vvwtwjrxog. Rjhh vrit bp rarhwer iack ddk udyae vvvil vygh itfegto dyn yczj, pl yv apg uvc xg vyc tqcdnax dvjdp.
./vigenere.py < test_encrypted
This method cracks the message testing with statistical methods and bruteforce and then checks every result validating the language to guess which result is the original text. In comparison to previous methods this will only work if the original text language is English because of the statistics methods it is using. In addition, in relation to this it is worth noting that long keys may contain some wrong letters. Given this limitation the maximum key length allowed by default is 100. Nevertheless you can force the program to test all possible key lengths using the argument --exhaustive. Also take in account that for small texts this program may fail, as we can see in the example below.
Print cracked key and other information about the cracking process:
./vigenere.py -V < test_encrypted
Text IC (Index of Coincidence): 0.04250925751584761
Friedman test suggests a key length of 8
Testing key length 8
Key "SRAZSCKX" shifts: [18, 17, 0, 25, 18, 2, 10, 23]
Kasiki examination
Finding sequence duplicates and spacings...
100%
Extracting spacing divisors...
Testing key length 5
Key "CRYPT" shifts: [2, 17, 24, 15, 19]
SUCCESS
Key: CRYPT
This method cracks the message testing with statistical methods and bruteforce and then checks every result validating the language to guess which result is the original text. In comparison to previous methods this will only work if the original text language is English because of the statistics methods it is using. In addition, in relation to this it is worth noting that long keys may contain some wrong letters. Given this limitation the maximum key length allowed by default is 100. Nevertheless you can force the program to test all possible key lengths using the argument --exhaustive. Also take in account that for small texts this program may fail, as we can see in the example below.
Print even more information:
./vigenere.py -VA < test_encrypted
Frequencies: [('v', 39), ('c', 37), ('g', 34), ('k', 32), ('t', 32), ('r', 27), ('x', 26), ('y', 25), ('l', 24), ('e', 23), ('a', 23), ('p', 22), ('z', 22), ('i', 21), ('h', 20), ('f', 20), ('d', 20), ('w', 17), ('u', 17), ('n', 17), ('j', 15), ('q', 15), ('m', 15), ('b', 12), ('o', 7), ('s', 3)]
Text IC (Index of Coincidence): 0.04250925751584761
Friedman test suggests a key length of 8
Testing key length 8
Subgroup 1 (IC: 0.041851106639839035)
vdkzevbehfuhuxcyjfmkvjjeikmxbqvhzzgcwfglahxzqkmcecparaknaxzkwhrayvgzgcd
Testing subkey 'A' with match score 57%
Testing subkey 'B' with match score 56%
...
Key "SRAZSCKX" shifts: [18, 17, 0, 25, 18, 2, 10, 23]
Kasiki examination
Finding sequence duplicates and spacings...
100%
{'kft': [55, 220, 165], 'rdu': [65], 'vzqi': [90], 'nrlvncxc': [105], 'hfgkfdw': [245], 'rxl': [15, 50, 155, 140, 105], 'zkey': [125], 'kgdg': [10], 'dgvf': [125], 'rwbu': [110], 'uzrxl': [35], 'rxg': [155], 'iag': [230, 290, 60], 'aeq': [200], 'vttk': [320], 'vyc': [210, 270, 60, 355, 145, 85], 'uhttc': [355], 'qit': [400], 'pxrw': [355], 'lwjgcz': [200], 'czvy': [85], 'yae': [130], 'vvv': [40], 'vygh': [520]}
Extracting spacing divisors...
Kasiki possible key lengths (sorted by probability):
[5, 10, 2, 4, 15, 20, 3, 7, 11, 35, 40, 6, 25, 30, 55, 9, 13, 14, 50, 65, 16, 18, 21, 22, 26, 29, 45, 70, 80, 90, 100, 12, 17, 23, 27, 28, 31, 32, 33, 42, 44, 46, 49, 52, 54, 58, 60, 64, 71, 85]
Testing key length 5
Subgroup 1 (IC: 0.07237673830594185)
vgeuggkvvevcwtfeuankpnciytvgkgetvxofuqqvknncguchvvguwkkkcvuypingacopvkjokgoavqawpvuetgtvnumpwvivwgvpwcuvvtocyuvcd
Testing subkey 'A' with match score 53%
Testing subkey 'B' with match score 54%
Testing subkey 'C' with match score 69%
Testing subkey 'D' with match score 54%
Testing subkey 'E' with match score 53%
Testing subkey 'F' with match score 46%
Testing subkey 'G' with match score 60%
Testing subkey 'H' with match score 54%
Testing subkey 'I' with match score 58%
Testing subkey 'J' with match score 52%
Testing subkey 'K' with match score 54%
Testing subkey 'L' with match score 55%
Testing subkey 'M' with match score 52%
Testing subkey 'N' with match score 59%
Testing subkey 'O' with match score 58%
Testing subkey 'P' with match score 52%
Testing subkey 'Q' with match score 48%
Testing subkey 'R' with match score 58%
Testing subkey 'S' with match score 56%
Testing subkey 'T' with match score 56%
Testing subkey 'U' with match score 54%
Testing subkey 'V' with match score 53%
Testing subkey 'W' with match score 54%
Testing subkey 'X' with match score 52%
Testing subkey 'Y' with match score 52%
Testing subkey 'Z' with match score 51%
Best subkey is 'C' with match score 69%
...
Subgroup 4 (IC: 0.06368520859671302)
hdrtphliibsquptrthpiwvihrjixaxedgjwwanxdcmvxatttxhdxcsdtdwxixpvbchgthcapimzcasuxtaduidiidaawcptwxhtridaihgnppxtad
Testing subkey 'A' with match score 63%
Testing subkey 'B' with match score 58%
Testing subkey 'C' with match score 53%
Testing subkey 'D' with match score 57%
Testing subkey 'E' with match score 59%
Testing subkey 'F' with match score 57%
Testing subkey 'G' with match score 57%
Testing subkey 'H' with match score 57%
Testing subkey 'I' with match score 55%
Testing subkey 'J' with match score 56%
Testing subkey 'K' with match score 60%
Testing subkey 'L' with match score 56%
Testing subkey 'M' with match score 54%
Testing subkey 'N' with match score 51%
Testing subkey 'O' with match score 58%
Testing subkey 'P' with match score 77%
Testing subkey 'Q' with match score 58%
Testing subkey 'R' with match score 54%
Testing subkey 'S' with match score 52%
Testing subkey 'T' with match score 57%
Testing subkey 'U' with match score 57%
Testing subkey 'V' with match score 57%
Testing subkey 'W' with match score 56%
Testing subkey 'X' with match score 53%
Testing subkey 'Y' with match score 52%
Testing subkey 'Z' with match score 53%
Best subkey is 'P' with match score 77%
Subgroup 5 (IC: 0.0586283185840708)
fwdfzmbtbxlkhggdknebxnhlaeazmgtgxlhbepyktmnlbvhllfwlzwgegblagmdtmhhmzmbmabxzeutlkxnhazhtlxxlzkgtohbhakelitylggqxp
Testing subkey 'A' with match score 56%
Testing subkey 'B' with match score 54%
...
Testing subkey 'R' with match score 54%
Testing subkey 'S' with match score 62%
Testing subkey 'T' with match score 75%
Testing subkey 'U' with match score 56%
Testing subkey 'V' with match score 52%
Testing subkey 'W' with match score 51%
Testing subkey 'X' with match score 58%
Testing subkey 'Y' with match score 60%
Testing subkey 'Z' with match score 62%
Best subkey is 'T' with match score 75%
Key "CRYPT" shifts: [2, 17, 24, 15, 19]
SUCCESS
Key: CRYPT
This method cracks the message testing with statistical methods and bruteforce and then checks every result validating the language to guess which result is the original text. In comparison to previous methods this will only work if the original text language is English because of the statistics methods it is using. In addition, in relation to this it is worth noting that long keys may contain some wrong letters. Given this limitation the maximum key length allowed by default is 100. Nevertheless you can force the program to test all possible key lengths using the argument --exhaustive. Also take in account that for small texts this program may fail, as we can see in the example below.
In addition, with the extra option -D you can check the language validation process in the same way that with previous cryptographic systems.
-A and -D options may be too verbose, avoid using them for long texts.
You can also set the permissiveness of the language validation process with the --threshold -T option. By default it is set to 50 (half of the text words must be written in the specified language in order to accept it as the original text). Values must be between 1 and 100. Values below 20% are not recommended (an encrypted text may be accepted as decrypted). Higher values indicate toughness, but 100% it is neither recommended (in the text may be non-english nouns and other original but non-english words).
As you can see the choice of the key is important. In order to have a secure encryption you will need to provide a key of at least the length of the text and it needs to be unique for that text. However, nowadays there are many others cryptographic systems more advanced and useful.
--> What is the AES Cipher? <--
./aes.py --help
usage: aes.py [-h] [-t TEXT] [-in INFILE] [-out OUTFILE] [-k KEY]
[-kf KEYFILE] [-m MODE] [--decrypt]
optional arguments:
-h, --help show this help message and exit
-t TEXT, --text TEXT text to read from (by default: standard input bytes).
If text contains non-printable characters (special
bytes) then you will need to use --infile and
--outfile in order to do the conversion.
-in INFILE, --infile INFILE
file to read from
-out OUTFILE, --outfile OUTFILE
file to write the result
-k KEY, --key KEY AES key used to encrypt or decrypt
-kf KEYFILE, --keyfile KEYFILE
file containing the AES key used to encrypt or decrypt
-m MODE, --mode MODE operation mode, by default CBC. Supported: ECB, CBC,
CFB, OFB, OPENPGP
--decrypt use the key to decrypt the text
This examples uses 128-bit key, but you can also use keys with 192 and 256 bits.
For text:
./aes.py -t "This is the AES tool from CryptTools!" -k "Sixteen byte key" -out test.enc
Bytes:
b'\x8f\x88NP\x8e^5&\x95\x92<\x9c\x93s\xa1D;LP\x01\x17\x83@\xeb\x8d\x7fM\xbeZ\xc4\x10\x15\xd8K\xe2\x03O\xa5\x98\xcf\xef\xc0\x1c\x8dg>\xa62,Gx\xb9E\xa7\xe7&r\xedb"\xf7\x05+\x05'
test.enc:
ʤÚ@�UXK*3p�IR��„$�bÄÕ�ó>Žx �·Ïô5S ªm¹ÿ’z>�3�Xe¿�þAÙí�ꀖÀR{å!vÕ
-t argument is not mandatory, so if you need to encrypt a long text you can skip it and paste your text in standard input. When completed press Return and then finish the input with Ctrl+D so the program will read it.
For any file format:
In order to encrypt images, binaries and other files you must provide input and output files with -in and -out parameters.
./aes.py -k "Sixteen byte key" -in test -out test.enc
Use the --decrypt parameter.
./aes.py -k "Sixteen byte key" --decrypt -in test.enc -out test
test:
This is the AES tool from CryptTools!
You can specify a key of 128, 192 or 256 bits which is inside a file:
./aes.py --keyfile test.key --decrypt -in test.enc -out test
test.key:
Sixteen byte key
You can specify the operation mode used for cipher multiple blocks.
./aes.py --keyfile test.key --decrypt -in test.enc -out test -m "ECB"
Defaults to CBC.