Exetools - Keygener Assistant 1.7 problem

Exetools (https://forum.exetools.com/index.php)

- General Discussion (https://forum.exetools.com/forumdisplay.php?f=2)

- - Keygener Assistant 1.7 problem (https://forum.exetools.com/showthread.php?t=13954)

Git	12-26-2011 21:09

Keygener Assistant 1.7 problem

Can anybody who is a member at www.at4re.com pass this to RobinHoodArab please.

AES encryption in keygener assistant V1.7 is sometimes producing incorrect encryption results. It will sometimes produce correct result on first run, then incorrect on next run.

Comparing to FIPS-197 standard test vectors :

Code:

Key: 000102030405060708090a0b0c0d0e0f

Plaintext: 00112233445566778899aabbccddeeff

Ciphertext: 69c4e0d86a7b0430d8cdb78070b4c55a

KA 1.7     : 04F6AA3FEB5CE4CD73BC9643D297743B



Key: 000102030405060708090a0b0c0d0e0f1011121314151617

Plaintext: 00112233445566778899aabbccddeeff

Ciphertext: dda97ca4864cdfe06eaf70a0ec0d7191

KA 1.7     : (no result)



Key: 000102030405060708090a0b0c0d0e0f101112131415161718191a1b1c1d1e1f

Plaintext: 00112233445566778899aabbccddeeff

Ciphertext: 8ea2b7ca516745bfeafc49904b496089 

KA 1.7     : (no result)

Maybe it is not using the correct number of Rounds which should be :
128bit key : 10
192bit key : 12
256bit key : 14 (maximum)

Also, encryption has to be done in 16byte blocks so maybe blocking routines are wrong. This app is a useful test base :
http://www.unsw.adfa.edu.au/~lpb/src/AEScalc/AEScalc.html

Git

mm10121991

12-26-2011 23:08

OK , It will be passed :)

Git	12-28-2011 00:49

Should also note that output length should be same as input length, and key length should alwys be 16, 24 or 32 bytes.

Git

mm10121991

12-28-2011 21:06

the developper of keygenner assistant is Mr Paradox/at4re but he is retired from their team
they didn't respond.

*RemedY*

12-28-2011 22:31

Git,

maybe you can give SnD Reverser Tool a try.
It has almost identical features and it works very well.

http://tuts4you.com/download.php?view.1923

The only shortcoming is it's GUI is not as userfriendly as that of Keygener Assistant but hey...

Regards
*RemedY*

H4X0R

12-28-2011 22:53

I think Mr Paradox used a ready-made library to encode AES.

Git	12-28-2011 23:04

Thanks Remedy, looks like a good tool.

H4X0r - AES isn't the problem I have AES code builtin to my app, I just wanted a few sources to check it again. They all agreed except KA.

Git

Git	01-29-2012 22:27

Another problem with KG 1.7 after using for a few calcs.

BigCalc tab (set to base 10)

P : 2
Q : 163
Click on P to the power of Q (P^Q)
Result : 256
(should be 11692013098647223345629478661730264157247460343808)

Git

mm10121991

01-30-2012 00:08

Check cryptokg.cjb.net you will find most of the tools you ever need. i think that ka is based on FGint lib so may be the bug is due to that lib

Git	01-30-2012 00:35

It wasn't a complaint, it was a bug report :)

Git

deepzero

01-30-2012 02:17

i dropped keygenner assistant a long time ago, lots of random bug and small mis calculations. I send a bunch of bug reports to at4re a long time ago, no response. :(

Ahmed18

01-30-2012 04:53

the problem is that the developer of this tool (Mr Paradox) left the reversing world :( .. No one has the source code of it to fix this bugs .. sorry!

sendersu

01-30-2012 14:36

Its not up to sorry, but its up to killed much amount of time and at the end you are cool - becaue you know that the tool was buggy.....
Mabye its good idea to put a comment near download link "use with caution" ? :)
To same these hrs for another REversers

Loki	01-30-2012 17:21

Quote:

Originally Posted by *RemedY* (Post 76564)

The only shortcoming is it's GUI is not as userfriendly as that of Keygener Assistant but hey...

Was written 2-3 years before KGA though and was only ever something for me to learn to code ASM a little better - started with a few routines and then me and Punk kept expanding it so the GUI isnt perfect.

Been considering making it open source if anyone wanted to learn from it, but the code isnt so great seeing as it was a learning piece.

deepzero

01-30-2012 23:51

never mind, go ahead! :)

Ahmed18

02-01-2012 00:50

Quote:

What is this stupid post ?!!!!!!!!! I am not the owner of this topic or the developer of this tool .. I also sorry for the owner of this topic [Git] not for you or anyone.
You don't like it so don't use it and don't write a silly comments again!!

JMI	02-01-2012 08:32

It appears that I must remind the group again that you are not to post comments and/or criticisms about eachother or posters on these Forums.

If you have a complaint about a Post or Threat, send me a Private Message or post in the Suggestions and Feedback Forum.

It is unnecessary to get upset with eachother and/or carry on a private spat (argument) here on these pages.

I hope this is enough warning. ;)

Regards,

Git	03-02-2012 23:54

I've made a start on writing a similar tool to KGA. At the moment, for some unknown reason, I've called it BNC. I'm writing it in C++ Builder in plain C. I'm using OpenSSL for its 'BN' big number library and for it's Crypto capabilities. I am supplementing BN with the GMP big number library.

I am not, at the moment, planning for more that 3 tabs, being Calculator, Encryption and Hashing. More could be added later. I want the first tab to be a very comprehensive arbitary precision calculator, so the 'conversion' part of kga at the bottom of it's first tab will not be implemented, in favour of being able to independantly convert any of the P, Q, N, R fields between base 2, 10 and 16. I've tested that conversion up to 800,000,000 bits so far!. Crypto will be as full an implementation of the OpenSSL functions as I can manage, probably without the X509 stuff. I am more interested in an accurate low bug tool than bells and whistles.

Any suggestion, comments or ideas welcome.

Git

Av0id

03-05-2012 12:53

Git, i know openssl is very powerful and complicated, but take a look at polarssl and miracl

Git	03-05-2012 20:41

I've used Miracl before and think GMP is better. Not heard of PolarSSL, I'll check it out, thanks.

Git

mrb!	03-06-2012 10:12

A warning though.

PolarSSL is GPL so you might need to open source your tool for it to be legal.

Git	03-06-2012 19:22

Same with Miracl, they both have commercial backgrounds too. I think I will stick with OpenSSL and GMP. They are both good, well established packages with a lot of current development. Converting between the two big number formats is quite easy too.

Git

Loki	03-06-2012 22:19

Not to put you off or anything (I dont think there is a 1-1 over lap here), but for awareness sake there is another new tool being developed by SADUFF

http://forum.tuts4you.com/topic/28538-sdf-tool-keygenning-tool/page__fromsearch__1

Git	03-06-2012 22:38

OK, more the merrier :). I'm not a member there so I can't look at it, but I'm sure they'll do a good job. Meanwhile, it would be good if anybody has any suggestions for what they would like to see please.

Git

Loki	03-07-2012 16:42

For anyone else interested but not a member over at t4u, the author is hosting on his own site

http://saduff.keygenme.com/downloads/SDF%20Tool.zip

dj-siba

03-08-2012 23:28

Quote:

Originally Posted by Git (Post 76531)

Can anybody who is a member at www.at4re.com pass...

Expect new Keygener Assistant version soon, Sorry for late reply.

Git	03-09-2012 00:31

Right, thanks. Pity somebody didn't say something earlier. Are they aware of the bugs?

Git

dj-siba

03-20-2012 19:53

Quote:

Are they aware of the bugs?

Yes, fixed see pics

Code:

http://i44.tinypic.com/29gc45.jpg

http://i40.tinypic.com/72zpcg.jpg

if there are more tell us
Any suggestion are welcome

Regards

deepzero

03-20-2012 19:58

perfect.
where can i grab that beta?

STRELiTZIA

03-22-2012 16:29

Hi,

Quote:

perfect.
where can i grab that beta?

Keygener Assistant v1.8

PHP Code:


		
			
http://www.2shared.com/file/JMkwZ6RS/Keygener_Assistant_v18.html

http://www.4shared.com/rar/V8gnNK82/file.html

http://www.mediafire.com/?ao7k8hb3j2vbok5

For any bug report, please feel free to post here:

PHP Code:


		
			
http://www.at4re.com/f/forumdisplay.php?18-ENGLiSH-SECTiON

Regards

uranus64

03-22-2012 18:32

@STRELiTZIA. Was posted already. Check this. Anyway, thank's.

Git	03-26-2012 00:49

Looks like 1.8 may still have some accuracy problems. Output of the Sapphire II hashes seem to be incorrect when compared to the authors test program (Michael Paul Johnson). Test program and Sapphire II hash are very old, so info may not be reliable. Also, author does not recommend using algo for hash bigger than 32 bytes (256 bits).

Code:

Sapphire II Hash of "Hello World"



bits   Author

128    DE0BCBC20DC0D9A8757D9F5A872691BE

160    DE0BCBC20DC0D9A8757D9F5A872691BE3B1386F1

192    DE0BCBC20DC0D9A8757D9F5A872691BE3B1386F149454785

224    DE0BCBC20DC0D9A8757D9F5A872691BE3B1386F149454785F33EF56A

256    DE0BCBC20DC0D9A8757D9F5A872691BE3B1386F149454785F33EF56A04C2CE57



bits   KGA1.8

128    CC5160EFFC89B049C7D98C621F40E962

160    9AB4E676C69D8D197A68FD91502B8C6BFEF7AE16

192    F594394418FD8F25842901D03584CD1927C8E267CDD5CB43

224    575A2E7589332F008514BC4BE7059F33AC7186734CDB4FC611F09322

256    7CE0CF43114C7F198098D08012E5E10BD0DE9E71B2C295607EBB094E3AAA1C4B

Source code of final hash generation is :

Code:

void sapphire::hash_final(unsigned char * hash,     // Destination

                          unsigned char hashlength) // Size of hash.

{

        int i;



        for(i = 255; i >= 0; i--)

                encrypt((unsigned char) i);



        for(i = 0; i < hashlength; i++)

                hash[i] = encrypt(0);

}

That last loop suggests to me that first N bytes of N+4 byte hash should indeed be the same.

Git

Git	03-26-2012 01:43

Also... most of the hashes should produce a valid hash of a zero length input string or byte array. You can see some examples in the Haval validation test program, where most of the 3 round hashes are validated against a zero length input string.

Git

All times are GMT +8. The time now is 07:03.