Improving the Rule-based attack
I have a problem regarding custom rules what I need is simple but it does not know if it's possible because currently the Hachcat v.0.47 is not supported, and come to the problem.

Need to create rules so

Example 1



i6-1 0123456789 -2 ABCDEF ?1?2 i7-1 ABCDEF -2 0123456789 -3 !"#$%&'()*+,-./:;<=>?@[\]^_`{|}~ -4 abcdef ?1?2?3?4


i6-1 0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ (?1)?1?1?1?1?1?1?1?1[?1]
between parentheses = i6X
between brackets = iAX

However is not supported by the rules and mask is not possible this way creates.

Only then is supported

Example 2

i60 i70 i80 i90
i60 i70 i80 i91
i60 i70 i80 i92
i60 i70 i80 i93
i60 i70 i80 i94
i60 i70 i80 i95
i60 i70 i80 i96
i60 i70 i80 i97
i60 i70 i80 i98
i60 i70 i80 i99
i60 i70 i80 i9A
i60 i70 i80 i9B
i60 i70 i80 i9C
i60 i70 i80 i9D
i60 i70 i80 i9E
i60 i70 i80 i9F
i60 i70 i80 i9G
i60 i70 i80 i9H
i60 i70 i80 i9I
i60 i70 i80 i9J
i60 i70 i80 i9K
i60 i70 i80 i9L
i60 i70 i80 i9M
i60 i70 i80 i9N
i60 i70 i80 i9O
i60 i70 i80 i9P
i60 i70 i80 i9Q
i60 i70 i80 i9R
i60 i70 i80 i9S
i60 i70 i80 i9T
i60 i70 i80 i9U
i60 i70 i80 i9V
i60 i70 i80 i9W
i60 i70 i80 i9X
i60 i70 i80 i9Y
i60 i70 i80 i9Z
i60 i70 i81 i90
i60 i70 i81 i91
i60 i70 i81 i92
i60 i70 i81 i93
i60 i70 i81 i94
i60 i70 i81 i95
i60 i70 i81 i96
i60 i70 i81 i97
i60 i70 i81 i98
i60 i70 i81 i99
i60 i70 i81 i9A
i60 i70 i81 i9B
i60 i70 i81 i9C
i60 i70 i81 i9D
i60 i70 i81 i9E
i60 i70 i81 i9F
i60 i70 i81 i9G
i60 i70 i81 i9H
i60 i70 i81 i9I
i60 i70 i81 i9J
i60 i70 i81 i9K
i60 i70 i81 i9L
i60 i70 i81 i9M
i60 i70 i81 i9N
i60 i70 i81 i9O
i60 i70 i81 i9P
i60 i70 i81 i9Q
i60 i70 i81 i9R
i60 i70 i81 i9S
i60 i70 i81 i9T
i60 i70 i81 i9U
i60 i70 i81 i9V
i60 i70 i81 i9W
i60 i70 i81 i9X
i60 i70 i81 i9Y
i60 i70 i81 i9Z

What makes the process of creating a Herculean labor rules

It would be possible to have support built-in charsets?
?l ?u ?d ?s ?a ?h ?D ?F ?R
to the rules in the next versions of hashcat?

Would be possible by support masks for the rules?

Is there another way I can create rules both in Example 1 as in Example 2 automated? If there is no way to create automated Example 1, Example 2 is possible in this version of Hashcat?

If not, someone here has a suggestion for how I create Example 2 automated rule?

Is there any limitation on the method that I propose?

Compensates to brute force based on rules?

How many percent acceleration implementation we could have with this method?

Just do not want to write a rule Example 2 manually :/

In theory, I want to break the "ualpha-numeric" key, but create "ualpha-numeric" size 7+ wordlist becomes strained by the size of the wordlist to be generated, so with that in mind I want to create the "ualpha-numeric" wordlist size 6 characters and make key with the rest of the rule ( i = insert ) or even make brute force exclusively by the rules only doubt i have is offset, if in practice we would have more computational efficiency in the process if the process of making brute force by rules is faster than the classical brute force. But to create the rule to make the rest of the key to be broken the way it currently ta becomes a Herculean task therefore becomes indispensable enhance the rules for support masks (or otherwise ) of the rules.
The unique rules that could benefit from this new method I propose are:

Insert N @ X iNX Inserts character at position N


Overwrite @ N oNX character at postion N with X

But as anybody there who are the geniuses behind the project feel free to create whatever they want I just thank.

I thank any other trick inform
is there a tl;dr version of this post somewhere? preferably in english?

(04-09-2014, 09:57 PM)AmadorBR Wrote: What makes the process of creating a Herculean labor rules

i have absolutely no idea what this is supposed to mean. so... 7?

(04-09-2014, 09:57 PM)AmadorBR Wrote: It would be possible to have support built-in charsets?

hashcat already has this, and more.

(04-09-2014, 09:57 PM)AmadorBR Wrote: Would be possible by support masks for the rules?


(04-09-2014, 09:57 PM)AmadorBR Wrote: Is there another way I can create rules both in Example 1 as in Example 2 automated?

use maskprocessor to generate the rules.

(04-09-2014, 09:57 PM)AmadorBR Wrote: Pays to brute force based on rules?

like... money?

(04-09-2014, 09:57 PM)AmadorBR Wrote: How many percent acceleration impementação we could have with this method?

Google translation :E

But the problem with creating the rules by maskprocessor example 2 is the size of the file to be generated, I know not of miracles to do with the numbers will be even higher 32 the size of the key in this case, why would I say that so masks better, but since it is not yet possible thanks anyway any news we're watching.
For me, the ideal would be if there was support for example 1 rules, but since it is not supported yet, I look forward to one day be possible maskprocessor per hour will use the same to create the rule example 2 there is the idea thank force !
the size would be the same regardless (in terms of keyspace), implementing a mask processor into the rule engine would not change this fact.
It's true, but in a test here, I noticed limitations, requires a lot of RAM to load the "rulelist" (created by maskprocessor), outside the size of the "rulelist" being generated (for cases of i0X for iZX), I know that maybe with the implementation of the support for masks for the rules also would use a lot of RAM to load the rules virtually (not using the "rulelist" as in example 1) so I thought maybe index the amount of plains, the be generated (as with the words) could work, but they are just ideas, we also know that to predict the amount of plains to be generated for the process, so to make the count of the plains as it appears currently with the words that would lead to the creation of the option to continue the plain where stopped in the previous section. I'm thinking about how to improve the process of Plains and the Permutation (which involves Plains) more ideas I describe.
feel free to post a feature request on trac ( ), but i doubt this will be implemented any time soon if ever.