also here check this out
https://twitter.com/80vul/status/1052856212574760961
#cybersecurity
7 messages ยท Page 8 of 1
Damn man
We're all just tryna make life worth living
Why do they have to mess with Tor project staff
@velvet isle You're into security stuff or just trying it for fun?
@thorn obsidian If it's on a lower port, it's much more likely to be scanned than something in the 20,000 to 60,000 range
Masscan makes that somewhat irrelevant, but it keeps out a lot of bots and people who don't know how to properly do bulk scans
wrong channel but you're welcome ๐
How does only root having access to create a service on the port improve security from a remote attack?
That's understandable.
We have published our second audit of @monero $XMR Bulletproofs sponsored by the Monero community. @quarkslab found 30 issues including 8 critical/high severity bugs, one of which impacted the live chain. All issues have been patched.
https://seclists.org/oss-sec/2018/q4/82
GCC Compiler Induced Vulnerability - affects programs compiled with GCC 7 and 8 containing nested functions
Noob programmers clearly
If a Brit installs colorama it's not unlikely he types colorama by accident though
There is some out commented subprocess
Conclusion
The worse written the malware the bigger the scandal
Wew another malicious PyPI package
They really need to implement some automated scanning of dependencies, and make it clear what dependencies a package has
Defcon26 talks are up : https://www.youtube.com/playlist?list=PL9fPq3eQfaaD0cf5c7wkzMoj2kifzGO4U
Any goodies
I did some decrypting on my end today of https traffic
I noticed that almost the whole of whatsapp connections are end to end encrypted
Something like that
For most of the connections I was getting a .enc file
Meanwhile Facebook Messenger messages are not encrypted but just sent over a secure connection ๐
I use Signal, as far as I'm concerned
whats app has been end to end encrypted for years
but afaik they still didnt leak info about the implementation
yeah that's why I prefer Signal
it's fully open source
and the big Facebook isn't behind
If WhatsApp was e2e years now
Why didn't they tell us
Until some update in 2016 I think
Also before I remember WhatsApp had a subscription thing
They removed it
@silent pier
Exploitation via Faxes (Didn't attend but heard it was good): https://youtu.be/qLCE8spVX9Q
Processor exploitation. I attended this one, guy is super sharp, and this continuation of work he did last year on discovery of undocumented instructions (SandSifter): https://youtu.be/XH0F9r0siTI
(SandSifter: https://github.com/Battelle/sandsifter)
Smart city exploitation (just watched, it's interesting): https://youtu.be/5z-rKz5ABgI
PLC hacking: https://youtu.be/-KHel7SyXsU
There was another really interesting one about the big bad ICS malware that happened recently (Triton) but it seems that isn't up on YouTube.
Awesome resource, Thanks.
@thorn obsidian
it depends
there's nothing wrong with storing raw credentials on disl
but they're only as safe as the machine is, the software and hardware you run
Where else would they store them?
Yeah this guy didn't even email them
K pleb just because they stopped responding to some unrelated feedback for reasons you don't know, doesn't mean you should post what you think is a security issue to the twitters for all the internet fame without contacting Signal first.
Can probably look through the project and see if the reason is documented anywhere since it's open source.
Do some due diligence first c'mon
I would argue that it should at least provide an option to have the key encrypted with a password that it prompts for on launch
and should use BitLocker or whatever so the file is at least encrypted with your user account
that's what I would do
if I were designing an app like that
What if they store key after performing "key wrapping"?
@safe bear it's not even a valid complaint, what does the dick want
store the encryption key on the server, nullifying the encryption your service touts as its primary feature?
Physical implants into your hips
I'm still looking for a good physical biological implant
Im hesitant in general for most technological implants
Other that things that help you live, I suppose
why are you hesitant?
Cause it's technology
it's one thing to implant myself, it's another for a government to require it
Technology is never safe
disagree .:P
AndI having it implanted in my own body
Idk, just seems like a big risk with todays security
depends, might be decent two factor
it'be especially cool if it had a button somehow that you can feel
imagine using a magnet to activate your implant. You'd feel your skin pulling towards the magnet.
Two factors seem to just be a pretty face for many
Just like this crypto wallet that didn't even timeout too many incorrect 2fa attampta
what crypto wallet?
I forgot its name, was on reddit a day or 2 ago
Someone brute forced a 2fa to a guy, who claims he lost over a mil $ in crypto
Something like 85k attempts were made
I don't believe in stealing from people, but if they don't take security seriously and put that sort of money at risk, they basically deserve to lose it.
Yee
II would never put that much money in an online wallet
If you have $1m you can afford a physical wallet
It's like The Fappening with 4chan. What, you gonna litigate 100,000 people because a dumb blonde actor/actress used the password "password" to secure their nudes?
If they're going to enjoy technology, power, and fame, they should do a bit of work to educate themselves in how to wield it.
Silly people not realizing the weakness of their security
yeah, I mean I'm of two minds about that sort of thing.
I do wish bigger tv channels and sites promoted security a bit more
I hate scammers, but you can't act like other people are responsible for your best interests.
I mean, if you fall for an online scam as someone in their late teens / twenties and use the internet regularly
You might just deserve it
Compared to if they try to scam a 70yo who only access their desktop to pay bills, or look at mail
That's just harsh
what's more, some cultures treat it like a game
that they earn what they steal, because stealing is a skill
in china, most shops haggle, and if you don't, its your fault you pay 10x
It's just sadistic.
A Wi-Fi router flogged by British mobile network EE has a hidden administration account with a hardcoded username and password โ and is accessible via SSH.
This root-level account, present in EE's 4GEE HH70 gateways, can be accessed by anyone on the local network, such as a malicious user or malware on a Wi-Fi-connected PC.```
lol
can people stop this shit please?
"It's important to note that for this vulnerability to be exploited, you need to have local access. So the risk of this being exploited is low." - EE Spokesperson```
@lusty flare Wait,from where you got this?
The Register
They must have thought: Let's downplay a gaping hole in our security; that'll show the world how much we care about the security of our clients!
They should have buried gold in their backyard like the smart people
@errant pilot smartest idea I've ever heard
@valid furnace """it's not even a valid complaint, what does the dick want
store the encryption key on the server, nullifying the encryption your service touts as its primary feature?"""
Encrypting it locally with a password the user has to enter on startup would mitigate the problem. And if you actually go to the thread that's what they're saying too.
Or maybe just don't save history
Prone to memory analysis?
๐คท
Does anyone know of a good data destruction program? I am wanting to format my main harddrive so I change to a UNIX os, and would prefer to download a legit ddp
you don't need to wipe your disk just to be able to install a new os
regularly just using the new os installer to delete and create new partitions is enough
if you want to securely wipe your data because it's confidential stuff (though I see no problem unless you're going to give the disk away), you can e.g. use DBAN (Darik's Boot And Nuke, or so)
On a modern HDD, a single pass of overwriting should be totally enough for consumer stuff. I guess you're not storing military secrets on there ๐
Thanks, this is an older computer which hasn't been touched in a while, so I'm not sure what malicious stuff is hanging around. I just want a clean reset.
Just write zeros to it if that's all you want. It'll be unrecoverable
tbh you can just use the secure erase function that most hdd's have
I have an iso for this
Darik's Boot and Nuke
or, you can just encrypt your hard drives and not worry about it. ๐
yeah, i'm not sure about SSD's either
lol
You just use the ATA Secure Erase command for SSDs
That triggers a full flush
Technically, you could have some data in cells that are marked as failed, which would possibly lead to data leakage.
I haven't read the ATA spec, so I don't know if secure erase is required to wipe failed cells or not. Probably implementation-specific.
@thorn obsidian Why are they scary?
Forensically, getting data out of a SSD is nearly fucking impossible
It varies by manufacturer, but usually when you issue a TRIM command any sections marked as free are zeroed. I believe this is true on SandForce, the most common controller, but it was not the case on certain controllers back in 2014 or so I think.
If you get a SSD to do forensics on, you basically: immediately disconnect it from the host (disconnect SATA connector) and preserve power (if possible)
^ read the bibliography links on that page
Yes, it's still possible, however it's much more difficult, error-prone, and unreliable than standard magnetic disk data recovery
tl;dr sorry
Forensics is cool 
Yes
Standard practice for any bulk storage medium is wipe using the appropriate secure erasure method, then destroy the drive physically
An admin friend of mine would destroy HDDs by drilling through the middle of them, and I think that's pretty standard practice (seen in a few other places too)
NIST is the golden standard in data destruction
"Clear" the data using a single pass zeroing, then using the SECURE ERASE UNIT command
Then purge using a cryptographic method (overwrite with scrambled data), use SANITIZE command, or discard the keys for a drive with built-in block level crypto
Then you destroy the medium physically
Up to and including incineration
There's actually a specific incineration temperature they recommend for magnetic media, but I can't seem to find it at the moment (and in-flight wifi is slow AF)
Also, 3 passes is generally enough for a HDD. If you're paranoid, 7 passes, since that's what the DoD uses.
35 passes is ridiculous and IIRC has been shown to not have any measurable improvement in the randomness of the data over DoD 7-pass
tl;dr
Forensics are kinda interesting.
realistically, no one's reading data off a destroyed drive
physically destroyed
certain temperatures, the magnetic properties of a hard drive break down.
certain fluctuacting fields can wipe platters instantly
SSDs use a digital stored charge, not analog magnetivity
Yes
Incineration ensures the components are completely inert though
Not just practically destroyed
Yes
"the data will rise like a phoenix from the ashes"
realistically, unless the NSA is knocking on your door, nobody will read any data from a single pass overwritten HDD.
i feel like an overwrite of /dev/urandom then /dev/zero is enough
for most cases, of course
but why
cause that makes individual bits of data harder to read, then clears them.
nothing is readable after a single pass of zeroes either
modern disks are so small with such tiny magnetic tracks and space between, that it's really almost impossible to find anything useful from before the last overwrite
Using multiple passes with random data was important on early disks with larger tracks, where you could still position the head in between those and have a good chance of reading something
hmm, fair enough.
For the average person one or two wipes is fine
Data erasure (sometimes referred to as data clearing or data wiping) is a software-based method of overwriting the data that aims to completely destroy all electronic data residing on a hard disk drive or other digital media by using zeros and ones to overwrite data onto all ...
However, nation-states are likely to have developed drive and vendor-specific recovery techniques
I doubt, but to everyone their own freedom.
Details on the cryptocurrency VB script one ("colourama"): https://medium.com/@bertusk/cryptocurrency-clipboard-hijacker-discovered-in-pypi-repository-b66b8a534a8
Potential solution: https://theupdateframework.github.io/overview.html
You already have though
colourama, that guy was clever, poor British people
imagine
cant you make a tea emoji real quick
no blob though ๐ฆ
1.ย Arbitrary File Overwrite Vulnerability Leads to Privilege Escalation Details: ====== X.org X Server application is vulnerable to...
so get this
a lot of modern hard drive platters are made of glass.
destroying them fully may be as easy as a tuned sonic blast
or a strong impact
or a drill
yup
DEFCON23, they had a guy who runs a datacenter experimenting with nailguns, directional explosives, and thermite
nailguns look very promising
I'd prefer the explosives
platters in desktop / server drives are usually metal
not ceramics
not to say they can't be, it's just usually
laptop drives are more often ceramics than desktop drives
@errant pilot the problem is the explosives largely didn't work
and, by using explosives you commit tons of felonies in the process
so the law doesn't "need" your computer anymore to convict you
I bet degausing would be most effective
Not as fun
I mean
destroying evidence is illegal anyway
and beyond that, destroying your own property isn't illegal
ceramic, same thing
glass is legit btw
they most certainly can be made of glass
:P
and have been.
and still are.
er, no.
Ceramics and glass have been / still are used
@thorn obsidian Its IP based
You won't be using cell phone credit
Lol
Its internet based
ยฏ_(ใ)_/ยฏ
Hm
Idk
do you have to trust your ISP for Tor to work?
Lol
I didn't know they always had to be a non-magnetic material
yah
they coat it with a magnetic substrate
so that it responds consistently to magnetic fields
didn't they used to be iron? i thought the magnetic fields are too "small" for a magnetic or metallic material to interfere with it
even if you had a solid bit of magnetic metal it'd be hard to control the read/write
yes but you've got to think about storing and holding that magnetic information
it's easier to store data on a thin magnetic tape instead of a thick iron bar
Yea
Stuff is encoded to the sim bro
That'll enable you do do that
And some AP you'll have to configure for traffic to go through ur isp
Maybe orbot ye
Brass Horn
๐
This is for people who care about privacy and security
It should improve in the future
Was tor fast when it first came out ?
Yes, regular calls and texts can be intercepted
If you're not running Tor, then the Sim wont send data.
Once its encrypted yes
That SIM card can also not use calls / texts, it's data only.
^^
It's an interesting concept
Who doesn't care about that
I don't want my isp to monitor me
Tho I do nothing bad
get some of those data stuffers
things that basically browse on random shit in your name
Plus you guys know that facebook messanger texts are not encrypted right ?
yes
Just sent over ssl
Its terrible, I decrypted my traffic once and my token and all that was exposed
Uhuh
Wait
Stallman is still alive
Wooo
We have emotes of him in programming discord server
Yea
๐
Hold on
@thorn obsidian
lol
Hmm
Fantasy can be great
gtg now
That thumbnail tho
It looks so sad ๐ฅ
@thorn obsidian is that on top of your package?
hi guys
Like Sims are pay-as you go
Right?
If that is the rate for that sim... That's a better deal than anything else
Be careful what you install folks!
Tl;dr: Malware deployed on pypi as colourama to trick people trying to pip install the legitimate package colorama
ugh what
old news
Clearly I have the wrong news aggregator
Heh. I got it from Naked Security today, I blame them
"For over five years" Damn
Addition of Wireguard VPN was the big thing for .4
Along with an experimental R-Pi 3 B+ 64-bit image
Yup.
Heyo, I'm wondering if it is possible to send encrypted requests? ( I don't want my program to be listened to by wireshark and have the requests revealed )
if you are talking about http requests
just use https
the request will still be seen in wireshark though
the content is just gonna be nonsens
alright, ill get a quick cert for the subdomain then
And for obfuscating a .py (will end up as a .exe), any recommendations?
you can compile it to a .exe using for example pyinstaller
Yeah I'm aware of pyinstaller but do you know of some good way to obfuscate it?
A good obfuscator
I'll search, cheers ^_^ I was wondering if there was a way to compile and obfuscate at the same time
also when you start thinking about obfuscation you should think if you really have to do this
-> install kernel driver to hide your files
no i mean
install a kernel driver to hide the program files from anything the user does
or can i do that with LUKS
@thorn obsidian it's a problem if you haven't patched
and considering the majority of android phone producers don't sell the products pre-patched with latest versions etc
the android market place can be a security nightmare
also pretty sure @thorn obsidian never said it was a problem, just posting security related news and shit :3
was the full writeup available at that point in time?
Hmmmmmmm?
zimperium were the ones who reported it
^
it's a writeup of a security issue
not everything i post has to be a new critical security problem lol
@upper phoenix If you want to obfuscate Python code, you need to transform it into C-code, then compile the C-code into an executable. Cython is the general go-to tool for this, as well as Nuitka.
Cython: http://docs.cython.org/en/latest/src/quickstart/index.html
Nuitka: http://nuitka.net/doc/user-manual.html#usage
Nuitka Home
Nuitka User Manual
Contents
Nuitka User Manual
Overview
Usage
Requirements
Command Line
License
Use Cases
Use Case 1 - Program compilation with all modules embedded
Use Case 2 - Extension Module
Nuitka can be run on your existing code without any modifications, so I'd give it a try before reaching for Cython. Cython seems simple, but I've heard it tends to be a lot more work than you initially estimate going in.
@safe bear Thank you so much ๐
Hi All! There is a new interesting CVE. An incorrect permission check for -modulepath and -logfile options when starting Xorg X server allows unprivileged users with the ability to log in to the system via physical console to escalate their privileges and run arbitrary code u...
Se Linux for da win
you could also argue that microsoft has written lots of kernel stuff and because of that you shouldnt use the kernel anymore
NSA everywhere
๐
hey @thorn obsidian have you got the LAIRETAM DEI?
K
==ATPRVSQF0QgUGa0BCdhBydvJncv12b0BSdvlHIlV2U
CVE-2018-5407: new side-channel vulnerability on SMT/Hyper-Threading architectures
https://seclists.org/oss-sec/2018/q4/123
almost everything you say feels like shitposting and i'm not sure why you keep doing it
it's a stack based overflow in bluetooth chip firmware
just because you need to be in bluetooth range doesn't mean it's not valid
not everything has to be a remote RCE 0day in a google product
jesus
also priv esc is an entire class of exploits and you constantly ignore them because "if you can run code that's obviously enough already"
which is just dumb
"kill chains"
Rarely are exploits used in isolation
In other words MS08-067 and friends are unusual
Usually you are combining multiple vulnerabilities and design flaws to achieve an objective
This The Atlantic ย article asks why hasn't there been a cyber-terrorist attack for the last 15 years, or as it phrases it: National-secur...
Stealing Chrome Cookies without root or password on OSX, Linux, and Windows via Remote Debugging Protocol.
sounds good
write about which ones are easily crackable and which of them can have DoS problems or whatever
i know for a long time anything written in python2 that was hashing user input was vulnerable to DoS attacks because of the hashing algorithm
ICYMI, hashcat v5.0.0 comes with a new "brain" feature. You won't have to "crack" previously computed hashes https://t.co/OEmXnmJq58 #password #cracking
Some SSDs have had their full-disk encryption password be 32 NULL bytes, no matter your input ๐คฆโโ๏ธ https://t.co/fb60lcxbMo (PDF) #encryption #fail
Likes
112
Has anyone used ZAP to simulate attacks of applications?
Question moved
There is someone who was a string for example :
String = โBztVsvjNgsnsisMvkdbโ
Okay so that string is very encoded with base64 and then with rot13
But... he added letters to the string so basically when you rot13 decode it then base64 decode it it will give random stuff because he basically messed with letters.
He did it on purpose
If youโre wondering how he decrypts it he take the extra letters he put in the string and then decode it from rot13 to base64 and get the correct code
Hereโs what I mean
This guy a has astring like I said...
String = โjskeheldiwnfolwroGoVjCkfโ
(Not exact string lol just example ^^)
So this guy encoded it with base 64 right?
THEN
He added random letters to that string to corrupt it or whatever
THEN
He used rot 13 to encode that base64 string right?
THEN
He did the same thing he added like random letters random places just to corrupt it
SO
If you try to rot13 decode it and then base64 decode it wonโt give you the actual code it will be messed up
How can I know which letters to remove from each encoding
Any help is appreciated
By idk if itโs possible because itโs gonna take ages maybe to know which letters are the extra ones
Is this an assignment or something you found in the wild?
I would check for characters that aren't in base64
@native edge itโs a challenge for $50
Also he can use characters like GLbcKfX or whatever how am I suppose to know ...
What are the clues
you could try bruteforcing it if the string isn't too long
not without any of the info no
@native edge if I give u the string would u try
I could give it a quick look but it's 2:30 so i'm off soon
Ok imma send it tomorrow on this channel because rn on Iโm on my phone
And I forgot what the link is because itโs on my pc
@silent pier ZAP is more of a reverse proxy with plugins not an attack tool, and it doesn't simulate attacks it just attacks stuff if you run the plugins
ยฏ_(ใ)_/ยฏ
Hello
Have an important question
Anyone who is familliar with python internals here?
!t ask
ask
Asking good questions will yield a much higher chance of a quick response:
โข Don't ask to ask your question, just go ahead and tell us your problem.
โข Try to solve the problem on your own first, we're not going to write code for you.
โข Show us the code you've tried and any errors or unexpected results it's giving
โข Keep your patience while we're helping you.
You can find a much more detailed explanation on our website.
Ok so
Python processes a string in a certain way say I write
In python 2.7
Print '{}'.format(a)
If i do something that resembles sql injection it just prints the statement as a string
But in sql injection it doesnt it like actually prints The code itself there even though it asks it in like $_GET
I'm wondering why is it
And how does these two different language process strings and how python defends from it
You are confusing an SQL query with a string
I know the data structure is a pystringobject
But why I'm just trying to understand it more in depth to understand why sql doesn't do the same or why it can't
A string is just some characters with no special meaning attached to them.
If you want to execute the code in a string there is eval() and exec() but those are generally adviced against
But the sql query recieves a string isn't it?
and executes it
Sorry just still don't understand why sql actually takes the input and doesn't apply it as a string but as something else
If it wouldn't take the string it would do nothing at all
Cause it compares it to something else isn't it?
Hmm I understand that just not the logic behind it
Like why would they no santize it all I'll look into it thougj
Thanks ๐
How would SQL know what to sanitize
txtUserId = getRequestString("UserId");
txtSQL = "SELECT * FROM Users WHERE UserId = txtUserId;
Let say I have this
I don't understand why if i enter OR it actually parses it as a comamnd
Cause usually in the queries the stuff you need to get is a string int etc.. not a command
So why does it even allows it
Maybe I'm missing something big here
But this is the reason i'm looking into it
That stuff doesn't happen in SQL itself, it happens in php, python, js etc and then gets passed to SQL
You would do the sanitizing in those parts
Ok cool thanks
basically, the programming langauge (python, PHP, etc) will substitute values in the string txtSQL = "SELECT * FROM Users WHERE UserId = txtUserId; for actual values (in this case txtUserId is replaced with the value of a variable) and then passes that to the SQL database, which executes it. SQL has no idea anything has been injected, the manipulation of the values happens in the programming language not the SQL server itself.
the reason python strings are not vulnerable to injection is the same reason SQL isn't vulnerable if you use parameterized queries, basically it just tells the parser that what it's reading is text and shouldn't be interpreted as commands
parameterized queries, is that what they're called?
This is a perfectly legitimate program.
On the topic of sql, whats the package for detecting sql injections, written in python
Uh- i mean it's a software
Not a package
gah Imma have to dig through knowns logs
sqlmap 
Huh
ZAP did things to this notes part of the app im testing
Interesting 
So I took this code from online source
from Crypto.Cipher import AES
password ="Yp2s5v8y/B?E(H+MbQeThWmZq4t6w9z$" #generated password
obj = AES.new(password, AES.MODE_CFB, 'This is an IV456')
message = "Answer is this"
ciphertext = obj.encrypt(message)
print ciphertext
obj2 = AES.new(password, AES.MODE_CFB, 'This is an IV456')
dec = obj2.decrypt(ciphertext)
print dec
So
new question
so basically idk why this doesnt work
so basically ive got the encryption thing right?
the
ciphertext = obj.encrypt(message)
so i put it on gist for a test I made a secret and got the link
and i did this...
from Crypto.Cipher import AES
from urllib2 import urlopen
connection = urlopen (link).read()
IV = "93KMsZC8914MXaWQ"
password ="""GZz9ex8WZ#Q$XuPdQJCRVTsDx+j$G!^x""" #generated password
message = connection
obj2 = AES.new(password, AES.MODE_CFB, IV)
dec = obj2.decrypt(message)
print ("Decryption : %s"%(dec))
but it gives me another thing how could i fix it am i doing something wrong
like bunch of random letters
@thorn obsidian not really helpful in this situation?
@thorn obsidian if you are URL opening the URL to the gist what you receive is the html containing the stuff submitted to the gist platform. And as the html does of course not equal the normal text and as CFB makes the blocks depend on each other you just get a bunch of non sense
Also why are you using python 2.7 xample?
Pythons cryptography library support in 3.x is very good too, there is no reason to stay with 2.7 in this situation
lol district health software @silent pier
also here is a thing
https://www.helpnetsecurity.com/2018/11/07/virtualbox-guest-to-host-escape-0day/
Yes it was dhis or some other software that was handed down to us
@orchid notch because I started at code academy and they started python 2.7 with me so I got used to it and everything
I want to switch to python 3.x but idk where to start or whatever lol
Also idk which 3.x version to use
Where do I really start @orchid notch
Start with what
The thing is Iโm mostly scared of my old projects not working on python 3.x too
Where do I start learning 3.x I was hoping like codeacadmey where they ask you to build the code itself
Also what if there modules python 3.x donโt have
The steps from 2.7 to 3.x are really minor
At least syntax wise
You will be able to Google the vast majority of errors you get if you are already familiar with 2.7
Hm well
Iโm probably gonna get 3.7 today
We've released an advisory to address the concerns around #BitLocker and the recently disclosed vulnerabilities in self-encrypting #SSDs. See https://t.co/YJUChp2Nwr to see how to turn on software encryption. You will not need to reformat the drive or reinstall application...
Retweets
242
Likes
263
Bitlocker on windows sometimes used hardware encryption, which is literally none-existant on some SSDs
because of course it was
muh 32 bit null encryption password amiright
Has anyone written a Threat model for some software / a web application before?
yah @thorn obsidian
that's actually a problem for us
We used Bitlocker on a couple of laptops with Corsair SSD's in
had to Veracrypt them
chinanumberone
@silent pier never had the need to write a threat model but owasp is generally a good resource for all things security related to applications
Yeah, I've read through it
But I'm kind of stuck as I have to provide a Threat model to something I'm not involved in designing nor developing
hmm thats tricky
So i'm supposed to write one based on results from a few analysis and pentests
But it's sort of too late to write one.. as the software is finished i feel
ah, well did they resolve the issues from the pentest ?
There weren't too many to mention
and i feel like the results from the pentests itself would be a great start for the threat model, essentially a summary of what is exploitable and how badly they can be exploited might do it
It's been through quite a bit of mud so far i'd assume
i see
and you were able to verify em too ?
source code tools are known to provide a lot of false positives
hm, can't run your own local instance of it ?
They just have the potential to be an sql injection
I can
But it's one of those, if something goes wrong, this could have a much bigger impact because they didn't follow prepared statements standards
etc
right
I have no idea how to actually break it in that way
And ZAP didn't provide much more than a few xss and buffer overflows on the api
in that case, i'd stick to theory and go down the path of what sqli is capable of doing
which is dumping databases, OS command exec etc
and explain it in that manner
eh
he doesnt have to know how AES works in order to use it
surely you don't need access to the source to figure out the threat models?
and the wikipedia articles about the depths of AES are confusing to say the least
even the original paper is a better read
Just identifying weaknesses and where you're likely to be attacked, right?
and the potential from that
Yeah I've already dug a few holes for what poorly built sql queries can do
cool, yeah i think that is your best bet. explaining what they can do if they go unfixed
for example @thorn obsidian this is how we multiply when calculating AES stuff
pub fn multiplication(a: u8, b: u8, m: u16) -> u8 {
let mut res = 0;
let mut a:u16 = a as u16;
let mut b:u16 = b as u16;
let mask = !(((!0) >> 8) << 8);
for _ in 0..8 {
res ^= a * (b & 1);
let carry = (a >> (8 - 1)) & 1;
a = (a << 1) & mask;
a ^= m * carry;
b >>= 1;
}
return res as u8;
}
Based on how i figured a threat model is to be designed is as a look up for devs to make sure all vulnerable points are covered with the correct controls
and yea @lusty flare you dont need the source, but at least a high level of what the app does, workflows and architecture is needed
I mean for a threat model you could literally just point out where the important shit is and how it's exposed
and the most vulnerable point of attack
yea basically just point out where in the app it is exposed and how bad it can get
but then theres this whole thing The technical steps in threat modelling involve answering questions: - What are we working on - What can go wrong - What will we do with the findings - Did we do a good job?
yes but if you say he doesnt have to understand it in order to use the packages why post the link
what profit would he gain from that
actually AES itself doesnt give a fuck about IV
the IV is part of the modes of operation
not AES
that this does not work has nothing to do with the miss understanding of AES but with the miss conception that a web page from github gist contains the html
and not only their encrypted message
That is fine @silent pier , imo i'd just stick to what you found and provide an overall recommendation to the dev team or whomever is reading this threat model
i'd say with that you've done your due diligence and answered the questions as best as you could
Well shit, i should restart discord more often. 14 updates.
and it's still fecking laggy
lol thats why i stick to the web client
it will be if you import those sonarqube results ๐
Oh I have, they were all screenshots of my report
sweet
how would you secure a sqlite db
against what
@silent pier make a threat model
I did 
make one for sqlite dbs
No.
@silent pier well users not able to read the data
worst case situations
Do they have access to the db file directly, or indirectly?
Or are you thinking about querying for data
well i just simply want to extract the data inside the database safely
because the user can view the database
or sqlite
whatever u call it
I'm not quite following. You want to extract data from a db without the users seeing it?
well yes
but they can just simply open it and there u go
this is what i mean
like this
i dont want them to able see these things
these things as well
I guess my question stands, Can they access the file then?
yeah
You could encrypt it I suppose. I haven't done any database file security
mostly queries and avoiding sql injections
https://www.zetetic.net/sqlcipher/ for instance
SQLCipher is an Open Source SQLite extension that provides transparent 256-bit AES full database encryption.
i wish if i can install it
but it wont let me
sqlcipher
ive been trying for ages doesn work
did you clone the repo?
git clone https://github.com/sqlcipher/sqlcipher.git
i dont have git clone
git is fairly common, you should probably get it if you don't have it.
uh
how do i get it im searching rn
ok imma install and brb
ok i got git
now imma execute that command
well its finished but still says
ImportError: No module named pysqlcipher
@silent pier any other ideas?
I know just as much as you at this point
https://www.elttam.com.au/blog/ruby-deserialization/
Ruby 2.x Universal RCE Deserialization Gadget Chain
could anyone give me a quick rundown on basic string encryption that can also be decrypted (something like AES I think)? I'm a newbie at encryption and would like to learn how to do the basics ๐
Today weโre going to talk about how to keep information secret, and this isnโt a new goal. From as early as Julius Caesarโs Caesar cipher to Mary, Queen of S...
@buoyant maple
๐ thanks
also, what is the use in SHA one-way? seeing that you will never get data that you encrypted back, whats the point?
for things like storing user passwords, they're very useful. those hashing algorithms will always give the same hash for the same text (i.e "hello" would always hash to "2bd782f..."), so you may store the hash of a user's password rather than the password itself. that way, their password isn't stored plainly in the database.
>>> import hashlib
>>> # create a new password hash:
>>> pass_hash = hashlib.sha512(b"password123").hexdigest() >>> pass_hash
'bed4efa ... b6ac4bf' # spooky, it isn't plaintext now.
>>> # attempt to log in using a different password
>>> attempted_hash = hashlib.sha512(b"password321").hexdigest()
>>> pass_hash == attempted_hash
False
>>> # now log in with the correct password
>>> correct_attempt_hash = hashlib.sha512(b"password123").hexdigest()
>>> pass_hash == correct_attempt_hash
True
>>>```
@buoyant maple
Never store plaintext password 
aaah, that makes sense
(was going to ask a question but realized it was bad)
ยฏ_(ใ)_/ยฏ
hashing is not perfect on its own, but i must go. someone else may explain rainbow tables and salting, or you could research it yourself. :)
๐
i have heard of rainbow tables before, something like pre-made hashes?
ah ๐ (didnt see you where going)
Rainbow tables is a start hash, and an end hash with an algorithm to generate the next hash
or start password
I forgot how the start goes
Nontheless you hash the password, and use an algorithm to generate a new plaintext password from that hash then hash that password and reapeat
why dont we just convert all commonly used paswords into hashes
once you reach x amount of hashes you save the end hash and start password
and try them
Later to look up you run the same algorithm on a hash and if you get the end hash you know that password is in that chain of passwords you generated
Thats a dictionary attack
yeah
Benefits of rainbow is it takes more computational power but is considerably smaller in size
while dictionaries of passwords can get insanely large
Its easier to explain rainbow tables with graphics tbh
but you can personalize the dictionary depending on their character
i think thats how it goes atleast
Yeah, but you can do the same for a rainbow
๐
Thats dependant on the algo you use
I couldnt find any of the resources i used for it
but theres an article about them
one more question: will the SHA algorythm itself (not talking about how many bits it has or quantum computers) ever have a security weakness or is it something special about the way its made? I have seen a wikipedia article on the diffrent hash types and MD5 and SHA use the same "constructuion"
because them 1 time pads are said to be un-crackable with SHA/AES, that must mean that them algorythms are impenitrable to weaknesses
You can guess the plaintext and convert that to conpare with the hash
But I guess what they mean is that you cant get the plaintext from the hash
๐
not sure what freenode is but assuming its either a discord channel or 3rd party chat thing
IRC
there will always be some people still using it...
mm
I didn't mean to convey surprise
I know many use it
But I don't need to understand why it is still as popular as it is
anyway, yeah Owez, IRC is basically one of the first internet chat applications
it's been around since - i want to say 1988 (Freenode itself has existed in some form since 1995)
Been wrking on this for anybody who is new to pt or python it may be a nice project to check out
Nice ASCII art ๐
๐
@thorn obsidian thats weird ๐ค
@thorn obsidian its due to me having that as the hostname in /etc/hosts
Interface addresses should be in /sys/net/
@thorn obsidian @safe bear i fixed it anyways
s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
s.connect(("8.8.8.8", 80))
ip = s.getsockname()[0]
s.close()
print(banner)
print("")
print(" [i] Current Local IP: %s [i]" % ip)```
what if you dont have internet access macs
8.8.8.8 could be blocked too in some networks?
@chilly elk /sys/class/net/<iface-name>/address. That file will contain the IP address of the named interface (e.g. eth0 or enp0s1). However, I don't know if it will have multiple addresses if an interface has multiple addresses (which, yes, is a thing, and something I've encountered many times).
Also BTW, making a random network request buried in the code of a red teaming code is NOT good practice.
@safe bear fair point
In case anyone missed it
a Wordpress plugin for GDPR is vulnerable to some shit
The most casual use of the attack is to just rewrite the wp_option site_url to redirect to malicious shit
however combined with woocommerce and some other shit allows PHP shells
vulnerable to some "shit", you're hilarious @lusty flare
i mean if you want a full break down
i'm happy to tell you :P
basically the plugin allows someone to submit an ajax request or some bollocks that then allows them to leverage wordpress internals to create new user accounts with administrator privileges
from that point they can modify the wordpress config etc etc
I'd love to hear them.
in the cases i've had to clean up this week it's thankfully only been people modifying the siteurl option
which is like the "base url"
e.g. http://{{baseurl}}/whatever
so they changed that so it redirected to a load of dick pills and porn
bad look for a school
In that case, LiveOverflow URL video is good for this reference.
Redirected to spam?
yah
they could've done much worse
with WP admin creds they could've installed a PHP shell or some shit
pwn'd the container
That's suspicious.
Reminds me of MITM attack.
I understood it, quite a good discovery.
eh, yeah kind
it gets worse if the Wordpress site had the GDPR plugin and Woocommerce installed
woocommerce being one of the most common store fronts for Wordpress
could've managed to get some shit in wp-cron.php and created some kind of persistence
Patch released?
Well, there's a difference between there being a patch and the patch being applied
patch came out on 6th Nov, the sites I had to look at were compromised on the 8th Nov
First signs anyone actually did anything to the sites was on the 12th
heh
seems like they did the ajax leverage thing to create a pair of administrator accounts and then disabled the flaw
sneaky sneaky, close the hole after you get in
Didn't see that coming too.
I wonder if it's still vulnerable or not.
If I'd knew some exploit development I should've tried it.
Hey guys.
Do you consider putting secrets in your environmental variables/.env a bad practice?
yeah since any sub processes might end up seeing it
also putting them directly in the command line arguments is bad, since you can see it in the process information on some os
and ofc the shell history
It depends on the context
If you're running a service in a container, environment is acceptable practice AFAIK
For a multi-use server/host, though, it's definitely considered bad practice, since any arbitrary process could read them
ah i did not think about containers with just one thing in them
Take that with a heavy dosage of Salt, though, since I'm not very knowledgeable or experienced in web security.
personally i would still not do it just to avoid accedents and leaking it in to reporting software but its an interesting way to make it safer
Not really that new, but y'all heard about this last month right?
https://medium.com/@bertusk/cryptocurrency-clipboard-hijacker-discovered-in-pypi-repository-b66b8a534a8
@gentle heron I've done it in the past but I'm transitioning to aws ssm
I like the idea of having a declaration of secrets somewhere but not in anything that could get accidently pushed to github etc
I think what I'm going to do is use a env file in VSC for local testing with non sensitive configuration settings, and use aws ssm to get the secrets during runtime.
And add the .env to my .gitignore but if it leaks the worst it will do is leak my region and company name.
This is all new to me though so I just wanted to idiotcheck my methods
@wooden vigil Might know something about best practices
@velvet isle Had dengue so...
@thorn obsidian im finding another way
@upbeat palm Oh that's sad to hear man
@thorn obsidian I'm using serverless
I'm trying to follow 12 factor principles as well
Basically a backend app that automates some stuff between two third party SaaS APIs
With a front end API to allow us to call some of these custom functions too
I'm probably over engineering it, but I'm using it as a portfolio piece to try and break into software dev
So I'm trying to learn how to do everything properly with devops/agile principles
CI/CD, Web frameworks, on AWS infrastucture seems to be the in demand skills in my area.
If I go with VPS that's a whole load of things I have to worry about that I don't have to worry about with serverless/lamdba
True
True but I also learn alot about serverless, infrastructure as code, API gateway, secret manager, continuous deployment, automation of testing, qa, deployment ;)
Anyways I don't want to turn this into a #career-advice discussion :p
Ops roles really don't seem to have much demand or salary as dev, devops roles in my area
@velvet isle I'm good now, I guess.
What have you been learning since then?
@thorn obsidian serverless at our scale would be practically free
A couple of dollars a month at most.
Bypassing disabled exec functions in PHP via imap_open
#0day #exploit
Retweets
248
Likes
380
[04:58] Scott: So how much are you projecting to spend on AWS?
at the beginning of the month about $12bn
@upbeat palm Just reading random articles about how to grab this and how to break that
Anything specific?
Nah. Just tweets from people and inspecting repos
@thorn obsidian Check the pins for this channel
@safe bear Can you link me to good CCNA security course on YouTube just like you did last time for CCNA?
No sorry, I don't know any other than Shrike, which is in the channel pins
In addition to the end-to-end encryption that protects every Signal message, the Signal service is designed to minimize the data that is retained about Signal users. By design, it does not store a record of your contacts, social graph, conversation list, location, user avatar...
Signal is well known for its security.
gdude posted this in OT but I also feel it belongs in here too: https://nginxconfig.io/
OT?
Off topic
Oh.
@upbeat palm You know about i2p?
well
that and the fact that the key to decrypt the local database is stored in cleartext on your local machine
๐ค
Why do people care so much about this plaintext thing
Don't you store private keys in key files?
lol
back to making my web scraper
haha
@velvet isle The anonymous network software?
Features are good.
Glad I caught you while you're online
I tried doing some sslstrip on my phone today but apparently some sites have updated to not load without https
Like facebook
I tried burp to intercept the traffic
But ssl pinning on the phone is blocking me
Version of your android OS?
That's strange.
This is happening because of improper implementation?
Um
I don't know the design
I was at a cyber security event and I heard this guy had some people open their facebook app or something
and his face appeared on their screen like video chat
๐
Idk how
Dig deeper.
comments on what that guy did ?
Yup, you went to blackhat event?
Oh.
I live in south america
Cool.
What are some great tools you use ?
No...no, it has everything from web application pentesting resources to Cryptography.
Like everything you need to become a red teamer.
BTW checkout Empyre repository on GitHub, was referenced in Hacking:AoE.
I'm off, ciao.
Okay later bro
https://antichat.com/threads/463395/#post-4254681
Interesting read, but it is in Russian. Basically exploits a vulnerability in the library PHP uses for IMAP. Specifically, it is a vulnerability with connecting using rsh (he couldn't manage to connect with ssh for some reason). Funnily enough, it actually takes advantage of the -o option with ProxyCommand in ssh. Therefore, it only works on Debian-based distros because, for them, rsh is basically an alias for ssh. He bypasses command parsing with the use of $IFS$() for spaces and base64 for slashes.
He mentions that, for the rsh call, we know we at least have influence over the hostname. The hostname is taken from the imap_open call in PHP. This seems to be how the -o option gets passed to the rsh/ssh call.
๐
@thorn obsidian For one thing, with pgp private keys you can/should use a passphrase. For another thing, the target audience of Signal does not, in general, have the information security education to know they should be using FDE.
And lastly and most damningly, the fact that they bothered encrypting the database at all shows that the people behind Signal know full well it should be encrypted, and were relying on security through obscurity.
Damn
Its 12:42am
Wasn't expecting a ping lmao
Frida is something I heard about but didn't read up on
didn't truecrypt shut down a few years ago with a confusing message telling people to use bitlocker instead
anyway, signal is aimed at non technical users so i think they have a duty to be responsible about these things, and that includes having the option to encrypt the database with a password
i don't know if the mobile app does anything like that or not
How else would you suggest to encrypt the desktop?
Anything that encrypts and stores in userland, is vulnerable to something running in userland
the point is
storing the encryption key side by side with the encrypted data without relying on any user input to decrypt
is no better than no encryption at all
this isn't about a sophisticated attack involving something "running in userland" at the same time as it gets decrypted
It prevents your mother from reading your texts I guess
it's about the fact that someone (a thief, law enforcement, etc) can get your data just by taking your computer, even if it's powered off and nothing is running and you never use it after you know they have been to your house
That's why you use disk encryption
Some sort of password-based decrypt would be a excellent feature
security software is worthless if it's not secure unless you use other security software
But Signal's focus is a balance between usability to attract the common person and privacy
a password doesn't make things unusable
people use passwords all the time
maybe warn people in advance that if they forget their password there's no way to recover the data but
More than they have to do for any other messaging client
They enter their password once and forget about it
Just the whole local DB thing seemed overblown
Their focus is on preserving privacy in transit
Not on the endless battle that is client machine privacy
Maybe they add a password, they if the user doesn't have FDE they're safe
But keyloggers still get it
And it can be read from memory
Then you have to add protections against those
KeePass does this fairly well, but it's not also trying to secure end-to-end communications and server-side
Key wrapping?
does anyone have a good video series or anything that explains dnssec on a technical level, i.e. not a high level brief overview
i think im not wrapping my head around different resource records in DNS and makes it hard to understand DNSSEC
Norton Security?
In terms of password protected files, how does the software detect whether it was correct or not. Does it prepend the pre-encrypted file with some kind of expected value/string that basically serves as a canary to immediately (or sooner at least) signal whether the key was correct or not? I apologize if this belongs in help, I just figured I'd ask here since this isn't Python specific.
some programs store a hash of the key inside the file
and others use other stupid and insecure methods to store the key in themselves
and if you just decrypt a file which has been encrypted using some algorithm before you actually cant know wether you were right
you will have to apply algorithms to check wether what you got looks like a valid language or w/e or just random gibberish
https is not a software
it's a protocol
just because almost everyone does something doesn't mean it's perfect either
I'm not arguing that you can't store passwords on a user's device
but there are (at least slightly) better methods than plain text
both sides of the argument are valid
@orchid notch thanks for responding, I'm just reading this. Imagine this... having a text string encrypted with the same key as the file that's an expected value (such as the name of the product), and if the password input decrypts to that expected value, it can then signal a success on the client? Would that be insecure at all? This wasn't mentioned but of course the AES key would come from the string being ran through PBKDF2 or scrypt.
Putting this in Security too because I thought it was cute.
Steps:
- Tamper with device (rip cover off and potentially remove alarm signal wires)
- Tamper alarm goes off because you're tampering with it.
- Take the fuse out of the fuse block next to it
- Re-insert the fuse back into the fuse block
- Alarm resets into the untampered and untriggered state.
Now you know how it works?๐ค
Hmm?
Nevermind it.
Is using a sanbox for executing payloads on my host machine a recommended thing ?
What do you mean with sandbox? And where does the payloads come from?
Are you worried that these "payloads" can cause problems on the host?
Doesnโt seem like much of an answer
@errant pilot I got what I wanted to know tho
Lmao
@sullen hazel I was worried about what the payload can do to the host
I don't wanna set up a vm to test them
Unless thereโs more to the conversation then all you got was shit on and no actual information
ยฏ_(ใ)_/ยฏ
All in all
I wanna test the reverse tcp payloads on my win10 machine
But Im afraid to execute them and I don't wanna use a vm
So I thought using a sandbox may be more safe
btw these slides are worth checking out: https://itsecx.fhstp.ac.at/wp-content/uploads/2018/11/02_Rene_Freingruber_Flying_under_the_radar_freingruber_v1.00.pdf
Preview
@safe bear when memes get trending
@chilly elk are you macs from that repo ?
@upbeat palm Here's some gud stuff
Believe me, you won't get anything better than Red Team Field Manual, but it's a good list. @velvet isle
๐
i know haha
But hey stars are stars
@velvet isle appreciate the support btw

@velvet isle i made a base framework template for another member in here that is basically the basics of BabySploit, NaviSat, and DarkSpiritz if you wanna check it out
@chilly elk no read me yet 
yeah literally threw it up like 10 mins ago haha

Welp
I got exams coming up . Can't say much of what I expect to see
Since I won't have time to even think of something great lmao
good luck man i remember school blech
Does it also have 10 nested if statements? 
@safe bear was that for the nmap thing?
It was in dark dpiritz
yeah
I gave you shit about it
bad code lmao
Because it took years off my life
i use different methods now
YEARS
(terminal tables ๐ )
What are those?
