WEBVTT 1 00:00:00.080 --> 00:00:02.439 Welcome to the deep dive, where we plunge into dense 2 00:00:02.480 --> 00:00:05.919 information and pull out those surprising, impactful nuggets of knowledge, 3 00:00:06.240 --> 00:00:09.599 all crafted just for you. Today, we're navigating a world 4 00:00:09.679 --> 00:00:14.000 that's more interconnected than ever. Our digital lives, every transaction, 5 00:00:14.119 --> 00:00:17.440 every connection, every piece of personal information flows through this 6 00:00:17.600 --> 00:00:20.480 vast network, making the art of safe collecting it not 7 00:00:20.559 --> 00:00:23.679 just important but absolutely critical. 8 00:00:23.920 --> 00:00:27.199 And at the heart of that safeguarding, ironically often lies 9 00:00:27.199 --> 00:00:31.640 its weakest point, what we call authentication hacking or password cracking. 10 00:00:32.119 --> 00:00:35.280 It's a fascinating blend of art and science, like finding 11 00:00:35.280 --> 00:00:37.679 the digital skeleton key. Whether you're an attacker trying to 12 00:00:37.679 --> 00:00:40.719 get in or an ethical security tester trying to keep 13 00:00:40.759 --> 00:00:44.920 them out, understanding this process is fundamental. 14 00:00:44.560 --> 00:00:46.759 Right, and our mission today is to pull back the 15 00:00:46.759 --> 00:00:49.799 curtain on this complex world. We're going to unveil the 16 00:00:49.880 --> 00:00:53.320 layers of security that shield our digital identities, exploring the 17 00:00:53.439 --> 00:00:57.280 very arsenal of tools wielded by malicious actors, but crucially 18 00:00:57.520 --> 00:01:00.560 also by the pen testers and red teams who assess 19 00:01:00.560 --> 00:01:04.040 and strengthen digital fortifications. Our goal for you is to 20 00:01:04.120 --> 00:01:07.040 walk away with knowledge that empowers you to bolster your 21 00:01:07.079 --> 00:01:10.599 own defenses or to better understand vulnerabilities in your environment. 22 00:01:11.000 --> 00:01:13.840 And we're drawing our insights from an excellent source, Daniel 23 00:01:13.959 --> 00:01:17.959 WD Delee's book Password Cracking with Kylie Lennox twenty twenty three. 24 00:01:18.519 --> 00:01:21.280 Daniel brings a wealth of practical knowledge with over twenty 25 00:01:21.400 --> 00:01:25.079 years in it and more than a decade focused entirely 26 00:01:25.159 --> 00:01:26.120 on security research. 27 00:01:26.159 --> 00:01:27.760 It's a truly valuable perspective. 28 00:01:28.040 --> 00:01:30.239 So why does all this even matter to you personally 29 00:01:30.359 --> 00:01:33.680 or professionally? What's truly astounding? Is it? Despite billions spent 30 00:01:33.719 --> 00:01:38.200 on sophisticated cybersecurity, the weakest link often isn't cutting edge tech. 31 00:01:38.560 --> 00:01:42.640 It's something we've known for decades. The human element, our brains, 32 00:01:42.640 --> 00:01:44.879 it turns out, are often the real Achilles heel in 33 00:01:44.959 --> 00:01:45.760 digital defense. 34 00:01:46.079 --> 00:01:49.560 That's a profound point. The most common attack methods haven't 35 00:01:49.599 --> 00:01:53.079 really changed because people continue to make predictable password choices. 36 00:01:53.640 --> 00:01:57.480 For instance, root force attacks are automated tools that systematically 37 00:01:57.599 --> 00:02:01.680 guess combinations. This is how one, two, three, four, five, 38 00:02:01.799 --> 00:02:05.640 six or password get cracked in seconds. It's a low effort, 39 00:02:05.719 --> 00:02:06.280 high reward. 40 00:02:06.400 --> 00:02:08.360 So they're just throwing everything at the wall, like literally 41 00:02:08.400 --> 00:02:10.560 trying everything. What's a step beyond that? 42 00:02:10.719 --> 00:02:14.360 Well, yeah, essentially automated guessing, but a step up are 43 00:02:14.400 --> 00:02:18.039 dictionary attacks, which try commonly used words or phrases from 44 00:02:18.080 --> 00:02:20.840 pre compiled lists. If your password is Summer twenty twenty 45 00:02:20.840 --> 00:02:23.919 four or even something like Dragon Slayer ninety nine, chances 46 00:02:23.919 --> 00:02:25.360 are it's on one of those lists. 47 00:02:25.400 --> 00:02:28.599 Okay, And then there's something called credential stuffing that sounds 48 00:02:28.639 --> 00:02:30.680 particularly insidious. How does that work? 49 00:02:31.039 --> 00:02:31.400 It is? 50 00:02:31.680 --> 00:02:35.879 Credential stuffing takes leaked user name password pairs from one service, 51 00:02:36.120 --> 00:02:38.360 maybe a forum that had a data breach or an 52 00:02:38.360 --> 00:02:41.080 old gaming site you forgot about, and then attackers try 53 00:02:41.120 --> 00:02:44.039 to reuse those exact same combinations to access your accounts 54 00:02:44.039 --> 00:02:47.360 on other services, think like your banking or email. 55 00:02:47.680 --> 00:02:49.159 It totally exploits. 56 00:02:48.719 --> 00:02:51.439 The human tendency to reuse passwords across platforms. 57 00:02:51.639 --> 00:02:54.199 Wow. Yeah, that's a huge wake up call for unique 58 00:02:54.199 --> 00:02:57.000 passwords everywhere. Right. And of course there's phishing. We hear 59 00:02:57.000 --> 00:02:59.120 about it constantly, but how does it fit into the 60 00:02:59.120 --> 00:03:01.680 password cracking puzzle? Is it really cracking? 61 00:03:02.080 --> 00:03:05.159 Well, phishing is more of a social engineering attack. It's 62 00:03:05.199 --> 00:03:09.039 about deception right right, tricking individuals into revealing their passwords 63 00:03:09.080 --> 00:03:12.560 via fake emails or deceptive websites. While it's not directly 64 00:03:12.639 --> 00:03:16.719 a cracking method itself, it's a primary way attackers obtain credentials, 65 00:03:16.879 --> 00:03:19.759 often bypassing the need for any complex cracking at all. 66 00:03:20.319 --> 00:03:23.319 Users without you know, good security training, are far more 67 00:03:23.360 --> 00:03:25.599 susceptible to these elaborate scams. 68 00:03:25.680 --> 00:03:28.520 And when weak passwords are breached, what are the real 69 00:03:28.560 --> 00:03:32.319 world consequences beyond just a simple log in failure? What 70 00:03:32.439 --> 00:03:34.240 kind of damage are we actually talking about here? 71 00:03:34.319 --> 00:03:37.560 Oh, the risks are quite severe for individuals and organizations. 72 00:03:37.599 --> 00:03:42.560 Breaches often lead to significant financial loss and severe reputational damage. 73 00:03:42.919 --> 00:03:45.599 For you personally, it could mean identity taft, where cyber 74 00:03:45.639 --> 00:03:50.439 criminals impersonate you, leading to major financial and frankly emotional distress. 75 00:03:50.680 --> 00:03:53.080 And it doesn't stop there, does it. Once one account 76 00:03:53.120 --> 00:03:56.080 is compromised, what else can an attacker potentially do? 77 00:03:56.520 --> 00:03:57.039 Exactly? 78 00:03:57.360 --> 00:04:00.800 They can use your compromise account to send Mali emails, 79 00:04:01.240 --> 00:04:05.919 make unauthorized transactions, maybe even spread malware, effectively turning your 80 00:04:05.919 --> 00:04:10.360 digital presence into a weapon. For organizations, this means unauthorized 81 00:04:10.400 --> 00:04:16.120 access to critical systems, confidential data, proprietary information. The domino 82 00:04:16.160 --> 00:04:18.959 effect can be absolutely devastating. 83 00:04:19.439 --> 00:04:21.600 So the answer to this, at least in part is 84 00:04:21.639 --> 00:04:25.920 simply strong passwords. But what actually makes a password strong 85 00:04:26.160 --> 00:04:27.879 in this context? Is it just length? 86 00:04:28.120 --> 00:04:30.720 Length helps, But it's more than that. Passwords with a 87 00:04:30.759 --> 00:04:33.839 mix of upper and lowercase letters, numbers, and special characters 88 00:04:33.879 --> 00:04:37.639 significantly increase the time and effort attackers need. Unique passwords 89 00:04:37.639 --> 00:04:40.079 for each account are your best defense against that credential 90 00:04:40.079 --> 00:04:42.360 stuffing we mentioned. Think of it as having a different 91 00:04:42.439 --> 00:04:44.519 key for every door in your life. And you know, 92 00:04:44.560 --> 00:04:47.319 password managers are excellent tools to help you manage all 93 00:04:47.319 --> 00:04:50.199 these complex, unique passwords across all your services. 94 00:04:50.279 --> 00:04:51.160 Highly recommend them. 95 00:04:51.360 --> 00:04:55.000 It sounds like complexity and uniqueness really pay off. And 96 00:04:55.079 --> 00:04:59.439 how does that layered defense connect with, say, defending against 97 00:04:59.439 --> 00:05:02.439 those fishing attacks. Does a strong password help there too? 98 00:05:02.800 --> 00:05:03.519 Yeah? It does. 99 00:05:03.639 --> 00:05:06.279 Even if you accidentally click a malicious link in a 100 00:05:06.319 --> 00:05:10.319 phishing email, or robust unique password makes it much harder 101 00:05:10.360 --> 00:05:12.079 for attackers to actually gain entry. 102 00:05:12.120 --> 00:05:15.199 If they somehow get it, it buys you time to react. 103 00:05:15.639 --> 00:05:19.399 So yes, education, awareness and proactive password management are truly 104 00:05:19.480 --> 00:05:22.240 crucial for you in this ongoing battle against cyber threats. 105 00:05:22.439 --> 00:05:25.759 With that frightening why firmly in mind, let's pivot to 106 00:05:25.839 --> 00:05:28.399 the how how do these digital locks actually work. Let's 107 00:05:28.480 --> 00:05:30.439 peel back the layers a bit to understand the foundation 108 00:05:30.519 --> 00:05:32.160 of password security and computers. 109 00:05:32.399 --> 00:05:36.279 Okay, so, when you set a password, operating systems typically 110 00:05:36.319 --> 00:05:40.319 don't store the plain texts that would be terrible. Instead, 111 00:05:40.399 --> 00:05:43.519 they store it in an encrypted form called a password hash. 112 00:05:44.639 --> 00:05:47.519 Think of a hash as a unique, one way fingerprint 113 00:05:47.519 --> 00:05:50.560 of your password. You can create the fingerprint from the password, 114 00:05:50.560 --> 00:05:52.759 but you can't easily get the password back just from 115 00:05:52.800 --> 00:05:56.279 the fingerprint. That's the key during most security tests. Even 116 00:05:56.279 --> 00:05:59.279 if you recover user passwords, they'll be in this hash 117 00:05:59.319 --> 00:06:01.879 form need to be unencrypted or cracked. 118 00:06:02.240 --> 00:06:04.879 Though it's wild to think that some services still store 119 00:06:04.959 --> 00:06:08.560 or transmit passwords in plain text, making them incredibly vulnerable. 120 00:06:09.079 --> 00:06:12.519 But mostly we're talking hashes now. In the Windows world, 121 00:06:12.519 --> 00:06:14.959 there are a couple of key authentication protocols at play 122 00:06:15.040 --> 00:06:17.839 right like NTLM and Carberos precisely. 123 00:06:18.160 --> 00:06:21.160 First, there's NTLM or NTI and land Manager. It's kind 124 00:06:21.160 --> 00:06:24.439 of a legacy protocol. It uses a challenge response mechanism. 125 00:06:24.800 --> 00:06:26.759 While it's been around since the early days of Windows, 126 00:06:26.800 --> 00:06:29.680 it's known to be vulnerable to attacks like past the hash, 127 00:06:29.879 --> 00:06:32.600 where attackers can use the hash directly without ever needing 128 00:06:32.639 --> 00:06:36.399 the plain text password scary stuff. It's gradually being replaced 129 00:06:36.399 --> 00:06:38.519 by more robust methods, thankfully. 130 00:06:38.279 --> 00:06:41.319 And the more secure option is Carbero's. What's the fundamental 131 00:06:41.360 --> 00:06:42.720 difference there? How does that work? 132 00:06:42.879 --> 00:06:46.399 Carberis is a ticket based system. It's designed to be 133 00:06:46.439 --> 00:06:49.879 more secure, using encrypted tickets to grant access, kind of 134 00:06:49.879 --> 00:06:51.560 like temporary passes at an event. 135 00:06:52.120 --> 00:06:53.199 But what's fascinating is. 136 00:06:53.199 --> 00:06:56.439 That even though Carberra's is inherently more secure, many companies 137 00:06:56.480 --> 00:07:01.720 still use NTLM, sometimes alongside Carberos. And regardless, both Carbero's 138 00:07:01.759 --> 00:07:05.639 tickets and NTL and password hashes are frequently targeted by attackers. 139 00:07:06.199 --> 00:07:09.800 While their authentication protocols differ, the attack process in essence 140 00:07:09.839 --> 00:07:13.240 is quite similar. Attackers obtain this encrypted password information the 141 00:07:13.279 --> 00:07:15.879 hash or ticket, and then try to crack it offline. 142 00:07:15.959 --> 00:07:18.879 That makes perfect sense, obtain the encrypted thing, then crack it. 143 00:07:19.360 --> 00:07:22.279 And with that understanding of hashes and protocols, we can 144 00:07:22.319 --> 00:07:26.120 now zero win on a particularly clever attack that directly 145 00:07:26.160 --> 00:07:30.480 exploits Curbero's, something called kurber roasting. What exactly is happening 146 00:07:30.480 --> 00:07:31.519 in this attack right? 147 00:07:31.600 --> 00:07:36.160 Curk roasting it specifically targets the curbero's authentication protocol. Usually 148 00:07:36.160 --> 00:07:39.959 within active directory environments, that's the Windows Network management System. 149 00:07:40.560 --> 00:07:43.399 The true cutting of kerber roasting isn't just in cracking 150 00:07:43.399 --> 00:07:47.920 a password. It's how attackers leverage legitimate system functionalities, the 151 00:07:48.040 --> 00:07:50.920 very way active directory is designed to work. To turn 152 00:07:50.959 --> 00:07:56.160 a feature into a pretty formidable vulnerability, an attacker captures 153 00:07:56.240 --> 00:08:00.000 encrypted service tickets, often specifically targeting service accounts, and then 154 00:08:00.000 --> 00:08:02.240 and attempts to crack them offline to reveal the plain 155 00:08:02.360 --> 00:08:03.319 text passwords. 156 00:08:03.399 --> 00:08:05.720 Okay, so service tickets are key here. How do they 157 00:08:05.800 --> 00:08:08.360 fit into the carberra's structure. You mentioned tickets before. 158 00:08:08.480 --> 00:08:12.160 Let's quickly break down the relevant corberra's components. First, there's 159 00:08:12.240 --> 00:08:15.000 the ticket granting ticket or TGT. You get this when 160 00:08:15.040 --> 00:08:16.959 you first log into a window system. It's like your 161 00:08:16.959 --> 00:08:20.480 main entry pass. Then, using that TGT, you can request 162 00:08:20.519 --> 00:08:24.519 service tickets for specific network resources like file servers or 163 00:08:24.560 --> 00:08:29.839 web services. These tickets are temporary passes for specific things and. 164 00:08:29.800 --> 00:08:33.519 The real goldmine for attackers. You mentioned are these service accounts? 165 00:08:34.039 --> 00:08:37.360 What makes them such high value targets compared to say, 166 00:08:37.679 --> 00:08:39.039 a regular user account. 167 00:08:39.159 --> 00:08:42.840 Good question. These accounts are often associated with background services, 168 00:08:42.879 --> 00:08:46.360 not actual human users. They sometimes have highly privileged access, 169 00:08:46.440 --> 00:08:49.840 maybe even domain admin rights, which is like having keys 170 00:08:49.840 --> 00:08:53.360 to the whole kingdom. Crucially, their passwords often aren't changed 171 00:08:53.440 --> 00:08:56.000 very often, maybe because nobody thinks about them, and sometimes 172 00:08:56.000 --> 00:08:59.159 they're even set to the minimum allowed domain password length, 173 00:08:59.399 --> 00:09:02.080 making them we When a service ticket is issued for 174 00:09:02.120 --> 00:09:05.399 one of these accounts, it's encrypted using that service account's 175 00:09:05.519 --> 00:09:08.840 secret key, which is derived from its password. That's why 176 00:09:08.879 --> 00:09:11.799 capturing and cracking these tickets is so appealing to attackers. 177 00:09:11.840 --> 00:09:15.960 Okay, so they target these potentially powerful, potentially neglected accounts. 178 00:09:16.200 --> 00:09:20.000 How do attackers actually pull this off in practice? What's 179 00:09:20.000 --> 00:09:20.879 the typical flow? 180 00:09:21.200 --> 00:09:24.519 Well, they usually start by identifying and enumerting those service 181 00:09:24.519 --> 00:09:28.679 accounts in their associated service principal names or SPNs. These 182 00:09:28.799 --> 00:09:31.960 SPNs are unique identifiers that Kuberro's uses to find a 183 00:09:31.960 --> 00:09:34.720 specific service on the network. Think of them like a 184 00:09:34.720 --> 00:09:38.360 mailing address for a service. Then, using a TGT they've 185 00:09:38.399 --> 00:09:42.159 already pained somehow, they request and capture the service tickets 186 00:09:42.279 --> 00:09:43.320 for these SPNs. 187 00:09:43.480 --> 00:09:45.840 Once they've captured the ticket, what's the next step is 188 00:09:45.840 --> 00:09:46.639 the hard part. 189 00:09:46.440 --> 00:09:49.919 Over not quite the real work for the attacker begins. 190 00:09:49.960 --> 00:09:54.519 Then cracking the captured ticket offline. This usually involves brute 191 00:09:54.559 --> 00:09:58.080 force or dictionary attacks using tools like hashcat or John 192 00:09:58.080 --> 00:09:58.840 the Ripper. 193 00:09:58.559 --> 00:09:59.639 Which we'll talk more about. 194 00:10:00.039 --> 00:10:02.600 They hammer away at the encrypted ticket until they hopefully 195 00:10:02.600 --> 00:10:06.600 reveal the plaintext password. Sometimes they even employ pass the 196 00:10:06.679 --> 00:10:09.840 ticket attacks, where they use the captured ticket directly for 197 00:10:09.960 --> 00:10:12.799 unauthorized access without needing to crack the password at all. 198 00:10:13.080 --> 00:10:15.279 They just present the valid ticket and get in. 199 00:10:15.399 --> 00:10:18.519 Wow, it's pretty incredible how specialized the tools are for 200 00:10:18.559 --> 00:10:21.080 this specific attack. What are some of the go to 201 00:10:21.080 --> 00:10:23.519 tools for kerberosting that people mine encounter. 202 00:10:23.840 --> 00:10:27.519 You're right, it's a dedicated toolkit. Rubius, for instance, is 203 00:10:27.559 --> 00:10:32.440 a powerful C sharp based post exploitation tool specifically for 204 00:10:32.559 --> 00:10:36.799 interacting with and attacking creberos. It can request tgts and 205 00:10:36.879 --> 00:10:40.440 service tickets and even automatically perform a kerberos attack. It's 206 00:10:40.480 --> 00:10:42.679 a favorite for its versatility, definitely. 207 00:10:42.799 --> 00:10:45.360 And then there's the infamous mimicats. We hear that name 208 00:10:45.399 --> 00:10:46.840 a lot. What's its role here? 209 00:10:47.240 --> 00:10:50.600 Mimicats is notoriously well known for extracting all sorts of 210 00:10:50.600 --> 00:10:54.879 credentials from memory, including cerbero's tickets and NTLM hashes. It's 211 00:10:54.879 --> 00:10:57.240 also used to facilitate those pass the ticket attacks we 212 00:10:57.399 --> 00:11:00.480 just discussed. Beyond that, the Kerberos toolkit is a set 213 00:11:00.519 --> 00:11:04.799 of tools designed specifically to extract ESPNS request tickets and 214 00:11:04.840 --> 00:11:07.600 crack them using a component called tgs. 215 00:11:07.159 --> 00:11:10.000 Rep crack, and for the actual heavy lifting of cracking 216 00:11:10.039 --> 00:11:12.519 the ticket itself. I'm guessing we're talking about the big 217 00:11:12.600 --> 00:11:14.360 names like hashcat and John. 218 00:11:14.120 --> 00:11:17.600 The Ripper precisely, they are the workhorses for the offline 219 00:11:17.639 --> 00:11:21.960 decryption part. And finally, there's Bloodhound. Bloodhound is an active 220 00:11:21.960 --> 00:11:26.639 directory analysis tool that helps security professionals and unfortunately attackers, 221 00:11:26.679 --> 00:11:30.720 identify those kerbero ostable accounts. It maps out relationships and 222 00:11:30.759 --> 00:11:34.200 potential attack paths within the network, aiding in the planning 223 00:11:34.240 --> 00:11:36.039 phase before an attack even begins. 224 00:11:36.320 --> 00:11:39.200 So what does this all mean for you listening? These 225 00:11:39.200 --> 00:11:42.879 tools are indeed used by malicious attackers to compromise security, 226 00:11:42.919 --> 00:11:45.519 that's the scary part. But it's crucial to understand that 227 00:11:45.519 --> 00:11:48.440 security professionals also use these very same tools for ethical 228 00:11:48.480 --> 00:11:52.120 hacking or penetration testing. They employ them to identify and 229 00:11:52.159 --> 00:11:56.120 address vulnerabilities before the bad actors can exploit them. It's 230 00:11:56.159 --> 00:11:57.279 a constant arms race. 231 00:11:57.320 --> 00:11:58.960 Really, that's a great way to put it, and at 232 00:11:58.960 --> 00:12:01.360 the foundation of almost all all password cracking, whether it's 233 00:12:01.480 --> 00:12:05.559 NTLM hashes or Corbero's tickets, is one critical component the 234 00:12:05.639 --> 00:12:08.600 word list. These are the well the foundation to password tracking. 235 00:12:08.840 --> 00:12:12.519 They're essentially text files filled with potential passwords that cracking 236 00:12:12.559 --> 00:12:15.519 programs use to compare against the target hashes or tickets. 237 00:12:15.720 --> 00:12:19.000 So it's not just random guessing. Then you're essentially comparing 238 00:12:19.000 --> 00:12:23.279 your target against a known or generated list. How do 239 00:12:23.320 --> 00:12:26.080 these cracking programs actually use these lists? Do they just 240 00:12:26.200 --> 00:12:27.039 try each word? 241 00:12:27.720 --> 00:12:31.159 They work by either taking words directly from the list, 242 00:12:31.320 --> 00:12:34.240 hashing them using the right algorithm, and comparing them to 243 00:12:34.240 --> 00:12:37.080 the target hash. That's a simple way, or more advanced 244 00:12:37.120 --> 00:12:40.320 programs will manipulate these words using rules. Think of it 245 00:12:40.399 --> 00:12:45.159 like automatically adding prefixes, suffixes, numbers, dates, or even converting 246 00:12:45.200 --> 00:12:48.759 words to leats speak, you know, replacing an e with 247 00:12:48.799 --> 00:12:51.159 a three and oh a zero to create tons of 248 00:12:51.240 --> 00:12:52.440 new combinations to try. 249 00:12:52.519 --> 00:12:54.720 It Really all comes back to human behavior, doesn't it. 250 00:12:55.120 --> 00:12:59.360 Our habits, our predictability are the vulnerability these lists exploit. 251 00:12:59.480 --> 00:13:02.879 They absolute people are creatures of habit and patterns. When 252 00:13:02.879 --> 00:13:06.080 it comes to passwords, we often use names, important dates, 253 00:13:06.120 --> 00:13:10.000 maybe pet names, numbers. A very common pattern scene is 254 00:13:10.159 --> 00:13:12.919 passwords starting with a capital letter and ending with the symbol, 255 00:13:13.360 --> 00:13:15.559 or maybe incorporating a year like the current one. 256 00:13:15.799 --> 00:13:17.200 It's often quite predictable. 257 00:13:17.519 --> 00:13:20.080 Where do these massive word lists even come from. Do 258 00:13:20.159 --> 00:13:21.519 people just sit and type them out? 259 00:13:21.759 --> 00:13:22.080 Huh? 260 00:13:22.120 --> 00:13:22.200 No? 261 00:13:22.360 --> 00:13:22.440 Not. 262 00:13:22.559 --> 00:13:26.279 Usually they come from various sources. Publicly leaked password dumps 263 00:13:26.279 --> 00:13:29.559 from past data breaches are a huge source combinations of 264 00:13:29.600 --> 00:13:33.679 different leaks. You can also find foreign language word lists 265 00:13:33.919 --> 00:13:37.840 or even full dictionary and encyclopedia dumps. And ethical pen 266 00:13:37.879 --> 00:13:41.360 testers will also create custom word lists using company specific 267 00:13:41.440 --> 00:13:45.919 data like employee names, local sports teams, phone numbers, email addresses. 268 00:13:45.960 --> 00:13:49.720 Knowing people often incorporate these personal or local details into 269 00:13:49.759 --> 00:13:52.600 the passwords. It's about thinking like the user you're testing. 270 00:13:52.879 --> 00:13:56.399 I know Callie. Linux, the security focused operating system we mentioned, 271 00:13:56.720 --> 00:13:59.440 actually includes several word lists right out of the box. 272 00:13:59.480 --> 00:14:01.200 Which ones are commonly used or known? 273 00:14:01.639 --> 00:14:04.039 Yeah, Collee comes packed with useful stuff. One of the 274 00:14:04.120 --> 00:14:06.720 most popular is the Rocky word list, which is just 275 00:14:06.799 --> 00:14:09.840 a massive collection of millions of actual passwords recovered from 276 00:14:09.879 --> 00:14:13.120 a specific database dump years ago. It's a snapshot of 277 00:14:13.240 --> 00:14:16.360 real compromise passwords. There's also a word list included with 278 00:14:16.399 --> 00:14:20.120 John the Repper itself, and things like wfuzz wordlists, which 279 00:14:20.120 --> 00:14:22.879 are more useful for web stuff like finding hidden directories 280 00:14:22.960 --> 00:14:23.440 or files. 281 00:14:23.720 --> 00:14:26.279 Beyond these pre made lists, I hear there are powerful 282 00:14:26.320 --> 00:14:29.360 tools that let you generate custom word lists on the fly. 283 00:14:30.120 --> 00:14:32.000 Tell me about CWL for example. 284 00:14:32.080 --> 00:14:35.879 Right. CWL Custom Wordless Generator is pretty neat. It crawls 285 00:14:35.879 --> 00:14:40.200 target websites to build custom theme based word lists. It 286 00:14:40.320 --> 00:14:43.039 pulls words related to a company or its industry. So 287 00:14:43.039 --> 00:14:47.679 if you're targeting say Acmecorp, it might scrape their website 288 00:14:47.679 --> 00:14:52.639 for terms like widget innovation, maybe product names, executive names. 289 00:14:52.360 --> 00:14:54.679 Stuff like that. It makes the list more relevant. 290 00:14:54.759 --> 00:14:58.279 That's clever tailoring the attack. And then there's Crunch, which 291 00:14:58.320 --> 00:15:00.840 sounds like it can build lists completely from the ground up. 292 00:15:01.120 --> 00:15:04.080 Crunch is fantastic for that. It lets you create custom 293 00:15:04.120 --> 00:15:07.399 password lists totally from scratch. You get precise control over 294 00:15:07.399 --> 00:15:10.159 the length of complexity of the character sets used. It 295 00:15:10.200 --> 00:15:13.759 can build simple sequential permutations like going from a tog's 296 00:15:14.399 --> 00:15:18.159 or much more complex alphanumeric or even Unicode combinations by 297 00:15:18.240 --> 00:15:21.120 using its charset dot LST files. It's for when you 298 00:15:21.120 --> 00:15:23.399 have a very specific pattern or guests in mind that's 299 00:15:23.440 --> 00:15:24.559 not in a standard list. 300 00:15:24.840 --> 00:15:26.600 And here's a surprising fact that kind of blew my 301 00:15:26.639 --> 00:15:30.039 mind when I read it. Hashcat itself, the cracking engine 302 00:15:30.039 --> 00:15:33.360 we'll talk more about soon, can actually create word lists. 303 00:15:34.000 --> 00:15:34.679 How does that work? 304 00:15:34.840 --> 00:15:38.519 Yes, again, it's a bit counterintuitive, but you use its 305 00:15:38.559 --> 00:15:42.240 normal attack commands with a special switch start out. This 306 00:15:42.399 --> 00:15:46.080 essentially tells hashcat to output the passwords it's generating based 307 00:15:46.120 --> 00:15:48.879 on rules or masks directly to the screen or a 308 00:15:48.879 --> 00:15:51.759 file as a word list, instead of comparing them to 309 00:15:51.799 --> 00:15:55.120 a hash. It's a really versatile feature for generating candidates. 310 00:15:55.240 --> 00:15:57.960 And if you really want to get granular with combining lists, 311 00:15:58.279 --> 00:16:02.039 there are hashcat utilities like Common and Combinator three. What 312 00:16:02.080 --> 00:16:03.320 do those do? Specifically? 313 00:16:03.519 --> 00:16:07.080 These tools are all about combining existing word lists. For instance, 314 00:16:07.159 --> 00:16:09.759 combinator might take red from one list and bike from 315 00:16:09.759 --> 00:16:13.240 another to create red Bike. Combinator three is even more complex. 316 00:16:13.600 --> 00:16:15.960 Just be warned, and this is a serious warning. They 317 00:16:16.000 --> 00:16:19.840 can generate massive output files, absolutely huge. They will quickly 318 00:16:19.840 --> 00:16:22.559 fill up even a large hard drive if you're not careful. 319 00:16:22.840 --> 00:16:26.840 We're talking potentially gigabytes, even terabytes of potential passwords. 320 00:16:27.080 --> 00:16:29.919 Okay, noted, don't accidentally fill your hard drive trying to 321 00:16:29.919 --> 00:16:32.399 make a word list. And then there's a niche but 322 00:16:32.799 --> 00:16:37.399 surprisingly effective technique called keymp walking implemented by a tool 323 00:16:37.440 --> 00:16:40.960 called kW processor. Can you explain that when it sounds unusual? 324 00:16:41.120 --> 00:16:44.039 Yeah, it is a bit unusual, but based on real behavior. 325 00:16:44.200 --> 00:16:47.919 It creates passwords by literally walking across a keyboard layout. 326 00:16:48.159 --> 00:16:50.519 Imagine starting at the zec, then moving up and write 327 00:16:50.559 --> 00:16:53.480 to a Q one, maybe creating a password like zac one. 328 00:16:53.960 --> 00:16:56.879 It's based on the idea that people might unconsciously create 329 00:16:56.919 --> 00:17:01.000 simple patterns on their keyboard when making passwords. Aw processor 330 00:17:01.080 --> 00:17:03.879 generates word lists based on these physical patterns, even for 331 00:17:03.879 --> 00:17:05.440 foreign language keyboard layouts. 332 00:17:05.559 --> 00:17:07.759 What stands out to me here listening to all this 333 00:17:07.839 --> 00:17:10.839 is the sheer variety and specificity of how word lists 334 00:17:10.880 --> 00:17:14.440 can be generated. It really underscores how deeply attackers and 335 00:17:14.480 --> 00:17:18.440 defenders have studied human habits and patterns in password creation. 336 00:17:19.160 --> 00:17:21.839 It's a constant cat and mouse game rooted in like 337 00:17:22.279 --> 00:17:23.240 human psychology. 338 00:17:23.519 --> 00:17:26.400 That's spot on, it really is. So once you have 339 00:17:26.480 --> 00:17:29.720 your potential weapon, your word list, the next crucial step 340 00:17:29.759 --> 00:17:33.359 in password cracking is identifying the hash type you're up against. 341 00:17:33.839 --> 00:17:36.400 You absolutely need to know the hash type so you 342 00:17:36.440 --> 00:17:40.400 can tell your cracking program what decryption algorithm to use. 343 00:17:40.880 --> 00:17:42.960 It's like knowing what kind of lock you're trying to pick. 344 00:17:43.079 --> 00:17:45.559 You need the right tool, the right technique. 345 00:17:45.680 --> 00:17:48.039 So there are different types of hashes, not just one. 346 00:17:47.920 --> 00:17:51.680 Hash, absolutely loads of them. There's the old ELM hash, 347 00:17:51.759 --> 00:17:55.640 which is thankfully outdated and very insecure, easily cracked, not 348 00:17:55.759 --> 00:17:59.519 widely used anymore. Then the NTLM hash, which combines the 349 00:17:59.640 --> 00:18:02.720 LM hash and nt hash, commonly found in Windows SAM 350 00:18:02.799 --> 00:18:04.720 databases or domain controller databases. 351 00:18:04.720 --> 00:18:05.440 That's a big one. 352 00:18:05.519 --> 00:18:08.720 You also have NTLMB one and NTLMB two, which are 353 00:18:08.680 --> 00:18:12.279 a challenge response hashes often captured in network really attacks. 354 00:18:12.559 --> 00:18:13.079 They're different. 355 00:18:13.119 --> 00:18:13.319 Again. 356 00:18:13.880 --> 00:18:17.000 Luckily, there are hash identification tools like hash identifier and 357 00:18:17.039 --> 00:18:19.920 hash id and calie Linux that help determine the hashtype 358 00:18:19.960 --> 00:18:22.680 from a sample, which is critical before you even begin 359 00:18:22.720 --> 00:18:23.440 trying to crack it. 360 00:18:23.720 --> 00:18:27.480 And for cracking maybe simpler passwords, especially those older LM 361 00:18:27.599 --> 00:18:30.440 or NTLM hashes, you might not even need a powerful 362 00:18:30.440 --> 00:18:33.519 local tool. Right, I've heard of online crackers. Yeah, how 363 00:18:33.559 --> 00:18:36.559 do those work? And importantly are they safe to use? 364 00:18:36.759 --> 00:18:37.000 Yeah? 365 00:18:37.039 --> 00:18:40.359 There are online crackers like crackstation. They use massive pre 366 00:18:40.440 --> 00:18:44.720 computed lookup tables often called rainbow tables, to return passwords 367 00:18:44.799 --> 00:18:48.000 in mere seconds for common hashes. If a hash for 368 00:18:48.079 --> 00:18:50.839 password or one, two, three, four, five six is uploaded, 369 00:18:50.920 --> 00:18:54.000 it's cracked almost instantly because those hashes have been seen 370 00:18:54.039 --> 00:18:57.319 and cracked millions of times before. But and this is 371 00:18:57.359 --> 00:19:01.680 a strong word of caution, be extremely Some online crackers 372 00:19:01.759 --> 00:19:04.839 might run crypto miners like bitcoin miners in your browser 373 00:19:04.920 --> 00:19:07.920 as payment for their service, often without telling you. Plus, 374 00:19:07.920 --> 00:19:11.720 you're uploading potentially sensitive hash data to an unknown third party. 375 00:19:12.079 --> 00:19:15.880 For any serious secure cracking, especially in a professional context, 376 00:19:15.920 --> 00:19:18.319 you'll definitely want to turn to offline tools you control. 377 00:19:18.319 --> 00:19:20.880 Okay, good advice, avoid the sketchy online ones. So when 378 00:19:20.920 --> 00:19:23.000 we do need a local, powerful tool, where does someone 379 00:19:23.039 --> 00:19:25.559 even begin? Is there a classic first stop cracker For 380 00:19:25.640 --> 00:19:26.759 maybe easier targets. 381 00:19:27.000 --> 00:19:29.799 Absolutely, for the low hanging fruit as we call it, 382 00:19:30.039 --> 00:19:33.359 or maybe shorter passwords, you'd often turn first to John 383 00:19:33.400 --> 00:19:36.640 the Ripper or just JTR. It's a fast, open source, 384 00:19:36.880 --> 00:19:42.319 primarily CPU based password cracker. It's incredibly versatile, supports hundreds 385 00:19:42.319 --> 00:19:45.400 of hash types, performs dictionary and hybrid attacks, and it 386 00:19:45.480 --> 00:19:50.000 runs on multiple platforms Linux, Windows, Mac. You simply pointed 387 00:19:50.039 --> 00:19:52.720 out a file containing your hashes on password hash list 388 00:19:52.720 --> 00:19:55.240 and it goes to work. It even cleverly stores the 389 00:19:55.279 --> 00:19:57.839 cracked passwords it finds in its pot file, so you 390 00:19:57.839 --> 00:19:59.599 don't lose them and don't have to recrack them. 391 00:20:00.039 --> 00:20:02.039 Undroop is good for a quick win. Maybe it hits 392 00:20:02.039 --> 00:20:05.160 the easy stuff first, But when you need serious speed 393 00:20:05.240 --> 00:20:08.799 for those really tough, long, complex hashes, what's the go to? 394 00:20:08.880 --> 00:20:09.759 What's the big gun. 395 00:20:10.119 --> 00:20:13.599 When you need serious computational muscle, you turn to hashcat. 396 00:20:14.079 --> 00:20:17.119 This is generally considered the king. It's an all purpose 397 00:20:17.359 --> 00:20:21.000 GPU based cracker. It uses your graphics card, though it 398 00:20:21.039 --> 00:20:24.119 can use your CPU if needed. It's widely touted as 399 00:20:24.160 --> 00:20:27.720 the world's fastest and most advanced password cracker. We're talking 400 00:20:27.759 --> 00:20:31.680 potentially with high end hardware billions, even trillions of hashes 401 00:20:31.720 --> 00:20:35.559