<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Nasrin on Medium]]></title>
        <description><![CDATA[Stories by Nasrin on Medium]]></description>
        <link>https://medium.com/@ghassemiN?source=rss-15177f890ebb------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*hTsSUY5WVBQvnvN9MVrGxQ.jpeg</url>
            <title>Stories by Nasrin on Medium</title>
            <link>https://medium.com/@ghassemiN?source=rss-15177f890ebb------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 22:21:54 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@ghassemiN/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Human Presence in the Age of Machine-Written Replies]]></title>
            <link>https://medium.com/@ghassemiN/human-presence-in-the-age-of-machine-written-replies-62cb6cc80168?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/62cb6cc80168</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[ai-or-human]]></category>
            <category><![CDATA[machine-reply]]></category>
            <category><![CDATA[ai-memory]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Wed, 28 Oct 2026 07:56:34 GMT</pubDate>
            <atom:updated>2026-10-28T07:56:34.909Z</atom:updated>
            <content:encoded><![CDATA[<p>Who is answering?</p><p>We talk about AI a great deal these days: critical vulnerabilities, the future of certain jobs, deepfakes and misinformation, over-reliance on AI, the strange influence of algorithms, and what all of this means for society, the economy, democracy and politics.</p><p>There is another question I think deserves attention: what happens to our human relationships when AI enters them?</p><h3>A reply that doesn’t sound like your friend</h3><p>Imagine you have been friends with someone for years. You know each other well, you follow each other’s lives, and you share a history. One day you message them about something personal and get back a reply you didn’t expect. It may be perfectly reasonable, even well written, but something about it feels off. It reads as if whoever wrote it doesn’t really know you.</p><p>There are plenty of innocent explanations. Your friend may have been distracted, may have misread your message, or may have forgotten the story you were referring to. But today there is another possibility: they may not have written it at all. An AI might have.</p><p>That raises a question I find worth asking: when we talk to a friend, are we entitled to expect that the person on the other end is actually <em>them</em>?</p><p>I am not arguing that using AI is wrong. I use it myself, and I enjoy it. I built a personal tutor for myself, inspired by Asimov, and I use AI responsibly in my work, while keeping up with its risks. But there is a real difference between asking AI to polish your writing and handing it your entire conversation with a friend.</p><h3>Trust, and who is really answering</h3><p>I won’t try to define what makes someone a friend. Whatever the definition, a friend is someone we trust, and trust makes us vulnerable. We open up, we are influenced, we take risks.</p><p>Now suppose your friend uses AI to answer your messages. This is not far-fetched. Today’s AI agents have memory, and someone could feed them everything they know about you so the replies feel personal. The answers might even be better: more thoughtful, more helpful, more balanced than what your friend would have written alone. I don’t doubt AI’s ability to write them.</p><p>But is it right? I may be speaking openly under the assumption that I’m talking to this person, when in fact a machine is doing a large part of the talking. And if my friend never told me, does that matter? I think it does, because in friendship we expect to be heard and understood by the other person, and to offer the same in return.</p><h3>Does being heard still count if a machine did it?</h3><p>Consider a bad day. You tell your friend what happened and how you feel. A simple reply like <em>“I know exactly what you mean, remember when something like this happened a couple of months ago?”</em> can mean a lot. The feeling of being understood is what matters. The depth of a friendship has less to do with how much information two people have about each other and more to do with the feeling that passes between them. After all, your doctor probably knows more about you than many of your friends do.</p><p>So if a machine can produce a reply that makes me feel heard, is that enough? What if my friend reads my message and lets AI write the response? What if they don’t even read it, and have no idea what passed between me and the AI? And does it change anything for me whether I know that a warm reply was machine-written or not?</p><h3>Privacy</h3><p>There is a further problem. Suppose I share a personal memory with a friend who has connected their messenger to an AI service. My message goes to that service, which sends back a reply. I trusted my friend, but my words ended up somewhere I knew nothing about. The service might run locally on my friend’s machine, or in the cloud, where my messages may be analyzed and possibly used for training, without my knowledge.</p><p>That is not so different from telling a friend a secret and having them pass it on to someone else without telling me.</p><h3>What we actually want from a friend</h3><p>What are we looking for when we talk to a friend? Correct answers? Those are available everywhere now, instantly and often more accurately than a friend could give them. So it isn’t about the answer. Is it about the time someone gives us? An AI has no time to give, while a person does. What we want is a human presence, with understanding and feeling that run in both directions. Without that, the conversation loses its value.</p><p>I hope we never end up in friendships like that, and that we let our friendships stay human- It would be wrong to conclude that any AI involvement in personal writing is harmful. Fixing grammar, translating for a non-native speaker, or helping someone with social anxiety or a disability find words are different from delegating a relationship.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=62cb6cc80168" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI Is Making Security Negligence Harder to Hide]]></title>
            <link>https://systemweakness.com/ai-is-making-security-negligence-harder-to-hide-d3e9a0072581?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/d3e9a0072581</guid>
            <category><![CDATA[data-privacy]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[vulnerability]]></category>
            <category><![CDATA[security]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Mon, 10 Aug 2026 15:59:48 GMT</pubDate>
            <atom:updated>2026-08-11T11:17:33.324Z</atom:updated>
            <content:encoded><![CDATA[<p>A few years ago- 2023-, I was testing an application whose name I’d rather not mention. During that process, I discovered that I could bypass its two-factor authentication on Android.</p><p>The issue was serious enough that I tested several other applications, including other related banking and financial apps, to see whether they were vulnerable in the same way. Among the ones I checked, this application was the only one affected.</p><p>I looked for a bug bounty or vulnerability disclosure program. They did not have one.</p><p>So I contacted the company directly. I explained that I had discovered a critical vulnerability and that I was able to bypass the application’s 2FA mechanism on Android. I asked to be connected with their technical or security team.</p><p>I received no meaningful response.</p><p>Several weeks later, I happened to be at a conference where I learned that one of my friends was using the same application.</p><p>I asked whether he had 2FA enabled. He said ofc,this is required<strong>.</strong></p><p>I told him to enter his password himself on my mobile. I would not ask for his second-factor code, and I would still be able to access the account.</p><p>It took time, fortunately ofc- cause you should not expose your password, He entered the password eventually. The application asked for the second authentication factor. I bypassed it. And I was in.<br>I asked my friends there to email the company behind the app and tell them what you found.</p><p>This is an application people trusted with financially sensitive activity, yet one of its core account-security mechanisms could be bypassed that easily. like piece of cake.</p><p>I did not publicly explain how the bypass worked. I generally do not disclose exploitation details of vulnerabilities like this to anyone other than the affected company, or people who share the same concerns as I do and I trust deeply.</p><p>After the conference, I contacted the company again. I received a reply that sounded as if nobody had actually read my email: “How can we help you?” My response was simple: Did you read my email? I am not asking for your help. I am telling you that you need mine.</p><p>Once again, nothing happened.</p><p>I had completely forgotten about that incident until a few nights ago. I couldn’t sleep and decided to check something that had been on my mind. For some reason, I suddenly remembered that application and thought, let me see where things stand now.</p><p>They had changed the authentication mechanism, which was something they absolutely needed to do. I mean it was mandatory.</p><p>So I put that issue aside and started looking around the application again. I spent a little more time with it and found something else.</p><p>My reaction this time : I closed my laptop and went to sleep. I did not even bother emailing them. Yes, yes, sad but true.</p><p>But my response would be very different if I believed customer data was exposed or people were at immediate risk. Customers should never have to pay the price for a company’s negligence or lack of competence. I care about people who are in a position of less power, and I always try to help them whenever I can. I believe we should all do the same.</p><p>Anyway, what I see today, however, is that AI is making it possible for far more people — even those with relatively limited technical expertise — to discover vulnerabilities that once required much deeper knowledge and experience.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CbZHTw7bKn7Glp7nUZaFtA.png" /></figure><p>That obviously introduces new risks, but I think there is a positive side to it as well.</p><p>Companies are beginning to realize that they can no longer assume their security weaknesses will simply go unnoticed. The barrier to finding vulnerabilities is getting lower, which means organizations have much less room for negligence.</p><p>In that sense, AI may ultimately force companies to take security more seriously, invest in it earlier, and be far more careful with the data their customers trust them with.</p><p>Something improves when ignoring it becomes more expensive? hum,maybe. But let stay in the right side.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d3e9a0072581" width="1" height="1" alt=""><hr><p><a href="https://systemweakness.com/ai-is-making-security-negligence-harder-to-hide-d3e9a0072581">AI Is Making Security Negligence Harder to Hide</a> was originally published in <a href="https://systemweakness.com/">System Weakness</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Sensitive information of hundreds of individuals is publicly accessible on a web application…]]></title>
            <link>https://medium.com/@ghassemiN/sensitive-and-personal-information-of-hundreds-of-individuals-is-freely-and-openly-accessible-on-a-2c320d871c36?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/2c320d871c36</guid>
            <category><![CDATA[information-security]]></category>
            <category><![CDATA[data-exposure]]></category>
            <category><![CDATA[security-by-design]]></category>
            <category><![CDATA[sensitive-data-exposure]]></category>
            <category><![CDATA[digital-privacy]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Sat, 09 Aug 2025 12:09:40 GMT</pubDate>
            <atom:updated>2025-08-09T12:24:46.575Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>Sensitive information of hundreds of individuals is publicly accessible on a web application, including first names, last names, ID numbers, cell phone numbers, addresses, and much more</strong></h3><p>We live in an increasingly connected world where the internet plays a central role in our daily lives. Unfortunately, our personal and sensitive information is frequently exposed to risk, whether intentionally or unintentionally, by us or the online services we use. This exposure puts both users and application owners at risk of a wide range of threats. Users often fall victim to identity theft, online fraud, blackmail, cyberbullying, and more. Meanwhile, companies not only face these same attacks but also the potential loss of reputation, customers, and financial stability. These risks are well-known within the industry, aren’t they?</p><h4>Hundreds of Sensitive Information Records Exposed to Public Access</h4><p>During a routine search on the internet, I came across a considerable amount of sensitive personal information from multiple individuals, all freely accessible on a specific web application. This data included national ID numbers, phone numbers, full names, home addresses, parents’ names, and in some cases, even personal photos. Finding this information in public access raised serious concerns, as it could easily be exploited for malicious purposes by anyone.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/383/1*DAsLYTY8vkev_fAB3mzH4w.png" /></figure><p>After realizing that such sensitive data was freely available, I decided to report the issue to the website administrators. I sent an email to them, providing three examples of the exposed information. In my message, I explicitly stated that this sensitive data needed to be removed immediately to prevent potential misuse.</p><p>I received a response thanking me for reporting the issue, and they assured me that they removed the data, but it took time to remove from google search. However, after some days, I checked that the information was still accessible. I then followed up with another email, reminding them that not only the data I initially reported had not been removed, but there are hundred of records. I am still waiting for their reply- It’s 5 days, and the records are still there.</p><p>Websites and organizations must assume greater responsibility for safeguarding users’ personal information. We can’t simply say “it’s the users’ mistake, so it’s their problem.” Systems should be designed with robust security measures, ensuring that user mistakes do not lead to the exposure of sensitive information. The problem with this web application, as I see it, is that it makes it too easy for users to make mistakes, resulting in the inadvertent disclosure of personal data. A secure system should anticipate human error/mistake and include safeguards to prevent such vulnerabilities from being exploited.</p><p>I first reported this issue to a trusted cybersecurity expert who really cares about digital privacy and asked for their opinion. They advised me to inform the website admin first (which I did a week ago) then a reporter. Today, I also reached out to a reporter. These are the steps I’ve taken so far. However, I’m still following up on the emails to ensure that all the exposed information has been deleted.</p><p>I will not mention the website or its data. I will send this article to them of course. The purpose of this post is to emphasize the critical importance of data protection. Above all, systems must take responsibility for safeguarding data and be designed with security as a top priority. If we identify any issues or mistakes who put people at risk, the next step is to report them and do everything in our power to ensure the problem is properly addressed and resolved.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2c320d871c36" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Security risks often hide in features added later]]></title>
            <link>https://systemweakness.com/security-risks-often-hide-in-features-added-later-ec8e20824fe6?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/ec8e20824fe6</guid>
            <category><![CDATA[sms-otp-verification]]></category>
            <category><![CDATA[security-by-design]]></category>
            <category><![CDATA[bug-bounty-tips]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[vulnerability]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Sun, 27 Jul 2025 15:13:31 GMT</pubDate>
            <atom:updated>2025-07-28T11:28:48.139Z</atom:updated>
            <content:encoded><![CDATA[<p>A few days ago, I placed an order at a physical store. To track the status of my order, they told me I could either use their website or follow the link sent to my phone via SMS.</p><p>two days ago, I decided to check my order status. I went to the website and found a section for order tracking. There were two options: one for in-store purchases and one for online orders.</p><p>Since I made my purchase in-store, I followed that path. The system asked for my phone number — which I had already provided at the time of purchase. I entered the number, received an OTP on my mobile, entered it in the form, and was able to see the current status of my order.</p><p>While using the form, I noticed there was no re-captcha or any other verification mechanism. I decided to test how many times I could request a code for my own number. As it turns out, there were no restrictions — the system kept sending OTPs without any rate limiting or IP blocking.</p><p>Then I wondered what would happen if I entered random phone numbers. I had no idea whether those numbers had any orders associated with them. I tried it — and received no error messages. I even entered two or three phone numbers that I had access to, which I was sure had no orders placed under them — and they received OTPs too.</p><p>I searched to see if this company had a bug bounty program; They didn’t. I emailed them anyway, although to be honest, I wasn’t very hopeful — emails like that often go unread or unanswered. I wasn’t looking for a reward; I just wanted them to fix the issue. Even if they had replied, “You shouldn’t have tested this,” I’d be fine with it — just fix the problem.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*A1UidPMlW4vP4ETphT8xFg.png" /></figure><p>Anyway, I became curious about which company had developed this web service, and who their other clients were. After a bit of digging, I managed to identify the provider.</p><p>I reviewed their list of clients and started checking them one by one to see whether the same issue existed. Most of them had order tracking, but only for online orders — not for in-store purchases. And none of them appeared to suffer from this specific vulnerability — The online order tracking form in all of them required an order number and included a re-CAPTCHA.</p><p>That led me back to the original webservice. I realized that the “track in-store order” feature had likely been added later, as an extension to the core system.</p><p>This is something we’ve seen more frequently— and it’s becoming a common risk. A system might be designed securely from the beginning, but when a plugin, module, or feature is added later, that addition may not go through the same level of security review or design discipline. This is something that both developers and those who are bug hunters should strongly consider.</p><p>So if you’re trying to assess a system or look for potential vulnerabilities, it’s often a good idea to start with the later added features— they are more likely to have flaws. To find what features added later, you may want to check the Internet Archive, review official company announcements (as they often disclose new features), or inspect class names and other design elements — there are various ways to tell if a feature was added after the original design.<br> Of course, none of these methods are 100% reliable; I’m sharing this based on my personal experience.</p><p>Keep your curiosity alive, and your security stronger.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ec8e20824fe6" width="1" height="1" alt=""><hr><p><a href="https://systemweakness.com/security-risks-often-hide-in-features-added-later-ec8e20824fe6">Security risks often hide in features added later</a> was originally published in <a href="https://systemweakness.com/">System Weakness</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Artificial Intelligence Chatbots and the Concerns They Bring]]></title>
            <link>https://medium.com/@ghassemiN/artificial-intelligence-chatbots-and-the-concerns-they-bring-19272200d0d7?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/19272200d0d7</guid>
            <category><![CDATA[ai-chat]]></category>
            <category><![CDATA[ai-chat-bot]]></category>
            <category><![CDATA[privacy]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Fri, 06 Dec 2024 14:15:09 GMT</pubDate>
            <atom:updated>2024-12-07T09:58:40.138Z</atom:updated>
            <content:encoded><![CDATA[<p>It may soon be rare to find an online service that doesn’t have AI in it. When I browse tech news, I often come across new AI-powered tools and the results they deliver… exciting, but also somewhat scary to me. Maybe not “fear,” maybe it’s better described as “concern.”</p><p>Don’t take this from me alone — talk to your friends, especially those who regularly use the internet for their daily tasks. You likely have friends who interact with various AI chatbots, whether for simple Q&amp;A or casual conversations. Some even use them as psychological or career counseling tools.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/360/1*UeDdZMfXnbnmTdW6CggIKQ.jpeg" /></figure><p>We are now facing another phenomenon: a near-friendly relationship or 100% trust between users and AI chatbots. In a way, you might feel sorry for modern humanity, that people are now turning to AI chatbots for emotional support or psychological services instead of humans! But let’s leave that aside for now. Maybe it could be described as who/what to talk to, which can be personal. But 100% trust?</p><p>What worries me is the history leaks! When I use an online service, I ask myself what would happen if the data were leaked? My name, my home address? My ID? my bank info? What information can I fake so that I can still use the service?</p><p>These questions are important! You can’t trust companies and their products 100%. This is becoming more and more evident every day! So it’s always good to know what can be revealed about me and what the consequences are!</p><p>Imagine if your chat history were somehow leaked! Or if someone had unauthorized access to it.</p><p>How many people could be harmed by such leaks? Ordinary individuals who were just seeking family advice and ended up discussing the most sensitive topics with chatbots. Companies and organizations whose employees shared internal codes with chatbots, exposed passwords…</p><p>What I wanted to say was just to share my concern, to know what we are facing and what benefits and risks it could have for us. I know that companies should be careful with our data, but apparently it is not as good as it should be. It is also up to us to know who/what we are talking to and what information we are sharing!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=19272200d0d7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[An easy way to find bugs: Enter wrong data]]></title>
            <link>https://systemweakness.com/an-easy-way-to-find-bugs-enter-wrong-data-0765ed4e9afd?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/0765ed4e9afd</guid>
            <category><![CDATA[audit]]></category>
            <category><![CDATA[web-app-security]]></category>
            <category><![CDATA[pentesting]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Tue, 14 May 2024 11:18:34 GMT</pubDate>
            <atom:updated>2024-05-15T16:48:19.945Z</atom:updated>
            <content:encoded><![CDATA[<p>If you’re looking to find something, don’t follow the path that’s already been prepared, try a strange path</p><p>I want to write about a usual and still important way to find bugs; I myself found a very important bug using this method, which led to purchasing even VIP tickets without paying any money. It was on one of the online ticketing websites for events, concerts, and theaters. I’m not mentioning the website’s name, and after I reported it, fortunately they fixed the bug.</p><p>The reason I’m writing about it now is that I normally use this method to find flaws in other web apps as well. Not always, but it often works. This is a common method for finding bugs, and it’s a question for me why developers don’t test it themselves. The problem may lie in <a href="https://medium.com/system-weakness/security-by-design-329b2f1623da">weak design</a>, not your codes. So, test it.</p><p>The method is quite simple: in online selling systems, add a small item to the cart, maybe you have a valid discount code, add that too, proceed to the checkout stage, and observe and record all the requests and responses (<a href="https://medium.com/system-weakness/how-to-install-burp-suite-certificate-on-an-android-emulator-bb2972ba188c">Burp Suite </a>will help you a lot). Read them all and understand what’s happening. Then, try proceeding with an invalid discount code, and again, read the requests and responses. Understand what requests are being sent and what responses are being returned. Now start the game.</p><p>Does sending a successful response of a previous discount code validation or purchase trick the system? Can you use the ID of a valid discount code you used in a previous purchase? Some discount codes have group IDs; maybe you can use these IDs! Some systems have weak security designs.</p><p>Using the ID of a valid discount code in a subsequent purchase may lead to results, although this depends on the specific design and implementation of the system. Some discount codes may be specifically assigned to an individual or group and may only be usable for subsequent purchases by that individual or group.</p><p>However, some systems may have security weaknesses that make these methods positively impactful on purchases. For example, if a system has a security flaw in discount code validation and only checks the validity of the code without verifying the owner of the code, reusing a valid discount code might yield results.</p><p>These methods can be useful for security testing and identifying system weaknesses. By testing these vulnerabilities, programmers can enhance system security and make necessary improvements in design and implementation.</p><figure><img alt="An easy way to find bugs: Enter wrong data" src="https://cdn-images-1.medium.com/max/760/1*J2_7U0Er7EheTD46ODutfg.png" /></figure><p>As you can see, the method is simple: understand the system and try entering incorrect data, do something the system doesn’t expect, some times you need to compare the incorrect result with the expected one, see how you can play with that.</p><p>This behavior isn’t just about finding bugs, it’s valid for any system you want to bypass.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0765ed4e9afd" width="1" height="1" alt=""><hr><p><a href="https://systemweakness.com/an-easy-way-to-find-bugs-enter-wrong-data-0765ed4e9afd">An easy way to find bugs: Enter wrong data</a> was originally published in <a href="https://systemweakness.com/">System Weakness</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Use 127.0.0.1 instead of localhost]]></title>
            <link>https://blog.devops.dev/use-127-0-0-1-instead-of-localhost-1f756f283fc6?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/1f756f283fc6</guid>
            <category><![CDATA[develoopment]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[latency]]></category>
            <category><![CDATA[localhost]]></category>
            <category><![CDATA[devops]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Tue, 26 Mar 2024 14:25:34 GMT</pubDate>
            <atom:updated>2024-03-27T19:24:49.734Z</atom:updated>
            <content:encoded><![CDATA[<p>Let’s jump into code, try it in python for example:</p><pre>$ python3<br>&gt;&gt;&gt; import socket<br>&gt;&gt;&gt; socket.getaddrinfo(&#39;localhost&#39;,0)<br>THEN<br>&gt;&gt;&gt; socket.getaddrinfo(&#39;127.0.0.1&#39;,0)<br></pre><p>As you will see in the results in your terminal ofc, when you use “localhost” it returns both v6 and v4 IP addresses, but when you use ‘127.0.0.1’ it returns only v4 IP add.</p><p>So why is that important, you may ask! It’s about “wasting some time”:</p><ul><li>some program (programmers) may try IP v6 at first and when it fails they will go for IP v4. Wasting of time.</li><li>By lookup the “localhost” name, there will be an other wasting of time.</li></ul><p>As a result, this simple switch will improve latency depends on your project. Lets put it in this way: Depends on your organization scale, this small improvement may decrease cost in a big range.</p><figure><img alt="Use 127.0.0.1 instead of localhost" src="https://cdn-images-1.medium.com/max/748/1*we1r2gfuJPVUd_64wG8aFQ.png" /></figure><p>I was setting up the Nginx reverse proxy, <a href="https://medium.com/@saderi">my husband</a> saw the config file, I used “localhost”, then he explained me why It’s important to get to used to 127.0.0.1 instead of that. :) that&#39;s why I wanted to write this short article.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1f756f283fc6" width="1" height="1" alt=""><hr><p><a href="https://blog.devops.dev/use-127-0-0-1-instead-of-localhost-1f756f283fc6">Use 127.0.0.1 instead of localhost</a> was originally published in <a href="https://blog.devops.dev/">DevOps.dev</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to find open ElasticSearch databases using Shodan]]></title>
            <link>https://systemweakness.com/how-to-find-open-elasticsearch-databases-using-shodan-fb9314af604a?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/fb9314af604a</guid>
            <category><![CDATA[elasticsearch]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[shodan]]></category>
            <category><![CDATA[bugbounty-tips]]></category>
            <category><![CDATA[vulnerability]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Thu, 06 Jul 2023 18:10:51 GMT</pubDate>
            <atom:updated>2023-07-13T11:34:10.595Z</atom:updated>
            <content:encoded><![CDATA[<p>The power of shodan</p><p>A large part of the Internet is called the Deep web. Deep web contains important and sensitive information that belongs to a person or an organization, a small or large company and etc. In order to access this part of the internet, in a legal way I mean, you need to be authorized or/and authenticated. BUT sometimes these classified data expose due to application vulnerabilities, human errors, or etc.</p><p>These sensitive information exposed publicly, yes, but still we need to find them, they barely show up in our daily google/duckduck search. there are some way to find exposed data, of course for reporting them and maybe receive the reward.</p><p>in this article I will show how to find open/exposed ElasticSearch database with Shodan.</p><figure><img alt="How to find open ElasticSearch databases using Shodan" src="https://cdn-images-1.medium.com/max/627/0*RApkUeOLivxXy7wX.jpeg" /></figure><h3>How to find open ElasticSearch databases using Shodan</h3><p>To do this, let’s go to our favorite<a href="https://shodan.io/"> shodan.io</a>. The Shodan is something like the search engine of internet-connected devices. Just subscribe and if it possible use its Small Business/Corporate plan and hit start.</p><p>1- To begin, we will use the “product” and “port” filters of “shodan”. See the example below:</p><pre>product:elastic port:9200</pre><p>Notice ElasticSearch’s default port is 9200.</p><p>2- Of course, you can also use the “country” filter to restrict the search, for example:</p><pre>product:elastic port:9200 country:COUNTRY_CODE</pre><p>In this example I did not use “country” filter. We are looking for the open database, maybe it has users column, so lets search with this query this time:</p><pre>product:elastic port:9200 users</pre><p>the result:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZnGUs1Lg5dIBrpLg5NMjbw.png" /></figure><p>Now we can see databases with user columns. Look at the size of that column, the larger it is, the more information it has. I kept going with one of them.</p><p>To reach the “users” column, we must use the ElasticSearch query: change the browser address to:</p><pre>http://IP_ADDRESS:9200/_cat/indices?v</pre><p>Now let’s go back to using ElasticSearch queries to see what information is in the user column:</p><pre>http://IP_ADDRESS:9200/users/_search?size=1000</pre><p>What I found: (Of course I colored red to protect users privacy. After all my goal is not to hurt people but help them to be safe and secure.)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/760/1*7tGwB4rz2xyMneLzd3dHRA.png" /></figure><p>As it’s obvious, users private information are public.</p><p><strong>Conclusion</strong>:</p><p>This way we are able to find vulnerable websites, I mean open ElasticSearch db. If you find something, the best thing is to report it to the website administrator. You may get rewards by reporting, but if there is no reward, don&#39;t worry about that, don&#39;t seek revenge or other monkey business :P</p><p>This is the way.</p><p>P.S :I had wrote this article<a href="https://medium.com/@ghassemiN/elasticsearchi-kullanarak-a%C3%A7%C4%B1k-veritabanlar%C4%B1-nas%C4%B1l-bulunur-cc546382b3da"> in Turkish</a> before, today I decided to publish it in English too.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fb9314af604a" width="1" height="1" alt=""><hr><p><a href="https://systemweakness.com/how-to-find-open-elasticsearch-databases-using-shodan-fb9314af604a">How to find open ElasticSearch databases using Shodan</a> was originally published in <a href="https://systemweakness.com/">System Weakness</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to find Phishing websites with Favicon Hash]]></title>
            <link>https://systemweakness.com/how-to-find-phishing-websites-with-favicon-hash-ebbd42390621?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/ebbd42390621</guid>
            <category><![CDATA[phishing-attacks]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[shodan]]></category>
            <category><![CDATA[phishing]]></category>
            <category><![CDATA[favicon-hash]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Mon, 03 Jul 2023 17:38:54 GMT</pubDate>
            <atom:updated>2023-07-04T12:50:14.956Z</atom:updated>
            <content:encoded><![CDATA[<p>Favicon hashes can be use for finding vulnerable and phishing websites</p><p>Favicon hash can be a way to find vulnerable websites, you can use it especially in BugBunty programs. And this is because different versions of software use different favicons.</p><p>But “favicon’s Hash” is not just for finding vulnerable websites. An other useful example is find Phishing websites. Because fake websites try to look exactly like the original website, from top -<strong>favicon</strong>- to bottom. In this way we may find the IP of these phishing websites and report them.</p><figure><img alt="How to find Phishing websites with Favicon Hash" src="https://cdn-images-1.medium.com/max/1024/1*nQ_8YjD_ubW_cpAmZObNuw.jpeg" /><figcaption>source of the photo: <a href="https://dinahosting.com/blog/que-es-un-favicon/">https://dinahosting.com/blog/que-es-un-favicon/</a></figcaption></figure><p>Services like <a href="http://shodan.io/">Shodan</a> calculate and let us search <a href="https://en.wikipedia.org/wiki/MurmurHash">MurmurHash</a> values for all favicons. In this way it can be pretty easy to find specific services and devices.</p><p><strong>Ok lets explain it with an example:</strong></p><p>1- first we need a favicon. As you know to get Favicon’s url, we can get it by inspect element or check “View Page Source” of a website and look for “favicons”, for example:</p><figure><img alt="How to find Phishing websites with Favicon Hash" src="https://cdn-images-1.medium.com/max/950/1*yH1saZIXD-hlYDu9EqkxXA.png" /></figure><p>Click the link and there will be a favicon: <a href="https://en.wikipedia.org/static/favicon/wikipedia.ico">https://en.wikipedia.org/static/favicon/wikipedia.ico</a></p><p>2- Now we need it’s hash. In order to get the favicon’s hash we can use services like faviconhash.com, although there are <a href="https://gist.github.com/yehgdotnet/b9dfc618108d2f05845c4d8e28c5fc6a">python scripts</a> too. I got it through the online service- unfortunately “faviconhash.com” it’s not up anymore. So you should use the python script.</p><figure><img alt="How to find Phishing websites with Favicon Hash" src="https://cdn-images-1.medium.com/max/640/1*vpAlz-6iYL0JpQN6XzQ9hQ.png" /></figure><p>and the result is: 857403617</p><p>3- OK, so far so good. now we are going to use <a href="http://shodan.io/">Shodan</a> and it’s filters for searching this hash: — You need to login to your shodan’s account.</p><pre>http.favicon.hash:NUMBER</pre><p>and the result is:</p><figure><img alt="How to find Phishing websites with Favicon Hash" src="https://cdn-images-1.medium.com/max/1024/1*Z8Z3FSozwUzi390aDNmK3g.png" /></figure><p>As you see there are more than 800 websites that are using the same Favicon. We can now use other filters to find suspicious websites, for example:</p><pre>http.favicon.hash:857403617 product:&quot;Apache httpd&quot;</pre><p>You can learn more about Shodan filters<a href="https://www.shodan.io/search/examples"> from here</a> , for example check this filter for the real example:</p><pre>http.favicon.hash:------- hostname:------ http.html:login</pre><p>I just used Wikipedia to explain this method. I mean, I was not going to find nothing suspicious :) Try this method on real examples, like your org’s websites, or websites that you are working on them at the bug bounty programs.</p><p>Like what I did for “<a href="https://www.ziraatbank.com.tr/tr">Ziraat Bank</a>” (a bank in Turkey), and look what I have found:</p><figure><img alt="How to find Phishing websites with Favicon Hash" src="https://cdn-images-1.medium.com/max/1024/1*k4vRt4hHFCa1n9ff3n3bOA.png" /></figure><p>one of them:</p><figure><img alt="How to find Phishing websites with Favicon Hash" src="https://cdn-images-1.medium.com/max/1024/1*jo1j3IvGJgbCZ4THLgySsw.png" /></figure><p>I hope you find this article useful.</p><p>This is the way.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ebbd42390621" width="1" height="1" alt=""><hr><p><a href="https://systemweakness.com/how-to-find-phishing-websites-with-favicon-hash-ebbd42390621">How to find Phishing websites with Favicon Hash</a> was originally published in <a href="https://systemweakness.com/">System Weakness</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Mitigate bot attacks by Rate limiting]]></title>
            <link>https://systemweakness.com/mitigate-bot-attacks-by-rate-limiting-46c8a5382883?source=rss-15177f890ebb------2</link>
            <guid isPermaLink="false">https://medium.com/p/46c8a5382883</guid>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[security-by-design]]></category>
            <category><![CDATA[rate-limiting]]></category>
            <category><![CDATA[brute-force]]></category>
            <category><![CDATA[ddos]]></category>
            <dc:creator><![CDATA[Nasrin]]></dc:creator>
            <pubDate>Mon, 19 Jun 2023 13:11:15 GMT</pubDate>
            <atom:updated>2023-06-21T08:10:34.031Z</atom:updated>
            <content:encoded><![CDATA[<p>How I accidentally found a vulnerability that costs for the business</p><p>Some days ago, I ordered food from a food app, I picked a restaurant by its ratings, I mean it was random. After 40 min my order delivered, and that restaurant sent a discount code for the next order. I wanted to check the code if it’s validate or not, sometimes they are not :) . I had to open an account on the restaurant’s own web application, instead of the food store application, in order to use discount code.</p><p>I started to create an account. It asked me name, surname, email, phone number, password in their registration form. After accepting terms and polices, for the next step it asked me to verify my phone number by entering the code they sent me through SMS.<br>I received the SMS, as usual I wanted to see if it limits the number of attempts of “Send Code” or not!</p><p>If I had permission to test, I would use tools, like BurpSuite. But I didn’t, so I tried almost 15 times, without any tools, without delay and every 15 times the code was sent to me by SMS. Without blocking my IP, showing captcha, or any other restrictions. It means, there was no limit on the number of times to send SMS.</p><p>This vulnerability can be used for bot attacks. In our case, DoS attack in addition to use server-side resources and “SMS pumping”<strong> </strong>makes extra cost for the business, because of sending SMS. They have to charge their SMS panel so sooner than usual.</p><figure><img alt="Mitigate bot attacks by Rate limiting" src="https://cdn-images-1.medium.com/max/824/1*oOC38pUmw_9ZIl37UTDExg.jpeg" /></figure><p>The solution is “Rate limiting” or using “Captcha” or detects bot by behavior and block them in the first steps; Rate limiting It’s a strategy for limiting network traffic. This strategy mitigate bot attacks like: Brute force and DoS/DDoS attacks. Let’s don&#39;t forget no solution can be 100% useful, this is no exception :)</p><p><strong>How this restaurant app can use rate limiting: It’s just an example.</strong></p><p>1- <strong>First try:</strong> first “Send code” try, and wait 60 seconds until user enter the code.</p><p>2- <strong>Second try</strong>: Second “Send code” try and wait 60 seconds until user enter the code.</p><p>3- <strong>Third try</strong>: Third “Send code” try and wait 60 seconds until user enter the code.</p><p>4- <strong>Fourth try</strong>: Ban IP for 5 min, OR send them code through email, or ask them contact to customer support, ask them pass the captcha or etc.</p><p>These are the security issues that should be considered when designing the application — <a href="https://medium.com/system-weakness/security-by-design-329b2f1623da">Security By Design</a>.</p><p>I wanted to report to them directly, then I decided to write this in the Medium post without mentioning the name and address. I will send this post’s link to the company that designed/developed/maintains their web App.</p><p>This is the way.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=46c8a5382883" width="1" height="1" alt=""><hr><p><a href="https://systemweakness.com/mitigate-bot-attacks-by-rate-limiting-46c8a5382883">Mitigate bot attacks by Rate limiting</a> was originally published in <a href="https://systemweakness.com/">System Weakness</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>