<?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 Harsh kothari on Medium]]></title>
        <description><![CDATA[Stories by Harsh kothari on Medium]]></description>
        <link>https://medium.com/@cyberhrsh?source=rss-3d0aab373e55------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*2pJvWjazjVBZujnOU9c9Tw.jpeg</url>
            <title>Stories by Harsh kothari on Medium</title>
            <link>https://medium.com/@cyberhrsh?source=rss-3d0aab373e55------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 12:11:46 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@cyberhrsh/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 to Get Your First CVE as a Beginner]]></title>
            <link>https://cyberhrsh.medium.com/how-to-get-your-first-cve-as-a-beginner-2d4122121ed4?source=rss-3d0aab373e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/2d4122121ed4</guid>
            <category><![CDATA[cve]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <dc:creator><![CDATA[Harsh kothari]]></dc:creator>
            <pubDate>Thu, 13 Nov 2025 04:56:01 GMT</pubDate>
            <atom:updated>2025-11-13T04:56:01.937Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>The Ultimate Guide for New Hackers Who Want Their Name in Cyber History</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NjKJ4NphAxyXwykUibEqCg.png" /></figure><p>Let me guess…<br>You’ve been hunting bugs for a while.<br>Reported some XSS here, some IDOR there.<br>Maybe got $50, $200, or $0 because “duplicate, thanks”.</p><p>And now, suddenly, you’re like:</p><p><strong>“Bro… I want a CVE. Like a REAL ONE. I want my name on MITRE. I want to be THAT guy.”</strong></p><p>Good.<br>Because today I’ll show you exactly how to do it - even if you’re a complete beginner.</p><p>But we’re doing it <strong>fun way</strong>, not the Wikipedia way.<br>Let’s go.</p><h3>Whoami?</h3><p>Name’s Harsh — aka <strong>cyberhrsh</strong>. I break stuff on the internet.</p><blockquote>Shoutout to <a href="https://medium.com/u/b35078dbf4f6">RivuDon</a>— the man, the myth, the reason this writeup exists.<br>Bro said ‘write it’, and here we are, dropping a full CVE roadmap.</blockquote><h3>Chapter 1: What Exactly is a CVE?</h3><p>CVE = <strong>Common Vulnerabilities and Exposures</strong><br>A CVE is a <strong>global identity number</strong> for a security vulnerability.<br> Example: CVE-2025-52390 (It consists of CVE-Year-Unqiue number)</p><p>Imagine you found a security bug in a library or software that thousands of companies use.<br>CVE ID ensures:</p><ul><li>The vulnerability gets public recognition</li><li>The vendor acknowledges your work</li><li>Tools (NVD, Nessus, Burp, etc.) can track it</li><li>Your name becomes part of security history</li></ul><p><strong>CVE = global fame + credibility</strong> (but not necessarily money).</p><h3>Chapter 2: Bug Bounty vs CVE (The Reality Check)</h3><p>Bug bounty is like dating apps.<br>Swipe left, swipe right, sometimes rejected, sometimes matched.</p><p>But a CVE?<br>That’s marriage.<br>Permanent. Public. Recorded.<br>Your name stays forever.</p><p><strong>Bug bounty = money.<br>CVE = reputation.</strong></p><p>Both are great.<br>But a CVE hits different.</p><p>In simple words,</p><p>Bug bounty = You find a bug in a company’s website/app and report it to THEM only. They fix it and may pay you.</p><p>CVE = You find a vulnerability in an actual software/product/library that MANY companies use. It gets a global ID and becomes publicly documented.</p><p><strong>Bug bounty = private + company-specific.<br>CVE = public + world-impacting.</strong></p><h3><strong>Chapter 3: Where Beginners Should Hunt ?</strong></h3><p>Don’t jump on Windows kernel, Cisco routers, Apache servers on day one.<br>Relax Batman.</p><p>Start with <strong>small open-source projects</strong> — the easiest place to earn CVEs.</p><p>Targets that beginners can handle:</p><ul><li>GitHub repos with 100–1000 stars</li><li>WordPress plugins</li><li>Node.js / Express apps</li><li>PHP projects (50% chances they’re already vulnerable)</li><li>Python Flask/Django projects</li><li>Random npm / pip libraries</li><li>Android open-source apps</li></ul><p>These devs are friendly.<br>They respond.<br>They fix quickly.<br>And they happily give you credit.</p><p><strong>Small project = big CVE opportunity.</strong></p><h3>Chapter 4: What Bugs Actually Become CVEs?</h3><p>Here’s the twist nobody tells you:</p><p>You DON’T need RCE.<br>You DON’T need buffer overflows.<br>You DON’T need hardcore exploitation.</p><p>Most CVEs are:</p><ul><li>Missing authentication</li><li>Default admin login</li><li>Sensitive info leak</li><li>IDOR in open-source</li><li>Broken password reset</li><li>Exposed databases</li><li>Bad JWT handling</li><li>Hardcoded API keys</li><li>Unsafe file upload</li><li>Directory traversal</li><li>SSRF</li><li>Insecure config in widely used software</li></ul><p><strong>Simple bug in popular software = CVE.</strong></p><p>You already know how to find these bugs from bug bounty experience.<br>Just aim at SOFTWARE instead of DOMAINS.</p><h3>Chapter 5: Your CVE Hunting Setup</h3><p>You only need:</p><ul><li>Burp Suite</li><li>Postman</li><li>VS Code</li><li>GitHub search</li><li>A local environment</li><li>Basic bug hunting skills</li><li>Curiosity (most important)</li></ul><p>If you can test a web app in bug bounty,<br>you can test an open-source product too.</p><h3>Chapter 6: The Real Sauce — How to Actually Get a CVE</h3><p>Alright hero, here’s the step-by-step roadmap.</p><h3>Step 1 — Pick a small GitHub project</h3><p>Don’t pick Linux Kernel.<br> Pick a cute little project like:</p><ul><li>“Simple CRM”</li><li>“Open-Source Inventory App”</li><li>“Basic File Sharing Tool”</li><li>“Node.js Authentication Boilerplate”</li></ul><p>Small projects = easier logic flaws.</p><h3>Step 2 — Run it locally</h3><p>You control everything here:</p><ul><li>no rate limits</li><li>no WAF</li><li>no restrictions</li><li>full access to code</li><li>full access to endpoints</li></ul><p>Feels like hacking in God Mode.<br>Enjoy.</p><h3>Step 3 — Attack it like a bug bounty target</h3><p>Try everything:</p><ul><li>IDOR</li><li>Broken auth</li><li>No access control</li><li>Password reset abuse</li><li>Direct file access</li><li>Sensitive info files</li><li>Weird parameters</li><li>API endpoints without login</li></ul><p>Check /admin pages.<br>Check /backup.<br>Check /config.<br>Check for .git/.</p><p>Half the time you’ll scream:<br><strong>“Bruh… is this even secure??”</strong></p><p>Congratulations.<br>That’s a CVE.</p><h3>Step 4 — Document properly</h3><p>Write:</p><ul><li>What’s affected</li><li>Steps to reproduce</li><li>Impact</li><li>Evidence (sanitized)</li><li>Suggested fix</li></ul><p>Be a gentleman hacker.</p><h3>Step 5 — Contact the developer</h3><p>Send a friendly message:</p><p>“Hey! I was using your project and found a security issue.<br>Would you like me to send details privately? Also, would you be open to assigning a CVE?”</p><p>Most devs reply:<br>“Yes please! Thank you!”</p><p>Developers LOVE security researchers who help them.</p><h3>Step 6 — Developer fixes → You get a CVE</h3><p>If the developer is a CNA (rare), they’ll assign the CVE.</p><p>If not, no problem →<br><strong>you request it through MITRE yourself.</strong></p><p>MITRE doesn’t bite.<br>Just fill out the form with details.<br>After some back-and-forth:</p><p>BOOM.</p><p>Your first CVE is live.</p><h3>Chapter 7: What Happens After Getting a CVE</h3><p>You become:</p><ul><li>Officially recognized cyber researcher</li><li>Someone whose work protects users</li><li>Someone with an ID in MITRE</li><li>Someone recruiters notice</li><li>Someone respected in infosec community</li></ul><p>Your name appears in:</p><ul><li>MITRE</li><li>NVD</li><li>GitHub advisories</li><li>Security blogs</li><li>Vendor changelogs</li><li>Google search</li></ul><p>Feels like god mode unlocked.</p><h3>Final Chapter: The Truth No One Says</h3><p>Getting a CVE is <strong>not about being a genius</strong>.<br> It’s about being:</p><ul><li>curious</li><li>consistent</li><li>friendly with maintainers</li><li>detailed in reports</li></ul><p>You don’t need insane hacking skills.<br>You need strategy.</p><p>If you can find a bug in bug bounty,<br>you can find a CVE in open-source way faster.</p><p><strong>Your first CVE is closer than you think.</strong></p><p>Now go hunt.<br>Go break something (legally).<br>Go leave your mark on the internet.</p><p>And when you get your first CVE,<br>message me:</p><p><strong>“BRO IT HAPPENED.”</strong></p><p>Signing off,<br><strong>cyberhrsh</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2d4122121ed4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I Reported a Pre-Account Hijack Affecting Any Gmail User (Even Google Employees)- My Bug…]]></title>
            <link>https://infosecwriteups.com/how-i-reported-a-pre-account-hijack-affecting-any-gmail-user-even-google-employees-my-bug-258180c8dd70?source=rss-3d0aab373e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/258180c8dd70</guid>
            <category><![CDATA[security]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[google]]></category>
            <dc:creator><![CDATA[Harsh kothari]]></dc:creator>
            <pubDate>Thu, 30 Oct 2025 08:53:44 GMT</pubDate>
            <atom:updated>2025-10-30T08:53:44.943Z</atom:updated>
            <content:encoded><![CDATA[<h3>How I Reported a Pre-Account Hijack Affecting Any Gmail User (Even Google Employees)- My Bug Report Journey with Google</h3><h3>Intro - a normal night that turned interesting</h3><p>Bug hunting looks cool on Twitter: big payouts, P1 RCEs, flex posts. Reality is usually half-cold coffee, many tabs, and small logic bugs that break real user trust.</p><p>I found one of those quiet but dangerous bugs on a Google subdomain. Short version: <strong>you could create a full account for any Gmail address, get a token, and change profile or cart - all without email verification and without rate-limiting.</strong> Bad, right? Very bad.</p><p>This is my story - how I found it, how I reported it, the follow-ups, the frustration, and what I learned.</p><h3>Whoami?</h3><p>Ek aur din, ek aur vulnerability. Same banda, naye emotions 😭. Intro ke liye purane writeups padho yaar.<br>(Translation - Another day, another vulnerability. Same guy, new emotions 😭. Read old write-ups for the intro, friend.)</p><h3>The “wait, that worked?” moment</h3><p>I was just poking around the Website. I found the website is using graphql , I already had learned how graphql works and stuff so without wasting time I runned Graphql Introspection query and I found many things. Interestingly i found an registeration mutation. I tried and results , Success.</p><p>No OTP. No verification. Just a token in response. I could update name, add address, and even pre-fill the shopping cart via GraphQL mutations.</p><p>Two feelings hit me: <strong>(1)</strong> “Yes - proof!” and <strong>(2)</strong> “Oh no - this is dangerous.”</p><p>The main reason it is more dangerous because they only have SSO for Sign-up / Sign-in , if the real user later logged in with Google OAuth, they would land inside the account I had already created. Zero-click takeover. Silent and nasty.</p><p>(Fun Fact- They can’t even revoke the account as they don’t have any reset password option or any other way they can revoke the JWT token.)</p><p>I tested carefully and kept all sensitive stuff private. I made a small, repeatable PoC: createCustomer → generateCustomerToken → updateCustomer/addProductsToCart → victim logs in via OAuth → victim ends up in attacker account.</p><h3>PoC and impact (short &amp; clear)</h3><p>I prepared:</p><ul><li>a private PoC video (redacted here),</li><li>reproduction steps (sanitized),</li><li>a clear impact list.</li></ul><p>Why it matters:</p><ul><li><strong>Zero-click account takeover</strong> — no phishing, no user interaction needed.</li><li><strong>Persistent access</strong> — attacker has a valid JWT; victim cannot revoke.</li><li><strong>Business logic abuse</strong> — attacker can prefill carts, spam, change profile, maybe place orders if payment saved.</li><li><strong>Internal risk</strong> — I tested behavior for an internal-style email during private testing (kept private here).</li></ul><p>This was a logic bug, not a shiny exploit - but the damage could be real.</p><h3>Reporting to Google VRP — timeline (real, no filter)</h3><p>I submitted the report and started the usual back-and-forth.</p><ul><li><strong>Jun 13, 2025 — </strong>Report submitted (triaged).</li><li><strong>Jun 18, 2025 — </strong>Sent detailed attack scenario and PoC video to Google on request.</li><li><strong>Jun 28, 2025 — </strong>Clarified that the victim cannot remove attacker or revoke token.</li><li><strong>Jul 7, 2025 — </strong>Shared additional impact (internal-style email test — private).</li><li><strong>Jul 10, 2025 — </strong>Status: accepted (filed to product team).</li><li><strong>Jul 24 &amp; Jul 31, 2025 — </strong>VRP panel’s reward decision: $0.</li><li><strong>Oct 15, 2025 — </strong>Final: closed — reason: application is operated by a third-party vendor (out of scope for VRP).</li></ul><p>Important: I also tried contacting the vendor directly about the bug, but I didn’t get a reply from them. Still, the vulnerability was <strong>eventually resolved </strong>now they have forget password option GraphQL is also secured (good - users protected), even though vendor communication was radio-silent from my side.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4kA-3TgvUnYWBxMIzueiGw.png" /><figcaption>Proof (I am not the fake medium writer ifykyk😂 )</figcaption></figure><h3>The emotional part - excitement and a bit of let-down</h3><p>That PoC moment - big dopamine spike. Getting triaged and accepted - validation. The follow-ups - expected. The closure because of “vendor-operated” scope - frustrating.</p><p>I pushed with WHOIS and ownership info, argued that a Google subdomain with user-facing security issues needs fixing regardless of vendor operation. The panel reviewed and still closed the report citing vendor operation. It felt like: I did the right technical job, but the procedural ladder stopped the reward path.</p><p>Still - the main win: the bug was fixed. That’s what matters for users.</p><h3>What I learned (practical, short)</h3><p>If you hunt logic bugs or build auth systems, keep these in mind:</p><ul><li><strong>Check signup &amp; SSO edges first.</strong> Signup, invite, change-email, SSO-linking - these often hide logic mistakes.</li><li><strong>Don’t issue long-lived tokens before verification.</strong> Tokens before verification = trouble.</li><li><strong>Give users session visibility &amp; revocation.</strong> If someone links via SSO, users should be able to see and remove sessions.</li><li><strong>Rate-limit / CAPTCHA unauthenticated endpoints.</strong> GraphQL is powerful — throttle public mutations.</li><li><strong>Collect ownership evidence early.</strong> WHOIS, registrar info, footer screenshots — they help if scope is questioned.</li><li><strong>Be ready for non-technical fights.</strong> Scope and ownership debates happen. Patience + evidence win.</li></ul><h3>Message for other researchers — short &amp; real</h3><p>Be curious more than flashy. Tools help, but thinking wins. Click around like a normal user. Try weird flows. Ask “what if” and follow it.</p><p>When you report:</p><ul><li>Make PoC repeatable and clean.</li><li>Explain impact in simple words engineers understand.</li><li>Back your claim with evidence (logs, screenshots, WHOIS) - especially if scope might be a problem.</li></ul><p>And emotionally: a $0 reward or closure doesn’t mean you lost. You protected users. You learned. You built credibility. Keep going.</p><h3>TL;DR</h3><p>I found an unauthenticated GraphQL flow on a Google subdomain that allowed account creation, token issuance, and profile/cart manipulation without email verification or rate-limiting - enabling 0-click pre-account hijack. I reported it to Google VRP (detailed PoC + follow-ups). I tried to contact the vendor but didn’t get a reply. The VRP eventually closed the report as vendor-operated, but the vulnerability was resolved. Lesson: logic bugs hide in plain sight - be curious, report clearly, and don’t stop.</p><p>Signing off,<br><strong>cyberhrsh</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=258180c8dd70" width="1" height="1" alt=""><hr><p><a href="https://infosecwriteups.com/how-i-reported-a-pre-account-hijack-affecting-any-gmail-user-even-google-employees-my-bug-258180c8dd70">How I Reported a Pre-Account Hijack Affecting Any Gmail User (Even Google Employees)- My Bug…</a> was originally published in <a href="https://infosecwriteups.com/">InfoSec Write-ups</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Acknowledged by Microsoft — For a Bug I Almost Didn’t Report]]></title>
            <link>https://cyberhrsh.medium.com/acknowledged-by-microsoft-for-a-bug-i-almost-didnt-report-d32033ef3fec?source=rss-3d0aab373e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/d32033ef3fec</guid>
            <category><![CDATA[hacker]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[microsoft]]></category>
            <category><![CDATA[hall-of-fame]]></category>
            <dc:creator><![CDATA[Harsh kothari]]></dc:creator>
            <pubDate>Sat, 19 Jul 2025 16:17:10 GMT</pubDate>
            <atom:updated>2025-07-19T16:17:10.327Z</atom:updated>
            <content:encoded><![CDATA[<h3>Acknowledged by Microsoft — For a Bug I Almost Didn’t Report</h3><h3>Introduction</h3><p>You know that feeling when you test something dumb… and it actually works?</p><p>Yeah. I had that.</p><p>This is the story of how Microsoft Advertising’s CORS config thought my fake domain was legit — just because it looked like it belonged.</p><blockquote><em>TL;DR: I didn’t hack Microsoft. But their server </em>thought I did<em>.</em></blockquote><p>Let’s break down how https://d3vredacted.microsoft.com became an “approved” origin — and how a simple CORS misconfig almost let me sip from the main app like an insider.</p><h3>Real Talk First…</h3><p>I wasn’t planning to write about this one.<br> It didn’t feel flex-worthy. No takeover. No dashboard access. Just… headers.</p><p>But <a href="https://medium.com/u/b35078dbf4f6">RivuDon</a>(legend) reminded me:</p><blockquote><em>“This might help someone who’s just starting out.”</em></blockquote><p>So here we are. If this bug makes one hunter think differently before hitting “Submit Report” — then it’s worth writing.</p><h3>Whoami?</h3><p>Name’s Harsh — aka <strong>cyberhrsh</strong>.</p><p>I find bugs that usually sit in plain sight. I don’t pop 0days every weekend or write ultra-elite exploits in Rust. But I <em>do</em> test what others scroll past.</p><p>Mostly into logic bugs, CORS oddities, and moments where a dev trusted something they absolutely shouldn’t have.</p><blockquote><strong>Note on Redactions:<br> Due to Microsoft’s disclosure policy and to avoid unintentionally exposing any sensitive infrastructure, I’ve redacted the actual subdomains and replaced them with </strong><strong><em>redacted.microsoft.com</em> or similar placeholders. Apologies for the inconvenience — I hope this still gives you enough context to understand the bug and learn from it.</strong></blockquote><h3>What Even Is This Bug?</h3><p>This one’s about CORS — Cross-Origin Resource Sharing.</p><p>Normally, when a browser sends a request to a different domain (like from evil.com to redacted.microsoft.com), <strong>SOP (Same-Origin Policy)</strong> steps in and says,</p><blockquote><em>“Yo. You can’t do that.”</em></blockquote><p>Unless… the target server explicitly says:</p><pre>Access-Control-Allow-Origin: https://evil.com<br>Access-Control-Allow-Credentials: true</pre><p>That’s what happened here — except Microsoft’s config was way too trusting.</p><p>They said:</p><blockquote><em>“If the origin ends with </em><em>redacted.microsoft.com, you’re good.”</em></blockquote><p>So I faked a domain that matched the suffix.</p><p><strong>https://d3vredacted.microsoft.com</strong></p><p>And guess what? Their server was like:</p><pre>Access-Control-Allow-Origin: https://d3vredacted.microsoft.com<br>Access-Control-Allow-Credentials: true</pre><p>Cookies sent. Session trusted. SOP bypassed.</p><h3>Technical Stuff (aka: The Boring But Important Part)</h3><p>No subdomain takeover.<br>No DNS tricks.<br>No shell access.</p><p>I hosted a page on my own domain and spoofed the Origin header like so:</p><pre>fetch(&quot;https://redacted.microsoft.com/some-api&quot;, {<br>  method: &quot;GET&quot;,<br>  credentials: &quot;include&quot;,<br>  headers: {<br>    Origin: &quot;https://d3vredacted.microsoft.com&quot;<br>  }<br>});</pre><p>Microsoft’s backend didn’t validate if the domain was real. It just checked if it <em>looked close enough</em>. endsWith(&#39;redacted.microsoft.com&#39;) logic gone wrong.</p><p>This is what made it vulnerable.</p><h3>Repro Steps</h3><ol><li>Host a site on any domain you control.</li><li>Send a fetch() or XHR request to redacted.microsoft.com</li></ol><ul><li>Use Origin: https://d3vredacted.microsoft.com</li><li>Add credentials: include</li></ul><p>3. Look for this in the response:</p><pre>Access-Control-Allow-Origin: https://d3vredacted.microsoft.com<br>Access-Control-Allow-Credentials: true</pre><p>Boom. The SOP is out the window.</p><h3>But Why Low Severity?</h3><p>Now here’s the part I want every new hunter to read twice:</p><blockquote><strong><em>I didn’t show a full attack scenario.</em></strong><em><br> I didn’t prove how the misconfig could actually leak user data or perform a dangerous action.</em></blockquote><p>That’s on me.</p><p>Because of that, Microsoft marked it as <strong>Low Impact</strong>, and it didn’t qualify for bounty — just an acknowledgment.</p><p>And honestly?<br> They were right.</p><p>I gave them a bug.<br>But I didn’t give them <strong>impact</strong>.</p><p>As one legend triager once told me:</p><blockquote><em>“Show us what you can do — as an attacker.”</em></blockquote><p>So if you find something like this, <strong>don’t rush to report.</strong><br>Try to chain it. Try to show what damage it can do. Think weird. Think creative.</p><h3>The Real Lesson Here</h3><blockquote><strong><em>Impact &gt; Bug</em></strong></blockquote><p>You could find a hundred misconfigs. But if none of them show risk, they’re just curiosities — not vulnerabilities.</p><p>This one? It <em>could</em> have been something bigger if I explored further.<br> But I didn’t. That’s the mistake I made.</p><p>Don’t make the same one.</p><h3>Disclosure Journey</h3><ul><li><strong>Jan 9, 2025</strong> → Reported to MSRC</li><li><strong>Triaged </strong>→ MSRC Case 93990</li><li><strong>Marked Low Severity</strong></li><li><strong>No bounty</strong></li><li>Asked for <strong>acknowledgment</strong></li><li>Got listed in <strong>May 2025 Online Acknowledgements</strong></li><li>Name live by <strong>June 16, 2025</strong></li></ul><h3>Takeaways (Save These Somewhere)</h3><ul><li>CORS misconfigs are real. Don’t sleep on headers.</li><li>Always test with fake origins. Look for reflection.</li><li>Access-Control-Allow-Credentials: true = game on.</li><li>Show the damn impact. Or don’t expect payout.</li><li>Impact &gt; bug, always.</li><li>Don’t be afraid to write about the small ones — someone might learn something big.</li></ul><h3>Acknowledgement</h3><p><strong>Listed</strong> on <a href="https://portal.msrc.microsoft.com/en-us/security-guidance/researcher-acknowledgments-online-services">Microsoft Online Services Security Acknowledgements — May 2025</a><br> <strong>MSRC Case ID</strong>: redacted(Can&#39;t share until the vulnerability get resolved)<br> Reported by: <strong>Harsh Kothari (@cyberhrsh)</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*duvyoAtLr-X_B0eibJGisg.png" /><figcaption><a href="https://msrc.microsoft.com/update-guide/acknowledgement/online">https://msrc.microsoft.com/update-guide/acknowledgement/online</a></figcaption></figure><h3>Wanna Chat Security?</h3><blockquote>I’m offering free 1:1 sessions for aspiring bug hunters:<br> <em>📅 </em><a href="https://topmate.io/cyberhrsh"><em>Book a session → topmate.io/cyberhrsh</em></a></blockquote><blockquote><em>👨‍💻 Let’s connect on LinkedIn: </em><a href="https://www.linkedin.com/in/harshh-kothari"><em>https://www.linkedin.com/in/harshh-kothari</em></a></blockquote><p>Stay curious. Break stuff (legally). And next time you find something boring — ask:</p><blockquote><em>“What could I </em>really<em> do with this?”</em></blockquote><p>Signing off,<br> <strong>cyberhrsh</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d32033ef3fec" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Recon Alone Won’t Make You a Top Bug Hunter (My 2025 Reality Check)]]></title>
            <link>https://infosecwriteups.com/why-recon-alone-wont-make-you-a-top-bug-hunter-my-2025-reality-check-4d7843e39019?source=rss-3d0aab373e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/4d7843e39019</guid>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[osint]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Harsh kothari]]></dc:creator>
            <pubDate>Sat, 12 Jul 2025 15:24:19 GMT</pubDate>
            <atom:updated>2025-07-13T05:56:09.709Z</atom:updated>
            <content:encoded><![CDATA[<h3>Introduction</h3><p>Bug bounty hunting in 2025 has become a wild mix of flashy recon tools, infinite browser tabs, and people posting $20K P1 RCE on Login Page writeups that make you question your entire existence. But here’s the truth no one tweets about: <strong>recon alone won’t carry you</strong>. In this write-up, I’m sharing what I learned the hard way — that understanding the app beats out running 10 tools overnight. Let’s cut the noise and get real about what actually helped me grow as a bug hunter.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OagwEO6HQqPfAcNa0kGzIg.png" /><figcaption>Credits: ChatGPT SORA</figcaption></figure><h3>Whoami?</h3><p>Name’s Harsh — aka <strong>cyberhrsh</strong>. I break stuff on the internet (legally, unless Company ghosting me counts as a crime). I specialize in logic bugs — the kind where you poke something with a dumb idea… and suddenly you’re inside someone’s account. Or their wallet. Or their therapist’s notes (true story, NDA tho).</p><p>I’m not a recon machine. I don’t spend 8 hours watching gau spit out 9000 URLs just to realize 8500 of them are robots.txt. If you vibe with that, you’re gonna like this post.</p><p>This isn’t a guide. It’s a <strong>reality check</strong>.</p><h3>The Recon Obsession Phase (We’ve All Been There)</h3><p>Let’s rewind.</p><p>It’s Nov 2024. I’m “serious” about bug hunting. I’ve got the tools, the Discord tags, and a Notion doc titled <strong>“$2K Bugs Incoming”</strong>.</p><p>The plan? Run <strong>Subfinder</strong>, <strong>Amass</strong>, <strong>gau</strong>, <strong>waybackurls</strong>, <strong>httpx</strong>, <strong>ffuf</strong>, <strong>katana</strong>, and maybe even <strong>shuffle DNS</strong> until the server cries.</p><p>I thought recon <em>was</em> hacking.</p><p>I even had a tmux window showing a cool recon dashboard like:<br> [+] Discovered 2,134 subdomains in 17.04 seconds<br> <em>“WOW. THIS IS IT.”</em></p><p>No bro. That was just <strong>foreplay</strong>.</p><h3>Recon ≠ Reward</h3><p>Here’s the plot twist.</p><p>Despite all those tools, all those wordlists, all those “juicy subdomains” — I had <strong>zero valid bugs</strong> to show. Not even a duplicate. Just screenshots of 403 pages and a /dev folder with one HTML file that said <em>&quot;test lol&quot;</em>.</p><p>And then I did something wild.</p><p>I stopped automating.<br>I opened the site in a browser.<br>I clicked things.<br>I signed up. Reset my password. Looked at what the app <em>actually</em> did.</p><p>And boom.<br>Within a week, I had my first triaged bug — not from recon, but from <strong>understanding</strong>.</p><h3>Bugs Hide in the App — Not in Your Recon Folder</h3><p>Look, recon is cool. It makes you feel like Batman with a terminal.</p><p>But the truth is, most impactful bugs aren’t hiding behind some test-vault-qa-staging4.internal.blabla.io.</p><p>They’re sitting right there on the main app, sipping coffee, waiting for someone to click the “Change Email” button and wonder,<br> <strong>“What if I change it to someone else’s email?”</strong></p><p>Recon can’t ask that question.</p><p>Only <strong>curiosity + logic</strong> can.</p><h3>The Mindset Shift That Changed Everything</h3><p>When I stopped treating bug hunting like digital mining and started treating it like problem-solving, things clicked.</p><ul><li>Instead of brute-forcing URLs, I asked: “What’s the business logic here?”</li><li>Instead of fuzzing 500 parameters, I asked: “Which 3 are actually sensitive?”</li><li>Instead of chasing deep recon, I watched how the app <em>handled users</em>.</li></ul><p>That’s when I found:</p><ul><li>Account takeover without verifying email</li><li>Token issues in GraphQL</li><li>Shopping cart abuse through logic skips</li><li>Pre-hijack scenarios <em>with zero interaction</em></li></ul><p>None of these came from recon. They came from slowing down and using <strong>brain &gt; Burp</strong>.</p><h3>Let’s Talk Human</h3><p>You know what recon doesn’t prepare you for?</p><ul><li>The <strong>anxiety</strong> of waiting 27 days for a triage update</li><li>Getting a <strong>“Won’t Fix”</strong> for an actual working bug</li><li>Debugging your PoC like it’s an OSCP exam</li><li>The <strong>dopamine crash</strong> after a duplicate</li><li>Forgetting what day it is because your tabs say POST /api/v1/register</li></ul><p>But that’s what real bug hunting looks like.</p><p>Not cool dashboards.<br>Not 0-day flexes.<br> Just persistence, curiosity, and lots of <em>“wait… what if?”</em></p><h3>The Point</h3><p>Recon is a <em>tool</em>, not a religion.</p><p>It should <strong>support</strong> your thought process, not <strong>replace</strong> it.</p><p>If you’re spending 90% of your time finding hosts and 10% thinking about how they work — flip that.</p><p>Some of my best bugs came from:</p><ul><li>Clicking around like a normal user</li><li>Testing weird input flows</li><li>Assuming devs make mistakes (they do. they <em>really</em> do.)</li><li>Asking dumb questions… and following through</li></ul><p>Forget recon glory. Aim for bug <strong>clarity</strong>.</p><p><strong>So here’s my take:</strong></p><p>Recon is like Tinder. It helps you discover new options.<br> But if you never talk, observe, or understand what you’re dealing with — you’re not going anywhere.</p><p>This year taught me that mindset beats methodology.</p><p>When I shifted from “find everything” to “understand something,” things finally started to click.</p><p>So if you’ve been stuck in recon loops, constantly scanning but rarely reporting — maybe it’s time to step back and ask:</p><blockquote>What am I really looking for?</blockquote><p>Because most of the time, bugs aren’t hiding in your recon results.<br>They’re hiding in plain sight — inside the app, waiting for someone curious enough to poke the right edge.</p><p>This writeup isn’t some secret recipe. It’s just what worked for me — and maybe, if you’re feeling stuck, it might work for you too.</p><p><strong>A perspective from someone still learning, still messing up, and still showing up.</strong></p><p>Signing off,<br> <strong>cyberhrsh</strong></p><blockquote>🚀 Offering Free 1:1 Mentorship on Cybersecurity &amp; Bug Hunting!<br> Stuck somewhere? Just starting out in cyber? Need guidance, feedback, or just want to chat?<br> I’m also learning every day, and that’s why I’m offering free mentoring sessions — to grow together.</blockquote><blockquote>📅 Book a session here on Topmate: <a href="https://topmate.io/cyberhrsh">https://topmate.io/cyberhrsh</a><br> 👨‍💻 Let’s connect on LinkedIn: <a href="https://www.linkedin.com/in/harshh-kothari">https://www.linkedin.com/in/harshh-kothari</a></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4d7843e39019" width="1" height="1" alt=""><hr><p><a href="https://infosecwriteups.com/why-recon-alone-wont-make-you-a-top-bug-hunter-my-2025-reality-check-4d7843e39019">Why Recon Alone Won’t Make You a Top Bug Hunter (My 2025 Reality Check)</a> was originally published in <a href="https://infosecwriteups.com/">InfoSec Write-ups</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Flashback Fuzzing: How I Found JWT Tokens in the Past & Got Paid for It]]></title>
            <link>https://infosecwriteups.com/flashback-fuzzing-how-i-found-jwt-tokens-in-the-past-got-paid-for-it-8da88d6daf09?source=rss-3d0aab373e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/8da88d6daf09</guid>
            <category><![CDATA[jwt]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[thriller]]></category>
            <dc:creator><![CDATA[Harsh kothari]]></dc:creator>
            <pubDate>Mon, 26 May 2025 04:56:03 GMT</pubDate>
            <atom:updated>2025-05-26T04:56:03.235Z</atom:updated>
            <content:encoded><![CDATA[<p>— a cyber-thriller story by Harsh Kothari</p><blockquote><em>“Pura system hi hack kar diya.”</em><br> — Me, after scrolling through the past on Wayback Machine.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*htl6Wl8lID_OxMHlgpxT1A.png" /></figure><h3>Prologue: The Boring Day That Wasn’t</h3><p>It started like any other day:<br> Low energy, 3 tabs open, 1 chai down.<br> I wasn’t looking for a bug. I was looking for <strong><em>“timepass”</em></strong>.</p><p>So I whispered to myself:</p><blockquote>“Chalo, kuch endpoints extract karte hai. Kuch toh milega.”</blockquote><h3>Act I: The Whisper from the Past</h3><p>It started as just another routine recon session — me, a cup of chai, and https://account.proton.me, a domain from a company renowned for its dedication to online privacy.</p><p>I fired up my trusted browser extension to extract API endpoints, expecting the usual list of requests. But then, one endpoint caught my attention — <strong>/api/core/v4/auth/jwt</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mHEWqq9SB0V8WkOCo-cUXQ.png" /></figure><p>Hmmm. 👀<strong>An auth endpoint with ‘jwt’ in it?</strong><br>That’s either gold… or a ghost town.</p><p>But I wasn’t done.<br>This is where the story turns…</p><h3>Act II: “Flashback Fuzzing” Begins</h3><p>I fed this endpoint into my old friend — <strong>Wayback Machine</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WmuXdd0X3FlBhxUHPPpa7Q.png" /></figure><blockquote>“Kal ka data, aaj ka attack surface.”</blockquote><p><strong>58 snapshots</strong> appeared. Some from 2024. Some from 2025.<br> I clicked on one from <strong>May 4, 2025</strong>.</p><p>Lo and behold.</p><blockquote>{<br> “AccessToken”:”u6m7nhrktfiieg2lvko*****”,<br> “UserID”:”D6Gcoc1Suq5e…”,<br> “ExpiresIn”:1800,<br> “TokenType”:”Bearer”,<br> “Scopes”:[“user”,”newssettings”]<br>}</blockquote><p><strong>JWT tokens. Live once. Archived forever.</strong></p><p>At this moment, I was:</p><blockquote>🧠 Thinking: “Is this real?”<br> 💀 Feeling: “Bro, mujhe goosebumps ho rahe hai.”</blockquote><h3>Act III: The Risks No One Thought Of</h3><p>Here’s what was at stake:</p><p>⚠️ Risk ‎ — ⚡ Details‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎<br><strong>Access Token Exposure — </strong>Tokens exposed in plaintext</p><p><strong>User Metadata Leaked — </strong>Included user IDs, scope, expiration</p><p><strong>Session Hijacking Risk — </strong>If accessed while live, APIs could’ve been abused</p><p><strong>Archive-wide Exposure — </strong>50+ versions = bigger attack surface</p><p><strong>Misconfig Evidence — </strong>Likely bad caching headers or insecure GET usage</p><h3>😂 Meme Break</h3><p>Me entering Wayback Machine &amp; finding JWTs in 2025:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/739/1*ojJ5zLITmdviuNA-vbVuEw.jpeg" /></figure><blockquote><em>“Apun hi bhagwan hai.”</em><br> — Every hacker who pulls tokens from thin air</blockquote><h3>I Call It: Flashback Fuzzing</h3><p>This wasn’t just dumb luck. I followed a method — and now it has a name.</p><blockquote>Flashback Fuzzing</blockquote><blockquote><em>Recon that uses </em><strong><em>past endpoints</em></strong><em> + </em><strong><em>archive data</em></strong><em> to dig out forgotten sensitive responses.</em></blockquote><h3>Steps:</h3><ol><li>Extract API endpoints from JS files or browser tools</li><li>Feed them into <strong>Wayback Machine</strong></li><li>Review archived responses</li><li>Look for anything sensitive — tokens, keys, IDs, misconfig clues</li><li>Report. Repeat. Rejoice.</li></ol><h3>Act IV: The Responsible Disclosure</h3><p>I wrote a respectful, clear report to the <strong>Proton Security Team</strong>.</p><p>Told them:</p><ul><li>JWT tokens were publicly accessible from web archives</li><li>They were tied to real scopes like user and newssettings</li><li>The issue still exists in the archive — a <strong>security time bomb</strong></li></ul><h3>Act V: The Unexpected Reward</h3><p>Few days later…</p><p>🛎️Inbox pinged.</p><blockquote>“Thank you for your report… As a token of our appreciation, here’s a <strong>$100 coupon</strong> to use on Proton services.”</blockquote><h3>Takeaways</h3><ul><li><strong>Old APIs never die</strong>, they just get archived.</li><li>Sensitive data + poor caching = sleeping bugs.</li><li>The <strong>past is a goldmine</strong> — learn to dig it right.</li><li>Respectful reporting = community respect + rewards 💸</li></ul><blockquote>We have to uphold the internet’s integrity.</blockquote><blockquote>Because…</blockquote><blockquote><strong>“Jab system fail hota hai, hacker hi system banta hai.”</strong></blockquote><h3>Final Thoughts</h3><p>This wasn’t a CVE. This wasn’t a critical 0-day.</p><p>But it was a reminder that <strong>recon is an art</strong>, and <strong>curiosity is your biggest tool</strong>.</p><p>And</p><blockquote>“To every silent warrior out there — your time is coming.”</blockquote><p><strong>The browser extension i used for endpoint extraction:-</strong></p><blockquote><a href="https://chromewebstore.google.com/detail/endpoint-extractor/jpkbcgdafcpbnjhdaecnakgjoijiecph">Endpoint Extractor</a></blockquote><p>If you loved this story and want more <strong>hacking thrillers, memes &amp; methods</strong>, follow me or hit me up for collabs.</p><p><strong>Jai recon. Jai fuzzing.</strong> 🫡<br> — <em>Harsh Kothari, aka the Flashback Fuzzer</em></p><p>🥱 <strong>Still reading? You’re a real one.</strong></p><p>And since you made it this far…</p><p>🎁 <strong>Bonus Level Unlocked: The Mentorship Side Quest!</strong></p><p>Ever felt like:<br> <em>“Bhai, recon toh kar liya… ab aage kya?”</em><br> or<br> <em>“Payloads toh hai, but report likhne mein JEE ka paper lagta hai!”</em></p><p><strong>I got you.</strong> 🚀</p><blockquote><em>🚀 </em><strong><em>Offering Free 1:1 Mentorship on Cybersecurity &amp; Bug Hunting!</em></strong><em><br>Stuck somewhere? Just starting out in cyber? Need guidance, feedback, or just want to chat?<br>I’m also learning every day, and that’s why I’m offering </em><strong><em>free mentoring sessions</em></strong><em> — to grow together.</em></blockquote><blockquote><em>📅 Book a session here on </em><strong><em>Topmate</em></strong><em>: </em><a href="https://topmate.io/cyberhrsh"><em>https://topmate.io/cyberhrsh</em></a><em><br>👨‍💻 Let’s connect on </em><strong><em>LinkedIn</em></strong><em>: </em><a href="https://www.linkedin.com/in/harshh-kothari"><em>https://www.linkedin.com/in/harshh-kothari</em></a></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8da88d6daf09" width="1" height="1" alt=""><hr><p><a href="https://infosecwriteups.com/flashback-fuzzing-how-i-found-jwt-tokens-in-the-past-got-paid-for-it-8da88d6daf09">Flashback Fuzzing: How I Found JWT Tokens in the Past &amp; Got Paid for It</a> was originally published in <a href="https://infosecwriteups.com/">InfoSec Write-ups</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Broke Authentication — Without Exploiting Anything]]></title>
            <link>https://infosecwriteups.com/i-broke-authentication-without-exploiting-anything-9b1d7905b70f?source=rss-3d0aab373e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/9b1d7905b70f</guid>
            <dc:creator><![CDATA[Harsh kothari]]></dc:creator>
            <pubDate>Wed, 14 May 2025 04:56:31 GMT</pubDate>
            <atom:updated>2025-05-14T04:56:31.504Z</atom:updated>
            <content:encoded><![CDATA[<h3>I Broke Authentication — Without Exploiting Anything</h3><blockquote><em>How a simple logic flaw let me access a real user account without email verification</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uXpAeSXhMXaxRxFhPwo81g.png" /><figcaption>Image credits:- Chatgpt</figcaption></figure><h3>Introduction</h3><p>What if I told you that you could log into a user account on a major platform without verifying the email address — no payloads, no tools, no brute force?</p><p>No fancy scripts.<br> No XSS, no IDOR, no SQLi.<br> Just… broken logic.</p><p>In this post, I’ll break down a real authentication flaw I found in a live system — and how such issues are more common than you think.</p><h3>Whoami?</h3><blockquote>Name’s Harsh — aka <strong>cyberhrsh</strong>. I break things on the internet (legally) and scream “<strong>BROOO</strong>” when I find bugs that shouldn’t exist in 2025. I’m not here to run Burp for 9 hours or fuzz 800 parameters. I like logic bugs. Chill bugs. Bugs that just fall into your lap because someone forgot a return false;.</blockquote><blockquote>This one? It let me log into an account without verifying my email. Yes. In production. In 2025. Let’s talk about how I did absolutely nothing… and still got in.</blockquote><h3>The Vulnerability</h3><p><strong>Target:</strong> Private crypto trade platform<br><strong>Issue:</strong> Account access granted <em>without verifying the email address</em></p><blockquote><em>Due to disclosure limitations, I can’t reveal the name of the platform — but yes, it handled real crypto transactions, live wallets, and personal user data.</em></blockquote><h3>Step-by-Step Breakdown</h3><h4>1. Registration Flow</h4><p>I went through a typical registration process:</p><ul><li>Entered my email</li><li>Set a password</li><li>Got a message like “Please verify your email”</li></ul><p>So far, so good.</p><h4>2. The Silent Bypass</h4><p>Out of curiosity, I went to the login page <strong>without verifying my email</strong>.<br> Typed in my email and password.<br> Hit “Login.”</p><p>✅ Boom — I was inside the account dashboard.</p><p>No confirmation check.<br>No token enforcement.<br>No restrictions at all.</p><blockquote><em>I had full access to the dashboard — including features like trading, profile settings, and wallet previews.</em></blockquote><h3>Why This Is Wildly Dangerous</h3><p>This isn’t just a UX bug.<br> It’s a <strong>critical business logic flaw</strong> on a live platform where security should be airtight.</p><h4>Real-World Impact:</h4><ul><li>Spam/bot account creation at scale</li><li>Attackers can automate fake accounts and abuse promos or airdrops</li><li>Financial actions possible before account trust is established</li><li>Bypasses a key <em>identity verification layer</em></li></ul><p>On a crypto platform, that’s a <strong>disaster waiting to happen</strong>.</p><h3>Bug Hunter Takeaway</h3><p>This bug taught me that sometimes…</p><blockquote><em>The most dangerous bugs are the ones that </em>don’t look like bugs at all<em>.</em></blockquote><p>As a bug hunter, always:</p><ul><li>Skip steps — see what breaks</li><li>Ask “What’s missing here?” instead of “What can I inject?”</li><li>Treat <strong>logic as your attack surface</strong></li></ul><h3>What Happened Next?</h3><p>I reported this issue directly to the team (it was part of a private program).<br>They acknowledged it quickly and pushed a fix.</p><blockquote>Status: <strong>Patched</strong></blockquote><h3>Final Thoughts</h3><p>This was one of those bugs that made me stop and say: <em>“Wait… that actually worked?”</em></p><p>It’s proof that in bug bounty, tools are cool — but <strong>understanding how systems work (and fail)</strong> is your ultimate advantage.</p><blockquote>If this blog helped you, hit <strong>clap</strong>, leave a <strong>comment</strong>, and <strong>follow</strong> for more real-world hacks! 🔥</blockquote><blockquote>🚀 <strong>Offering Free 1:1 Mentorship on Cybersecurity &amp; Bug Hunting!</strong><br> Stuck somewhere? Just starting out in cyber? Need guidance, feedback, or just want to chat?<br> I’m also learning every day, and that’s why I’m offering <strong>free mentoring sessions</strong> — to grow together.</blockquote><blockquote>📅 Book a session here on <strong>Topmate</strong>: <a href="https://topmate.io/cyberhrsh">https://topmate.io/cyberhrsh</a><br> 👨‍💻 Let’s connect on <strong>LinkedIn</strong>: <a href="https://www.linkedin.com/in/harshh-kothari">https://www.linkedin.com/in/harshh-kothari</a></blockquote><blockquote>Thanks for reading — see you in the next one! ✌️</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9b1d7905b70f" width="1" height="1" alt=""><hr><p><a href="https://infosecwriteups.com/i-broke-authentication-without-exploiting-anything-9b1d7905b70f">I Broke Authentication — Without Exploiting Anything</a> was originally published in <a href="https://infosecwriteups.com/">InfoSec Write-ups</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How a Curious Curl and a Ghost Domain Got Me into the Dutch Government’s Hall of Fame]]></title>
            <link>https://osintteam.blog/how-a-curious-curl-and-a-ghost-domain-got-me-into-the-dutch-governments-hall-of-fame-bf1537f4a737?source=rss-3d0aab373e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/bf1537f4a737</guid>
            <category><![CDATA[hall-of-fame]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[hacker]]></category>
            <dc:creator><![CDATA[Harsh kothari]]></dc:creator>
            <pubDate>Tue, 15 Apr 2025 11:25:44 GMT</pubDate>
            <atom:updated>2025-05-16T06:38:36.517Z</atom:updated>
            <content:encoded><![CDATA[<h3>Introduction</h3><p>I was just casually running Nuclei on a list of Dutch government subdomains when something caught my eye.</p><p>Among all the usual HTTP responses and cookie flags, one <strong>“informational” finding</strong> showed a strange domain in a CORS header: <a href="http://sbelastingdienst.nl./">http://sbelastingdienst.nl.</a></p><p>The origin domain looked a lot like an official one — but a tiny voice in my head said: “Wait… what even is <strong>sbelastingdienst.nl</strong>?”</p><p>So I paused everything, manually opened the URL… and it <strong>didn’t even exist</strong>. That’s when things got real.</p><p>Here’s how that curiosity led me to a serious CORS vulnerability — and eventually landed me in the <strong>Dutch Tax Department’s Hall of Fame</strong> 🇳🇱</p><h3>Whoami?</h3><p>Before diving into the hack, a quick intro. My name is Harsh Kothari aka <strong><em>cyberhrsh</em></strong>, and I’m a security researcher, bug hunter, and all-around tech enthusiast. I spend my time uncovering security flaws, automating workflows, and occasionally ranting about bad bounty payouts haha😂.</p><p>Now, back to the hack.</p><h3><strong>Recon 101: Nuclei + Manual Digging</strong>🧭</h3><p>I was doing a basic recon sweep of *.belastingdienst.nl using tools like Subfinder, httpx, and <strong>Nuclei</strong>. During the scan, I came across:</p><blockquote><a href="https://cpapps.belastingdienst.nl/">https://cpapps.belastingdienst.nl</a></blockquote><p>Looked like a standard login portal — so I opened it in the browser. While Nuclei was still running in the background, I started manually checking a few of these subdomains.</p><h3>Suspicious CORS Header💡</h3><p>After a few minutes, Nuclei flagged something interesting on https://cpapps.belastingdienst.nl — an Access-Control-Allow-Origin header pointing to:</p><blockquote><a href="http://sbelastingdienst.nl/">http://sbelastingdienst.nl</a></blockquote><p>That looked super close to the real domain… but it was missing just one character — probably a typo, or some internal misconfiguration. So I tried visiting <a href="http://sbelastingdienst.nl./">http://sbelastingdienst.nl.</a></p><p><strong>No website. No content. Domain not even registered.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RXSjYM_I0M73eTT2JPoE1Q.png" /><figcaption>Unregistered Domain (Now it is resolved)</figcaption></figure><p>That’s when the lightbulb went off 💡</p><h3>Quick Repro with Curl 🧪</h3><p>To confirm, I sent this curl request:</p><blockquote>curl -I -H “Origin: <a href="http://sbelastingdienst.nl/">http://sbelastingdienst.nl</a>&quot; “<a href="https://cpapps.belastingdienst.nl/">https://cpapps.belastingdienst.nl</a>&quot;</blockquote><p>And got this in response:</p><blockquote>Access-Control-Allow-Origin: <a href="http://sbelastingdienst.nl/">http://sbelastingdienst.nl</a> <br>Access-Control-Allow-Credentials: true</blockquote><p>Boom 💣</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BBDHV3O20GZdhodRBWB1JQ.png" /></figure><p>That’s a serious CORS misconfiguration. Since the domain wasn’t registered, <strong>anyone could buy it</strong>, host a site, and make credentialed cross-origin requests to cpapps.belastingdienst.nl.</p><p>The server would happily respond with sensitive data — thinking it’s talking to a trusted origin.</p><h3>Exploitation Potential</h3><p>Here’s what an attacker could do after registering sbelastingdienst.nl:</p><ul><li>Host a fake site and bait authenticated users</li><li>Use JavaScript to send cross-origin requests to the real API</li><li>Steal cookies, tokens, or headers.</li><li>Potentially hijack user sessions and access sensitive data</li></ul><p>All because the server blindly trusted a <strong>typo’d domain</strong>. Oof.</p><h3>Responsible Disclosure 📬</h3><p>I reported the issue to the Dutch Tax Department’s security team with:</p><ul><li>Steps to reproduce</li><li>Explanation of the unregistered domain</li><li>Potential impact and risk</li><li>Suggested fixes (restrict origins, disable credentials, etc.)</li></ul><p>The team was quick to respond, fixed the issue responsibly, and… ✨</p><h3>🏅 Hall of Fame</h3><p>They added me to their <strong>Responsible Disclosure Hall of Fame</strong> — and honestly, that felt good.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZwdBwd4-AZLAUdFRDmWbvg.png" /></figure><p>It wasn’t the flashiest bug, but it was clean, real, and impactful. And it started with good old recon + curiosity.</p><p><strong>You can check the hall of fame :-</strong></p><p><a href="https://www.belastingdienst.nl/wps/wcm/connect/bldcontenten/standaard_functies/individuals/contact/data-leak-vulnerability-abuse-computer-systems/hall-of-fame-cvd">HALL OF FAME</a></p><h3>🧠 Final Thoughts</h3><p>This bug was a great reminder that:</p><ul><li>Automation is great (shoutout to Nuclei),</li><li>But <strong>manual testing + curiosity still wins</strong></li><li>And yes, CORS misconfigs can be <strong>really bad</strong></li></ul><p>Sometimes, just opening a weird URL during recon can lead to a full disclosure and a Hall of Fame mention.</p><p><strong><em>Keep hunting 👊</em></strong></p><p><strong><em>COMMUNICATION LOGS:</em></strong></p><ol><li>Reported through email on 24–03–2025 11:40 AM</li><li>First response on 24–03–2025 06:51 PM</li><li>Second response stating Hall Of Fame on 25–03–2025 02:49 PM</li></ol><p><strong>Feel Free to connect with me on linkedin:</strong></p><p><a href="https://www.linkedin.com/in/harshh-kothari/">https://www.linkedin.com/in/harshh-kothari</a></p><p><strong>If you find this article helpful, do clap, leave a comment and follow to encourage me to write more. ♥️</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bf1537f4a737" width="1" height="1" alt=""><hr><p><a href="https://osintteam.blog/how-a-curious-curl-and-a-ghost-domain-got-me-into-the-dutch-governments-hall-of-fame-bf1537f4a737">How a Curious Curl and a Ghost Domain Got Me into the Dutch Government’s Hall of Fame</a> was originally published in <a href="https://osintteam.blog/">OSINT Team</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Duplicate to Letter of Appreciation: How I Hacked NASA]]></title>
            <link>https://infosecwriteups.com/from-duplicate-to-letter-of-appreciation-how-i-hacked-nasa-3f8b5a6c105c?source=rss-3d0aab373e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/3f8b5a6c105c</guid>
            <category><![CDATA[nasa]]></category>
            <category><![CDATA[swag]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[vdp]]></category>
            <category><![CDATA[hacking]]></category>
            <dc:creator><![CDATA[Harsh kothari]]></dc:creator>
            <pubDate>Sun, 09 Feb 2025 06:54:39 GMT</pubDate>
            <atom:updated>2025-05-15T04:54:57.289Z</atom:updated>
            <content:encoded><![CDATA[<h3>Introduction</h3><p>One day, while browsing <strong>LinkedIn</strong>, I saw a post from a cybersecurity researcher sharing an appreciation letter they got from NASA. The curious kid in me waked up, so I decided to check out <strong>NASA’s Vulnerability Disclosure Program</strong> to see what kind of security testing they allowed.</p><p>To be honest i was <strong>not expecting</strong> what is going to be happened next !!</p><h3>Whoami?</h3><p>Before diving into the hack, a quick intro. My name is<strong> Harsh Kothari </strong>aka <strong><em>cyberhrsh</em></strong>, and I’m a security researcher, bug hunter, and all-around tech enthusiast. I spend my time uncovering security flaws, automating workflows.</p><h3>Recon: Finding the Perfect Target</h3><p>I followed the <strong>recon process</strong> (will share soon) which included finding the subdomains, filtering the interesting subdomains, finding vulnerable URLs etc. When this was running I decided not to waste time and check manually to find something <strong>jhakass</strong> 😂</p><p>So, I turned to Google dorks, but I quickly realized my dorks weren’t unique and what I was finding had probably already been reported. That’s when I started to find some unique dorks and I stumbled upon a writeup titled</p><blockquote><strong>Advanced Google Dorking</strong></blockquote><p>There were plenty of write-ups on the topic too! If you get a chance, check out <a href="https://medium.com/u/9e41d1b8a839">AbhirupKonwar</a> articles — they’re fantastic and I’m sure you’ll love them!</p><p><a href="https://medium.com/@abhirupkonwar04/list/advanced-google-dorking-8817a1178836">List: Advanced Google Dorking | Curated by Abhirup Konwar | Medium</a></p><p>After using some Google dorks, I stumbled upon an XLS file that was leaking PII details of registrants from some meetup or event. The file was located at</p><blockquote><a href="https://*.*.nasa.gov/*/tally.xls">https://*.*.nasa.gov/*/tally.xls</a></blockquote><p>without wasting a single second I reported it on Bugcrowd. Ahh you think now i’ll get the LOR ? Nah, how can my luck be this good? Turns out, I got a duplicate submission.</p><p>I was a little disappointed at first, but I didn’t let that stop me. I kept testing, and after a few days, I received an email saying the vulnerability had been resolved. Curious, I went back to check the XLS file link, and sure enough, the data had been removed. Instead, I saw this message:</p><blockquote>File Removed Removed; this exists to flush caches</blockquote><p>However, I wasn’t convinced that it was truly gone. I waited a bit, thinking the removal might just be part of the cache-clearing process. But then I decided to dig deeper.</p><p>Then, I went to the <strong>homepage</strong> and found that the <strong>same data </strong>was still available there. I cross-checked the details I had reported initially, and to my surprise, the data was identical — it was the same, legit information !!</p><p>So I submitted a response request from <strong>Bugcrowd</strong> stating <strong>Issue is reproducible </strong>and then in few days i got this reply from Bugcrowd !!</p><blockquote>Based on previous steps to replicate vulnerability, system owner has been remediated. Thank you please accept this letter as a token of our appreciation for your efforts in detecting this vulnerability.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*buocUbfaufv_YVXbLCYimQ.png" /><figcaption>WHATTTT I DID IT 🎉!!!</figcaption></figure><h3>Lessons Learned</h3><ol><li><strong>Don’t assume that a fix means a complete resolution</strong> — Just because something appears to be fixed doesn’t always mean it is. Always double-check and re-test after a fix has been applied.</li><li><strong>Persistence pays off</strong> — Sometimes, you’ll face setbacks, like duplicate submissions or early rejections, but persistence is key. Keep pushing forward, and you might discover something even more valuable.</li><li><strong>Leverage resources and community knowledge</strong> — When you feel stuck, refer to write-ups, blogs, and guides. You might discover new methods, like unique Google dorks, that lead you to a vulnerability.</li><li><strong>The importance of reporting</strong> — Don’t hesitate to report vulnerabilities as soon as you find them. Even if your submission doesn’t initially result in an immediate fix, it could lead to something much bigger down the line.</li></ol><p><strong><em>COMMUNICATION LOGS:</em></strong></p><ol><li>Reported through Bugcrowd on<strong> 04 Dec 2024 10:07 AM</strong></li><li>Bugcrowd placed a blocker on NASA on <strong>04 Dec 2024 12:08 AM</strong></li><li>Got duplicate on <strong>16 Dec 2024 9:00 PM</strong></li><li>Resolved on <strong>26 Dec 2024 10:03 PM</strong></li><li>Submit response request on <strong>29 Dec 2024 12:17 PM</strong></li><li>Got LOR on <strong>31 Dec 2024 02:37 AM</strong></li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bznVccrLz55zLMVJTV5csw.jpeg" /></figure><p><strong>Feel Free to connect with me on linkedin:</strong></p><p><a href="https://www.linkedin.com/in/harshh-kothari/">https://www.linkedin.com/in/harshh-kothari</a></p><p><strong>If you find this article helpful, do clap, leave a comment and follow to encourage me to write more. ♥️</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3f8b5a6c105c" width="1" height="1" alt=""><hr><p><a href="https://infosecwriteups.com/from-duplicate-to-letter-of-appreciation-how-i-hacked-nasa-3f8b5a6c105c">From Duplicate to Letter of Appreciation: How I Hacked NASA</a> was originally published in <a href="https://infosecwriteups.com/">InfoSec Write-ups</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I Hacked the Dutch Government and Got a Lousy T-Shirt]]></title>
            <link>https://infosecwriteups.com/how-i-hacked-the-dutch-government-and-got-a-lousy-t-shirt-bb0f5716bbe1?source=rss-3d0aab373e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/bb0f5716bbe1</guid>
            <category><![CDATA[dutch]]></category>
            <category><![CDATA[swag]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[dutch-government]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <dc:creator><![CDATA[Harsh kothari]]></dc:creator>
            <pubDate>Thu, 30 Jan 2025 13:47:56 GMT</pubDate>
            <atom:updated>2025-05-15T04:54:09.593Z</atom:updated>
            <content:encoded><![CDATA[<h3>Introduction</h3><p>I wasn’t expecting much when I started poking around a Dutch government website. Just another night of recon, running automated tools, and hoping to find some low-hanging fruit.</p><p>Over time, I had reported more than <strong>25 bugs</strong>, only to be met with the usual frustrations — <strong>duplicates, N/A responses, or low-severity dismissals</strong>. It was disheartening, but every bug hunter knows the game: persistence is key.</p><p>Then one night, while reading <strong>Medium write-ups and disclosed reports</strong>, I realized I had become familiar with the technologies the Dutch government relied on. That’s when <strong>an idea struck me</strong> — a potential attack vector that hadn’t been widely explored.</p><p>Excited, I put my theory to the test. A few hours of tinkering later, I had successfully exploited a vulnerability. This time, it wasn’t a duplicate. It wasn’t N/A. It was real, impactful, and couldn’t be ignored.</p><p>I finally had my moment… and that’s how I ended up with a <strong>t-shirt</strong>.</p><h3>Whoami?</h3><p>Before diving into the hack, a quick intro. My name is Harsh Kothari aka <strong><em>cyberhrsh</em></strong>, and I’m a security researcher, bug hunter, and all-around tech enthusiast. I spend my time uncovering security flaws, automating workflows, and occasionally ranting about bad bounty payouts haha😂.</p><p>Now, back to the hack.</p><h3>Recon: Finding the Perfect Target</h3><p>Bug bounty hunting is a mix of patience, automation, and knowing where to look. While reading <strong>Medium write-ups and disclosed reports</strong>, I noticed a common issue — <strong>Drupal installations are sometimes misconfigured, leaving the </strong><strong>/core/install.php endpoint accessible.</strong></p><p>If this page is publicly available, it could allow an attacker to <strong>reinstall Drupal</strong> and gain complete control over the website. That was all the motivation I needed to start hunting.</p><h3>Google Dorking for Vulnerable Targets</h3><p>Instead of manually checking websites, I built a <strong>Google Dork</strong> to streamline the process:</p><blockquote>inurl:”/core/install.php” site:.nl</blockquote><p>This query narrowed down results to <strong>Dutch websites</strong> potentially running an exposed Drupal installation page. After filtering out false positives, I found something promising — <strong>a live site with an accessible </strong><strong>/core/install.php page.</strong></p><p>At this point, my heart started racing. If this was truly unprotected, it could mean <strong>full site takeover</strong>. Time to validate the bug.</p><p><strong>BUT BUT BUT</strong></p><p>Here’s the catch, how would i know this is government related website or not?</p><p>Well, then i opened the website and started analyzing the content i found this</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NgL5pYKVmTnziuKpPzpGJA.png" /><figcaption>This is Of course a gov website</figcaption></figure><h3>Exploiting the Vulnerability</h3><p>With the target now identified and the vulnerability confirmed, I was ready to test the exposure. The /core/install.php page was still accessible, and it was clear that the website wasn’t properly secured.</p><p>At this point, I was faced with a decision — should I proceed and <strong>attempt a reinstall of the website</strong> through this endpoint? <strong>Could this result in a full takeover?</strong> I was aware of how easy it would be to trigger the installation process and potentially gain control. But despite how straightforward the exploit was, I wasn’t sure about the <strong>consequences</strong> of reinstalling the site.</p><p>I didn’t want to create chaos or damage the site in any way. Instead, I made the responsible decision to <strong>report the vulnerability</strong> rather than take advantage of it.</p><h3>Reporting the Vulnerability</h3><p>I took my time to write a <strong>clear, detailed report</strong>. After all, it wasn’t just about finding the bug — it was about showing the Dutch government how dangerous it could be.</p><p>In my report, I outlined:</p><ul><li><strong>The exposed endpoint</strong> (/core/install.php)</li><li><strong>How it could lead to a complete site takeover</strong></li><li><strong>The steps to reproduce</strong> the issue</li><li><strong>Suggested remediation steps</strong> (i.e., restrict access to the endpoint, ensure proper configuration)</li></ul><p>I submitted it through their official <strong>bug bounty platform</strong>, hoping that the severity of the bug would be recognized.</p><h3>The T-Shirt Response</h3><p>I won’t lie — I was on edge waiting for the response. After all, this was a serious vulnerability that could have compromised sensitive data and allowed an attacker to potentially control a government website. But will my dream to wear dutch tshirt this time will be fulfilled?</p><p>After what felt like an eternity of silence, the response came:</p><p><em>“The organization has resolved the vulnerability so as a token of<br>our appreciation we would like to offer you a T-shirt”</em></p><p><strong><em>WHAT WHAT WAIT? I DID IT 🎉</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/776/1*l20n-Ymy2Y7hhL15B6C9SQ.png" /><figcaption>BEST MOMENT FOR REAL !!!!</figcaption></figure><h3>Lessons Learned</h3><ol><li><strong>Recon is essential.</strong> Knowing the technologies and common misconfigurations can help you identify high-impact bugs.</li><li><strong>Patience and persistence matter.</strong> Even if your previous reports didn’t get much attention, every bug you find brings you one step closer to success.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4V4M9C21GCLFpW_BoarCzg.jpeg" /></figure><p><strong><em>WAIT WHATS THIS DOPPER BOTTLE AND HOW I RECIEVED IT? </em>Wait for next write-up 😉</strong></p><p><strong><em>COMMUNICATION LOGS:</em></strong></p><ol><li>Reported through <a href="https://english.ncsc.nl/contact/reporting-a-vulnerability-cvd">https://english.ncsc.nl/contact/reporting-a-vulnerability-cvd </a>on 23–12–2024 7:53 PM</li><li>Triaged on 24-12-2024 02:23 PM</li><li>Patched and asked for details for Tshirt on 27–12–2024 02:21 PM</li><li>Tshirt received on 24–01–2025 at 11:47 AM</li></ol><p><strong>DUTCH GOV SCOPE:</strong></p><p><a href="https://github.com/zzzteph/DutchGovScope">GitHub - zzzteph/DutchGovScope: Dutch Government Bugbounty scope analysis, subdomains and url lists</a></p><p><strong>Feel Free to connect with me on linkedin:</strong></p><p><a href="https://www.linkedin.com/in/harshh-kothari/">https://www.linkedin.com/in/harshh-kothari</a></p><p><strong>If you find this article helpful, do clap, leave a comment and follow to encourage me to write more. ♥️</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bb0f5716bbe1" width="1" height="1" alt=""><hr><p><a href="https://infosecwriteups.com/how-i-hacked-the-dutch-government-and-got-a-lousy-t-shirt-bb0f5716bbe1">How I Hacked the Dutch Government and Got a Lousy T-Shirt</a> was originally published in <a href="https://infosecwriteups.com/">InfoSec Write-ups</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>