<?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 Utaa on Medium]]></title>
        <description><![CDATA[Stories by Utaa on Medium]]></description>
        <link>https://medium.com/@mhrdkaa._?source=rss-2b668d28a0c9------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*f814BZAAksxCg7THcLkPfQ.jpeg</url>
            <title>Stories by Utaa on Medium</title>
            <link>https://medium.com/@mhrdkaa._?source=rss-2b668d28a0c9------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 21:08:27 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@mhrdkaa._/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 Made €200 Bypassing Cross-Portal Access Control]]></title>
            <link>https://medium.com/@mhrdkaa._/how-i-made-200-bypassing-cross-portal-access-control-209644918546?source=rss-2b668d28a0c9------2</link>
            <guid isPermaLink="false">https://medium.com/p/209644918546</guid>
            <category><![CDATA[pentesting]]></category>
            <category><![CDATA[bug-hunting]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[bugbounty-writeup]]></category>
            <dc:creator><![CDATA[Utaa]]></dc:creator>
            <pubDate>Tue, 29 Sep 2026 14:19:33 GMT</pubDate>
            <atom:updated>2026-09-29T14:19:33.343Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CDfSST27hznSOyy4lmGq4w.png" /></figure><p>Hey everyone! In this writeup, I’m excited to share my experience finding a <strong>Broken Access Control / Cross-Portal Authorization Bypass</strong> vulnerability in a private bug bounty program.</p><p>What made this finding interesting was how a simple logic flaw across two related workflow features netted a total of <strong>€200</strong> in bounties across two separate reports.</p><p>To respect the program’s privacy guidelines, we will refer to the target domain as <strong>redacted.com</strong>.</p><h3><strong>1. Understanding the Architecture: Multi-Portal Setup</strong></h3><p><strong>redacted.com</strong> operates as a multi-tenant B2B platform featuring isolated “Portals” for different organizations. Within this platform, organizational structures are managed through two main entities: <strong>Business Units (BU)</strong> and <strong>Group Business Units (Group BU)</strong>.</p><p>Under their Role-Based Access Control (RBAC) model, access should strictly enforce the following rule:</p><ul><li>An Admin assigned to <strong>Portal A</strong> should only manage Business Units within <strong>Portal A</strong>.</li><li>They must never be able to create, edit, or delete Business Units belonging to <strong>Portal B</strong>.</li></ul><p>On the user interface (UI) level, this control seemed to be in place. Buttons to add or edit units for organizations outside the user’s scope were disabled or hidden. But as bug hunters know: <strong>Client-side restrictions are not real security.</strong></p><h3><strong>2. Bug #1: Never Trust Client-Side Controls (€100)</strong></h3><p>While testing the <em>Group Business Unit</em> creation feature, I noticed that the web UI prevented my account (Portal A Admin) from selecting or creating groups for other organizations.</p><p>I bypassed the UI restriction and analyzed the backend request using Burp Suite. The creation request was sent directly to:</p><pre>/admin/organizations/group-business-units/new</pre><p>This endpoint required a key parameter: targetAccountId (an identifier representing the organization&#39;s account).</p><h4><strong>The Recon Step:</strong></h4><p>To exploit this, I needed a valid targetAccountId belonging to a victim organization (<strong>Portal B</strong>). I tested the listing endpoint: POST /business-units/list</p><p>By sending a standard payload {&quot;limit&quot;: 1000, &quot;offset&quot;: 0}, the server responded with a complete list of Business Units across <strong>all portals</strong>, exposing their respective targetAccountId values.</p><h4>The Exploitation:</h4><p>I grabbed a targetAccountId belonging to Portal B (e.g., 0017Qxxxxxx) and crafted a direct POST request:</p><pre>POST /admin/organizations/group-business-units/new HTTP/1.1<br>Host: api.redacted.com<br>Authorization: Bearer &lt;admin_token_portal_A&gt;<br>Content-Type: application/json<br><br>{<br>  &quot;targetAccountId&quot;: &quot;0017Qxxxxxx&quot;,<br>  &quot;groupName&quot;: &quot;got Hacked&quot;<br>}</pre><p>The backend <strong>failed to validate</strong> whether my authenticated session had management rights over targetAccountId. The server responded with HTTP 200 OK, and the new Group BU was successfully created inside the victim&#39;s portal.</p><p>I submitted the report, and <strong>€100 was awarded.</strong></p><h3><strong>3. Bug #2: Escalating to Edit &amp; Delete (€100)</strong></h3><p>After the first report was accepted, I asked myself: <em>“If creation lacks ownership checks, what about modification and deletion?”</em></p><p>I investigated the sister endpoints handling individual Business Units:</p><ul><li>/admin/organizations/business-units/new</li><li>/business-units/group/{id}</li></ul><h3>The Workflow:</h3><ol><li>Authenticated as an Admin for <strong>Portal A</strong>.</li><li>Extracted victim entity IDs from Portal B via /business-units/group/list.</li><li>Sent direct UPDATE and DELETE requests targeting those entity IDs.</li></ol><h3>Deletion Logic Flaw:</h3><p>When sending a DELETE /business-units/group/{victim_group_id} request, the server returned an error response:</p><pre>{&quot;code&quot;:&quot;FAILED&quot;,&quot;description&quot;:&quot;Can not delete group business unit&quot;}</pre><p>However, upon refreshing the target state, the entity was actually deleted! The backend executed the deletion query despite returning a failure status to the client — a classic example of flawed response handling.</p><h3>Modification Logic Flaw:</h3><p>I was also able to arbitrarily edit the titles and properties of Business Units belonging to other organizations.</p><p>Since this flaw affected distinct endpoints and directly impacted Business Unit assets (rather than just groups), I submitted it as a second report. Although initially flagged as a potential duplicate, I clarified the distinct scope and endpoint behavior with the triage team. The report was re-opened and accepted for an <strong>additional €100!</strong></p><h3>4. Key Takeaways</h3><ol><li><strong>Frontend Restrictions ≠ Backend Security:</strong> Disabling UI elements or modifying DOM state does not secure an API. Always test direct HTTP requests.</li><li><strong>Enforce Server-Side Ownership Checks:</strong> Ensure every sensitive API endpoint verifies: <em>“Does the authenticated user own or have explicit permission to manage the target object ID?”</em></li><li><strong>Communicate Clearly with Triagers:</strong> If your report is initially closed or miscategorized, politely provide step-by-step evidence explaining why the endpoint or scope is distinct.</li></ol><p>Happy hunting!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*d32lpltppU5cnQtFeY885Q.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=209644918546" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Hardcoded Credentials Lead to Internal Data Exposure & a $300 Bounty]]></title>
            <link>https://medium.com/@mhrdkaa._/a-couples-hunt-hardcoded-credentials-lead-to-internal-data-exposure-a-300-bounty-933db89ac98c?source=rss-2b668d28a0c9------2</link>
            <guid isPermaLink="false">https://medium.com/p/933db89ac98c</guid>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[bug-bounty-writeup]]></category>
            <dc:creator><![CDATA[Utaa]]></dc:creator>
            <pubDate>Tue, 11 Aug 2026 14:04:18 GMT</pubDate>
            <atom:updated>2026-10-01T03:16:19.528Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*coMxeXQPZ65-0hWJt9KFpA.png" /></figure><p>In the world of bug bounty and penetration testing, we often hyper-focus on complex vulnerabilities like RCE, SSRF, or SQLi. Many assume bug hunting is strictly a solo endeavor. However, in this write-up, we (my girlfriend and I) want to share the story of how our joint hunting session yielded a critical finding in a CMS backend architecture.</p><p>This is a story about how a token left behind in a production JavaScript bundle, combined with a little GraphQL manipulation, led to the exposure of internal data, secret draft content, and wide-open Public Registration.</p><p>Let’s break down how our collaboration uncovered this logic flaw and misconfiguration.</p><h3><strong>The Reconnaissance: Dissecting the JavaScript Bundle Together</strong></h3><p>It all started with a standard recon session on a target (let’s call it <strong>Platform X</strong>). The target is a modern application built with React and powered by a <strong>Strapi CMS</strong> backend with a GraphQL endpoint.</p><p>While I was busy mapping out subdomains and fuzzing directories, my girlfriend focused on manually analyzing the Single Page Application (SPA) in the browser. While examining the main.[hash].js file in the production environment, she spotted two highly interesting lines of code and immediately called me over:</p><ol><li>A static <strong>CH=”HARDCODE CREDENTIALS”</strong>.</li><li>A hardcoded x-include-drafts: true header inside the bundle.</li></ol><p>To a developer, this might just be leftover testing environment (staging/preview) code that accidentally slipped through. But to us, it was the master key.</p><h3>The Discovery: Piecing the Puzzle Together</h3><p>Armed with the API token and the header, we started testing the target’s GraphQL endpoint at https://cms.target.com/graphql.</p><p>Normally, published content (like blog posts) is publicly accessible. However, the x-include-drafts: true header instructs Strapi to respond with content that is still in <strong>DRAFT</strong> status (publishedAt: null).</p><p>We tried a simple GraphQL query using the token:</p><pre>curl -s -X POST &quot;https://cms.target.com/graphql&quot; \<br>  -H &quot;Authorization: Bearer &lt;TOKEN&gt;&quot; \<br>  -H &quot;Content-Type: application/json&quot; \<br>  -H &quot;x-include-drafts: true&quot; \<br>  -d &#39;{&quot;query&quot;:&quot;{ caseStudies { data { id attributes { title slug publishedAt } } } }&quot;}&#39;</pre><p>The server responded with dozens of case studies that had never been released to the public.</p><pre>{<br>  &quot;data&quot;: {<br>    &quot;caseStudies&quot;: [<br>      { <br>        &quot;data&quot;: { <br>          &quot;attributes&quot;: { <br>            &quot;title&quot;: &quot;example&quot;, <br>            &quot;slug&quot;: &quot;example&quot;, <br>            &quot;publishedAt&quot;: null<br>          } <br>        } <br>      }<br>    ]<br>  }<br>}</pre><p>We validated the slug on the live website, and it returned a 404 Not Found. This confirmed that we were accessing unreleased data meant only for the internal marketing team.</p><h3>The Impact: More Than Just Draft Articles</h3><p>The exploration didn’t stop there. With this open GraphQL access, we divided our tasks to explore vulnerability chaining and discovered a much more massive impact:</p><h3>1. Competitive Intelligence &amp; Confidential Data Leak</h3><p>Beyond just titles, we could pull the full HTML body of the draft articles. These contained confidential client business metrics. To competitors, this financial information and release strategy is a goldmine.</p><h3>2. C-Suite PII Exposure</h3><p>Through the Author relation in the GraphQL posts, the database structure exposed executive email addresses (like the personal emails of the CEO &amp; Founder). This data is highly susceptible to high-level spear-phishing attacks.</p><h3>3. Open Registration Without Verification</h3><p>The exploration continued when my girlfriend discovered that the plugin::users-permissions.auth.register action was left active for the Public role. What does this mean? Anyone could register a new CMS account without authentication or admin approval!</p><pre>curl -s -X POST &quot;https://cms.target.com/graphql&quot; \<br>  -H &quot;Content-Type: application/json&quot; \<br>  -d &#39;{&quot;query&quot;:&quot;mutation { register(input: { username: \&quot;poc-hacker\&quot;, email: \&quot;hacker@poc.com\&quot;, password: \&quot;Password123!\&quot; }) { jwt user { id username email } } }&quot;}&#39;</pre><p>The system responded with a valid JWT, and upon logging in, the account status showed confirmed: true and blocked: false. Although the Authenticated role didn&#39;t currently have elevated privileges for dangerous actions (all write/read actions returned 403 Forbidden), this flaw allowed attackers to perform massive user database pollution.</p><h3>4. GraphQL Permission Enumeration</h3><p>Finally, we successfully extracted the entire CMS permission model (66+ permission actions), revealing hidden internal custom endpoints like api::data-transfer.migrateHideForEmbedded.</p><h3>Inspiration &amp; Takeaways</h3><p>The findings from our couple’s hunting session might seem simple, but the impact was extraordinary. Here are some key takeaways:</p><p><strong>For Bug Hunters:</strong></p><ul><li><strong>Never ignore JavaScript files.</strong> De-obfuscate them and look for sensitive keywords like token, bearer, x-, or secret. Often, the keys to the kingdom are left in the most accessible places.</li><li><strong>Collaboration brings new perspectives.</strong> Sometimes, one person is hyper-focused on finding complex bypasses, while the other spots a blatant business logic flaw right in front of them.</li><li><strong>Understand the target’s architecture.</strong> Knowing how React interacts with GraphQL and Strapi is crucial for chaining exploitation scenarios (like understanding the magic behind the x-include-drafts function).</li></ul><p>Sometimes, the biggest vulnerabilities aren’t born from genius system bypasses, but from forgotten debug code left behind before pushing to production.</p><p><strong><em>Keep hacking, keep exploring, and always check those JS bundles!</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OkbCRXs8OeTiihomRl694g.png" /></figure><p>— <strong>Putra Mahardika (@mhrdkaa._)</strong> &amp; <strong>Rara Sindy Aprilia Putri</strong> (Instagram @rssnyy)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=933db89ac98c" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>