<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: HackerMe</title>
    <description>The latest articles on DEV Community by HackerMe (@hackerme).</description>
    <link>https://dev.to/hackerme</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4094444%2F9bc80efa-e28d-446e-b320-b5ccf9afd7d9.jpg</url>
      <title>DEV Community: HackerMe</title>
      <link>https://dev.to/hackerme</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/hackerme"/>
    <language>en</language>
    <item>
      <title>The Container Store Has About 100 Stores, and a Cyberattack Stopped Its Online Orders. How Many Days of Sales Could Your Store Lose?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Fri, 20 Nov 2026 20:02:12 +0000</pubDate>
      <link>https://dev.to/hackerme/the-container-store-has-about-100-stores-and-a-cyberattack-stopped-its-online-orders-how-many-5302</link>
      <guid>https://dev.to/hackerme/the-container-store-has-about-100-stores-and-a-cyberattack-stopped-its-online-orders-how-many-5302</guid>
      <description>&lt;p&gt;If you take orders on a website, this is worth a look this week. The Container Store's website stopped taking orders after a cyberattack. On Friday, 20 November 2026, parent company Neighborhood Intelligence said The Container Store had restored its website and systems after a cybersecurity incident the weekend of 14-15 November. Physical stores stayed open, using other ways to process sales. The company has not said how long online ordering was down. How many days of sales could your store lose if your website could not take orders?&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;Neighborhood Intelligence said in a 20 November 2026 statement that its wholly owned subsidiary, The Container Store, had restored its website and systems following a cybersecurity incident "this past weekend." Chain Store Age, reporting that announcement, dates the incident to the weekend of 14-15 November. The Record, Retail Dive and Chain Store Age all covered the same company statement. The company's own release was also listed the same day on the Neighborhood Intelligence investor-relations site and on BusinessWire; those full pages could not be opened directly for this draft (the IR pages returned a block, and BusinessWire timed out), so the wording below is taken from the outlets that quoted it.&lt;/p&gt;

&lt;p&gt;The company said that after it detected suspicious activity, it activated its security procedures and proactively suspended online order-taking at The Container Store to protect customer information and its systems. The Record described the event as a cyberattack that forced the company to shut down online ordering and limited in-store payment processing. All brick-and-mortar stores stayed open, the company said, and served customers through "alternative transaction-processing procedures."&lt;/p&gt;

&lt;p&gt;Neighborhood Intelligence said it hired independent cybersecurity experts to help with the investigation, containment and restoration. The investigation is still going. The company said it does not believe customer financial data was affected, because The Container Store does not store or maintain consumer financial or payment information. The Record, writing about the same statement, said that so far it does not appear the attackers stole customer financial data.&lt;/p&gt;

&lt;p&gt;The company said the incident was isolated to The Container Store. Other Neighborhood Intelligence businesses, including Bed Bath &amp;amp; Beyond, Overstock, buybuy BABY and Kirkland's, were not affected. It said it expects a minimal impact on consolidated revenue for the quarter, and that any impact on profitability is expected to be mitigated by business interruption insurance. CEO Marcus Lemonis said the company "made the deliberate decision to prioritize the protection of our customers and our systems while keeping our stores open and serving customers." The Record says The Container Store has about 100 stores across the U.S., and that Neighborhood Intelligence, a rebrand of Bed Bath &amp;amp; Beyond, bought the company in July for $150 million.&lt;/p&gt;

&lt;p&gt;What the company has not said matters as much as what it has. Retail Dive asked what caused the incident, whether it was a ransomware attack, and how long the website was down. The company did not answer those questions. It has not publicly confirmed ransomware, and it has not named a group.&lt;/p&gt;

&lt;p&gt;Separately, on 5 November 2026, the ransomware group n0n claimed, in a listing reported by DeXpose, that it had attacked The Container Store and taken a shopper chat corpus of 46,013 typed customer messages, search session analytics, 9,305 named customer email addresses from the account portal, and a staff directory. The listing was marked "NOT PAID â€” EXPOSED." That claim has not been independently verified. Chain Store Age and The Record both note that the company has not publicly addressed the n0n claims, and Chain Store Age says there has been no independent verification that the listing is related to the later weekend incident. This article treats those figures as an unverified claim, not as confirmed facts. The Record also reports that n0n emerged in September and, according to cybersecurity experts, specializes in ransomware and data extortion. That is background on the group, not proof that it caused this outage.&lt;/p&gt;

&lt;h2&gt;
  
  
  Who is affected
&lt;/h2&gt;

&lt;p&gt;The company says only The Container Store's website and systems were affected, and that its other brands were not. The Record says the chain has about 100 U.S. stores. Shoppers who tried to order online during the outage could not complete those orders. People in physical stores could still buy, using backup payment procedures.&lt;/p&gt;

&lt;p&gt;The wider lesson is for any store that takes orders on a website, including a much smaller shop. If checkout stops, online sales stop, even if the physical store stays open. A large chain may have business interruption insurance and a team that can switch to backup payment procedures. Many small stores do not.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why it matters for a small business
&lt;/h2&gt;

&lt;p&gt;The Container Store is a well-known retailer, and its parent says it still expects only a limited effect on the quarter, in part because of insurance. A small shop often does not have that cushion. If your website or checkout is down for a weekend, or for several weekdays, those online orders do not come in later by themselves. Customers go somewhere else. That is why the question is how many days of sales your own store could lose, not how a large chain will book the quarter.&lt;/p&gt;

&lt;p&gt;The company's statement about payment data is specific and limited: it says it does not store consumer payment information, and it does not believe that data was affected. The investigation is still open. That does not automatically protect other customer details, and it does not restore missed online sales. The unverified n0n listing, if it were ever confirmed, pointed at emails, chat messages and a staff directory, not payment cards. Even when card numbers are not stored on your site, a downtime event can still cost you days of sales.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check today
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Ask how you would take orders if your website went down
&lt;/h3&gt;

&lt;p&gt;Write down a simple backup: phone orders, in-store only, a second checkout tool, or a manual process. Know who decides to take the site offline, who talks to customers, and how you keep selling. The Container Store could keep its stores open with alternative payment procedures. If you sell mostly or only online, you need a plan before the site stops.&lt;/p&gt;

&lt;p&gt;We have also written about cases where a website itself was forced offline, including &lt;a href="https://security.hackerme.ai/hacked-homepage-hockey-federation/" rel="noopener noreferrer"&gt;a hacked homepage that took the world hockey federation's site down&lt;/a&gt; and &lt;a href="https://security.hackerme.ai/courts-california-vendor-went-down/" rel="noopener noreferrer"&gt;courts that went dark when one vendor was hit&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Check what on your store website is open to the internet, and keep checkout tools updated
&lt;/h3&gt;

&lt;p&gt;Ask your developer or host what on your site can be reached from the public internet: the store, the admin page, checkout plugins, staging copies, and any leftover tools. A &lt;a href="https://brandingme.ai/app/r/FSvE5vNtUVv6DkakDnPEyQ" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see from the internet, without logging in. It is not a full review of your business, and it will not by itself prove you are safe.&lt;/p&gt;

&lt;p&gt;Keep the store, checkout and plugins updated. We have covered store-specific risks such as &lt;a href="https://security.hackerme.ai/checkout-plugin-admin-login/" rel="noopener noreferrer"&gt;a checkout plugin that can hand a stranger your admin login&lt;/a&gt; and &lt;a href="https://security.hackerme.ai/rootways-magento-skimmer/" rel="noopener noreferrer"&gt;a credit-card stealer hidden in online store checkouts&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Do not treat "we do not store card numbers" as the whole plan
&lt;/h3&gt;

&lt;p&gt;The Container Store's statement about payment data is one important check. Also ask where customer names, emails, order history and staff directories live, who can access them, and how you would tell customers if something went wrong. If a scan finds a serious issue, our guide to &lt;a href="https://security.hackerme.ai/website-scan-found-critical-flaws/" rel="noopener noreferrer"&gt;what to do when a website scan finds critical flaws&lt;/a&gt; is a starting point.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. Keep backups and a basic incident plan
&lt;/h3&gt;

&lt;p&gt;You do not need to wait for a confirmed ransomware label to prepare. Have a recent backup you have actually tested, and a short list of who to call if the site or checkout stops. Our guide to &lt;a href="https://security.hackerme.ai/ransomware-preparedness-for-small-business-websites-defenses-isolation-and-verification/" rel="noopener noreferrer"&gt;ransomware preparedness for small business websites&lt;/a&gt; covers defenses, isolation, and how to check that a backup really works.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  How many days of sales did The Container Store lose?
&lt;/h3&gt;

&lt;p&gt;The company has not published an exact start or end time, and it has not given a figure for missed online sales. Chain Store Age, reporting the company's announcement, says the cybersecurity incident was the weekend of 14-15 November 2026. Neighborhood Intelligence said on Friday, 20 November, that the website and systems had been restored. Retail Dive asked how long the website was down and did not receive an answer. This article does not invent a day count. The useful question for a smaller shop is the one in the title: how many days of sales could your store lose if online orders stopped?&lt;/p&gt;

&lt;h3&gt;
  
  
  Was it ransomware, and was it the group called n0n?
&lt;/h3&gt;

&lt;p&gt;The company has not said the incident was ransomware, and it has not named a group. Retail Dive asked whether it was a ransomware attack and got no answer. On 5 November, about a week and a half before the weekend incident, n0n claimed a leak of customer messages, email addresses and a staff directory, according to DeXpose. Chain Store Age and The Record say the company has not publicly addressed that claim, and there is no independent verification that it is connected to the later outage. Treat it as a claim, not as a confirmed cause.&lt;/p&gt;

&lt;h3&gt;
  
  
  Was customer payment data stolen?
&lt;/h3&gt;

&lt;p&gt;The company says it does not believe customer financial data was affected, because The Container Store does not store or maintain consumer financial or payment information. The Record, reporting the same announcement, said that so far it does not appear the attackers stole customer financial data. The investigation is still going.&lt;/p&gt;

&lt;h3&gt;
  
  
  Were other brands in the same company affected?
&lt;/h3&gt;

&lt;p&gt;Neighborhood Intelligence said the incident was isolated to The Container Store, and that Bed Bath &amp;amp; Beyond, Overstock, buybuy BABY and Kirkland's were not affected.&lt;/p&gt;

</description>
      <category>security</category>
      <category>retail</category>
      <category>webdev</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>Hackers' AI Found a Hole Nobody Knew About and Broke In Within 11 Minutes. What on Your Server Is Open to the Internet?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Fri, 20 Nov 2026 18:11:27 +0000</pubDate>
      <link>https://dev.to/hackerme/hackers-ai-found-a-hole-nobody-knew-about-and-broke-in-within-11-minutes-what-on-your-server-is-4fl</link>
      <guid>https://dev.to/hackerme/hackers-ai-found-a-hole-nobody-knew-about-and-broke-in-within-11-minutes-what-on-your-server-is-4fl</guid>
      <description>&lt;p&gt;If you run a website, an admin page, or any tool on a server that can be reached from the internet, this is a check worth doing this week. Amazon Threat Intelligence reported on 20 November 2026 that attackers used autonomous AI agents to find a hole nobody had reported yet in an open-source medical-imaging web app. BleepingComputer, writing about the same research, says the agents broke in within 11 minutes: they went from mapping the app's public interface to using that unknown hole. They later used the same hole to take full control of a university server, and then against a second university.&lt;/p&gt;

&lt;p&gt;Amazon's team saw the whole campaign because the attackers left their own dashboard open to the internet with no password.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;On 10 September 2026, Amazon Threat Intelligence found a public copy of Cairn, an open-source tool that can coordinate several AI agents to reach a goal written in plain language. The copy was reachable from the internet and had no login, so anyone could see the operators' prompts and the agents' work history.&lt;/p&gt;

&lt;p&gt;The attackers had given the agents lists of internet addresses and a simple instruction: "obtain shell access," meaning find a way in and take control of the machine. According to CJ Moses, CISO of Amazon Integrated Security, the agents then planned, carried out, and adjusted their own steps. The operator did not choose each next move. Amazon says the campaign ran for about 48 hours from 8 September 2026 and targeted universities in East Asia. The universities were not named.&lt;/p&gt;

&lt;p&gt;Not every attempt worked. One effort ran 148 actions over 16 hours and failed, including a long try against a database service. Moses said those agents "ran continuously for 16 hours without breaks, sleep, or working-hour constraints." In a separate attempt, the agents spent more than six hours trying more than 30,000 password combinations for about 80 real usernames, including against a remote login, an admin dashboard and a VPN, and never found a working login.&lt;/p&gt;

&lt;p&gt;One attempt did work. The agents found a previously unknown flaw in an open-source medical-imaging AI web app. BleepingComputer identifies that app as MONAI Label, a tool researchers use to label medical images. Amazon assesses with high confidence that the flaw was unknown at the time: there was no public fix or advisory. In about 11 minutes, the agents went from mapping the app's public interface to using that hole. In a later session, they used the same hole to take full control of a university server, then against a second university running the same software.&lt;/p&gt;

&lt;p&gt;Amazon contacted the maintainers, and they released a fix. BleepingComputer says that fix came out on 18 November 2026. AWS also worked with a government partner to notify the affected organizations, which were not AWS customers. Amazon assesses with moderate confidence that the actor was China-linked, based on the language of the prompts, the time zone on the scripts, and the choice of targets. This article follows Amazon's own wording and does not treat that as proven.&lt;/p&gt;

&lt;h2&gt;
  
  
  Who is affected
&lt;/h2&gt;

&lt;p&gt;The victims in this report are universities, and they have not been identified in public. The wider lesson is not limited to universities or to medical-imaging software. Anyone who runs a website, an admin panel, an API, a dashboard, or a self-hosted open-source tool that can be reached from the internet can face the same kind of attention. If you run MONAI Label, update it. If you run other tools on a server, assume a stranger can find them if they are left open.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why it matters for a small business
&lt;/h2&gt;

&lt;p&gt;You may not think of your login page, your admin panel, or a leftover test app as "open to the internet." If they can be reached without being on a private network, they are. In one case, the agents found a hole nobody knew about and moved in minutes. In others, they spent hours trying passwords and still failed, which is a reminder that rate limits and lockouts matter. The attackers themselves were caught because they left a dashboard open with no password.&lt;/p&gt;

&lt;p&gt;Moses warned that an agent that finds and uses an unknown hole in minutes can finish its work faster than many security teams can even review an alert. A small business often has the same pattern: a website, a store or WordPress admin page, a hosting control panel, or an old tool nobody uses anymore, all sitting on a server.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check today
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Check which apps, admin panels and APIs are reachable from the internet
&lt;/h3&gt;

&lt;p&gt;Ask your developer or host for a simple list: what on your server can be opened from the public internet? Include admin pages, customer logins, APIs, dashboards, staging copies, and any self-hosted tools. If you do not recognise an item, or nobody uses it, treat it as a problem until you know why it is there. A &lt;a href="https://brandingme.ai/app/r/4v3cOJg6sWlKixhBsuEEng" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see from the internet, without logging in. It is not a full review of your server, and it will not by itself prove you are safe.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Take down what you do not need, and update the rest
&lt;/h3&gt;

&lt;p&gt;Turn off or remove tools, admin pages and test copies you do not need. For what you still use, apply the latest updates, especially for self-hosted open-source apps. The maintainers of the medical-imaging app in this case released a fix on 18 November. We have also covered cases where a service was left reachable and then abused, including &lt;a href="https://security.hackerme.ai/port-1433-exposed-sql-server-passwords/" rel="noopener noreferrer"&gt;SQL Server logins left open on the internet&lt;/a&gt; and &lt;a href="https://security.hackerme.ai/web-server-flaws-no-password/" rel="noopener noreferrer"&gt;web-server flaws that can let attackers in without a password&lt;/a&gt;. If it does not need to be public, it should not be public.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Turn on login rate limits and lockouts
&lt;/h3&gt;

&lt;p&gt;The agents tried more than 30,000 password combinations for 80 real usernames over six hours and failed. That is the kind of guessing that rate limits and account lockouts are meant to stop. If your site, admin page, email, or hosting panel can be tried again and again with no pause, ask your developer to turn on limits and lockouts. Our guide to &lt;a href="https://security.hackerme.ai/brute-force-login-protection-what-to-configure-what-to-test-and-what-to-ask-your-developer/" rel="noopener noreferrer"&gt;brute-force login protection&lt;/a&gt; walks through what to ask for. A related case is a &lt;a href="https://security.hackerme.ai/password-reset-form-dooray/" rel="noopener noreferrer"&gt;password-reset form that attackers guessed their way into&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. Never leave a dashboard open without a password
&lt;/h3&gt;

&lt;p&gt;Amazon found this campaign because the attackers left their own Cairn dashboard reachable from the internet with no login. The same mistake is easy to make with a hosting panel, a store admin page, a database tool, a test site, or a monitoring dashboard. Every dashboard should require a strong login. If a page does not need to be on the public internet, keep it off the public internet. AI agents already look for things left in the open, as we have written about &lt;a href="https://security.hackerme.ai/exposed-api-keys-ai-agents/" rel="noopener noreferrer"&gt;how they hunt exposed API keys&lt;/a&gt; and about &lt;a href="https://security.hackerme.ai/gemini-agents-website-test/" rel="noopener noreferrer"&gt;Gemini agents hitting a website tens of thousands of times&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Did they really break in within 11 minutes?
&lt;/h3&gt;

&lt;p&gt;Amazon and BleepingComputer both describe an 11-minute window in which the agents broke in within 11 minutes, going from mapping the app's public interface to using a hole nobody had reported yet. Full control of the university server came in a later session, when they used that same hole again.&lt;/p&gt;

&lt;h3&gt;
  
  
  Did all of the AI attacks work?
&lt;/h3&gt;

&lt;p&gt;No. One attempt ran 148 actions over 16 hours and failed. Another tried more than 30,000 password combinations for about 80 real usernames over six hours and also failed. The successful case was the unknown flaw in the medical-imaging app.&lt;/p&gt;

&lt;h3&gt;
  
  
  How did Amazon see the attacks?
&lt;/h3&gt;

&lt;p&gt;The attackers left their Cairn dashboard open to the internet with no password. Amazon could read the prompts and the work history. That is also the simplest warning in the report for anyone who runs a dashboard of their own.&lt;/p&gt;

&lt;h3&gt;
  
  
  Is there a fix, and is there a CVE?
&lt;/h3&gt;

&lt;p&gt;The maintainers released a fix. BleepingComputer dates that release to 18 November 2026. Amazon says there was no public advisory at the time the agents found the hole. This article does not list a CVE or advisory number. If you run MONAI Label or any other self-hosted tool, update it and check what else on the server is reachable from the internet.&lt;/p&gt;

</description>
      <category>security</category>
      <category>ai</category>
      <category>webdev</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>Hacked Sites Are Showing Visitors a Fake Cloudflare Check. Would You Notice It on Yours?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Fri, 20 Nov 2026 08:14:37 +0000</pubDate>
      <link>https://dev.to/hackerme/hacked-sites-are-showing-visitors-a-fake-cloudflare-check-would-you-notice-it-on-yours-1acf</link>
      <guid>https://dev.to/hackerme/hacked-sites-are-showing-visitors-a-fake-cloudflare-check-would-you-notice-it-on-yours-1acf</guid>
      <description>&lt;p&gt;If your website runs on WordPress, this is a check worth doing this week. Researchers at Stormshield have described a still-active campaign in which hacked WordPress sites show Windows visitors a fake Cloudflare "verify you are human" page. That page copies a command and tells the visitor to paste and run it, which can install software that steals passwords and other data from their computer.&lt;/p&gt;

&lt;p&gt;Stormshield published the research on 19 October 2026. The Hacker News brought the same campaign up again on 20 November. So this is not a brand-new story today. It is still active.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;Attackers first get a script onto a WordPress site. Stormshield does not say how the sites were broken into, and there is no CVE number for this campaign.&lt;/p&gt;

&lt;p&gt;Once the script is there, it does two things that make it easy to miss:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;If you are logged in as a WordPress administrator, the script stays quiet. Stormshield says it looks for a typical logged-in WordPress cookie, or the black admin bar at the top of the page, and then it does nothing. You can look at your own site and see a normal page, while a customer on a Windows computer sees something else.&lt;/li&gt;
&lt;li&gt;For other Windows visitors, the script reads its next instructions from a record on a public blockchain, then shows a fake Cloudflare check. The fake page is written in several languages, including English, French, German, Spanish and Portuguese. It copies a command onto the computer and tells the person to paste and run it.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;If they do, the command can install a password-stealing program. Stormshield named that program from text found inside the file. This article does not repeat the command.&lt;/p&gt;

&lt;p&gt;Stormshield also found more than 70,000 websites compromised by EtherHiding-style campaigns in just a few days of searching. That figure covers this family of campaigns, not this one campaign alone. Stormshield does not say how many sites carry this particular script.&lt;/p&gt;

&lt;h2&gt;
  
  
  Who is affected
&lt;/h2&gt;

&lt;p&gt;Owners of WordPress sites, and anyone who visits those sites from a Windows computer. The fake page is not limited to famous brands. Any hacked WordPress site can become the trap that infects its own customers. Researchers have not published a public list of victim sites.&lt;/p&gt;

&lt;p&gt;A check from a phone, or while you are logged in to WordPress, may not show you the same page your Windows customers see.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why it matters for a small business
&lt;/h2&gt;

&lt;p&gt;Your website is often the most trusted page your customers open. If it shows a familiar "verify you are human" screen, many people will follow the steps. They will think they are helping your site load. In this campaign, they may instead be installing software that steals their passwords, including passwords they use with you.&lt;/p&gt;

&lt;p&gt;Because the script hides from a logged-in administrator, a site that looks clean to you is not proof that visitors see the same thing.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check today
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Look at your site the way a visitor does
&lt;/h3&gt;

&lt;p&gt;Open your website while logged out, or in a private window. Look for a "verify you are human" page, or any Cloudflare-style check, that you never added. If you can, also ask someone on a Windows computer to open the public site without signing in. If you see a check you did not put there, take the site offline if you can and call your developer or host.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Check theme and plugin files for scripts you do not recognise
&lt;/h3&gt;

&lt;p&gt;Ask your developer to review theme files, plugin files and any code that was added recently. Look for scripts you did not install. This is not a job for guesswork in a live admin screen alone, because a page that looks normal while you are logged in can still be unsafe for visitors.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Update or remove old plugins
&lt;/h3&gt;

&lt;p&gt;Old, unused or unpaid plugins are a common way attackers get a first foothold, even when researchers have not named the exact way this campaign starts. Our guide to &lt;a href="https://security.hackerme.ai/safe-wordpress-security-updates-how-to-rehearse-apply-and-harden-without-breaking-the-site/" rel="noopener noreferrer"&gt;safe WordPress security updates&lt;/a&gt; walks through a careful update. Plugin bugs are not rare: we have covered cases such as the &lt;a href="https://security.hackerme.ai/burst-statistics-vulnerability/" rel="noopener noreferrer"&gt;Burst Statistics plugin&lt;/a&gt; that needed a prompt update.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. Remember that a clean-looking page is not a clean bill of health
&lt;/h3&gt;

&lt;p&gt;A hacked homepage can show visitors a message you never wrote, as &lt;a href="https://security.hackerme.ai/hacked-homepage-hockey-federation/" rel="noopener noreferrer"&gt;the World Hockey Federation found&lt;/a&gt;. In this campaign, the fake check can be hidden from you and still be shown to customers. If something looks wrong, treat it as a real incident: change admin passwords, review users and plugins, and have a professional inspect the files.&lt;/p&gt;

&lt;p&gt;A &lt;a href="https://brandingme.ai/app/r/5LNfV0S4-UdO_GH5xcskLw" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see from the internet, without logging in. It is not a full cleanup, and it will not by itself prove a site is safe.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Is this a new attack from today?
&lt;/h3&gt;

&lt;p&gt;No. Stormshield published its analysis on 19 October 2026 and had been tracking the campaign since late September. The Hacker News wrote about it again on 20 November 2026. Treat it as still active, not as a story that started today.&lt;/p&gt;

&lt;h3&gt;
  
  
  Would I notice the fake Cloudflare check on my own site?
&lt;/h3&gt;

&lt;p&gt;Not necessarily. Stormshield says the script does not run when it detects a logged-in WordPress administrator. It also limited the fake page to Windows visitors and tried to ignore search-engine bots. Looking at the site while you are signed in, or only from a phone, is not enough.&lt;/p&gt;

&lt;h3&gt;
  
  
  Were 70,000 sites showing this exact page?
&lt;/h3&gt;

&lt;p&gt;Stormshield says it found more than 70,000 websites compromised by EtherHiding-style campaigns in a few days of searching. That number covers the family of campaigns, not this one campaign alone, and Stormshield does not say how many sites carry this particular script.&lt;/p&gt;

&lt;h3&gt;
  
  
  Is there a CVE or an official WordPress patch?
&lt;/h3&gt;

&lt;p&gt;No CVE has been published for this campaign, and researchers have not named the exact way the sites were first broken into. There is no single official patch that closes this story. The practical work is to inspect the public site while logged out, review theme and plugin files, and update or remove old plugins.&lt;/p&gt;

</description>
      <category>security</category>
      <category>wordpress</category>
      <category>webdev</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>OpenRouter Routed Some Chatbot Prompts to a Hacked Server for 18 Days. Were Your Passwords in Them?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Thu, 19 Nov 2026 19:40:59 +0000</pubDate>
      <link>https://dev.to/hackerme/openrouter-routed-some-chatbot-prompts-to-a-hacked-server-for-18-days-were-your-passwords-in-them-5e4c</link>
      <guid>https://dev.to/hackerme/openrouter-routed-some-chatbot-prompts-to-a-hacked-server-for-18-days-were-your-passwords-in-them-5e4c</guid>
      <description>&lt;p&gt;On 10 November 2026, Phala published a preliminary incident report. Phala is a company that OpenRouter sometimes sends chatbot requests to. The report says an attacker took over a server that was running the DeepSeek V4 Flash 0731 model. Anyone whose requests for that model were routed to Phala between 21 October 2026 at 00:00 UTC and 8 November 2026 at 05:35 UTC should treat the request contents as exposed, and should rotate any API keys, tokens, passwords or other secrets those requests contained.&lt;/p&gt;

&lt;p&gt;This is not a new leak from today. The report is about nine days old. The useful part now is checking whether your site's chatbot or AI app was in that window, and rotating anything secret that was in the prompts.&lt;/p&gt;

&lt;h2&gt;
  
  
  What Phala says happened
&lt;/h2&gt;

&lt;p&gt;Phala says the attacker reached internal management tools on a server on 21 October, and later took control of a new GPU server that was already serving live traffic. That gave the attacker a view of in-flight request data while the model was processing it. Phala says it took that server and the model offline, and that containment was completed on 8 November at 05:35 UTC. It has since restored the service with additional protections. A full root cause analysis has not yet been published.&lt;/p&gt;

&lt;p&gt;A third-party researcher had been planting unique test credentials, called canaries, in requests to Phala since 1 July. According to Phala, only canaries from requests sent on 21 and 22 October were later used by someone else. The earliest misuse Phala observed was on 22 October at 05:41 UTC, and the most recent was on 9 November. Phala says those tests sampled requests rather than watching every request, which is why it kept the wider 18-day window.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why you might have reached that server without choosing it
&lt;/h2&gt;

&lt;p&gt;Privacy Affairs, writing on 11 November, notes that OpenRouter is a gateway that sends a request to one of several companies hosting the same model. Unless you told it otherwise, it picks the provider for you. So you could have reached Phala without ever choosing it.&lt;/p&gt;

&lt;p&gt;Privacy Affairs also says this report concerns Phala's own deployment only. DeepSeek's own API, and the other companies serving that model through OpenRouter, are not part of it.&lt;/p&gt;

&lt;p&gt;Phala says the gateway is notifying affected users. Privacy Affairs wrote on 11 November that it found no public notice on OpenRouter's own website at the time.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why this matters if you run a site
&lt;/h2&gt;

&lt;p&gt;Site chatbots and AI-built apps often put API keys, customer details or internal notes inside prompts. A coding assistant can also pull secrets from files you did not mean to paste. You rarely see which company actually handled the request.&lt;/p&gt;

&lt;p&gt;We have written before about &lt;a href="https://security.hackerme.ai/docker-hub-leaked-keys/" rel="noopener noreferrer"&gt;how quickly exposed keys get used&lt;/a&gt; and about &lt;a href="https://security.hackerme.ai/ai-agent-permissions/" rel="noopener noreferrer"&gt;what to check when an AI assistant is connected to your tools&lt;/a&gt;. The same habit applies here: if a secret was in a prompt, treat it as known to someone else.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check today
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. See whether your chatbot or app used OpenRouter in that window
&lt;/h3&gt;

&lt;p&gt;If you have a site chatbot, a coding assistant or an app that sends prompts through OpenRouter, check whether it used the DeepSeek V4 Flash 0731 model between 21 October 2026 at 00:00 UTC and 8 November 2026 at 05:35 UTC. Privacy Affairs says OpenRouter's Activity page shows whether you used the model in that period, and that if your app logs OpenRouter generation IDs, the generation endpoint returns the provider name that served each request.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Rotate secrets that were in those prompts
&lt;/h3&gt;

&lt;p&gt;If your requests may have gone to Phala in that window, rotate every API key, token, password or other secret that appeared in the prompts. Include secrets a coding agent may have pulled in from your files, not only what you typed. Then review those accounts for activity you do not recognise, and contact the provider if you find any.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Keep secrets out of prompts from now on
&lt;/h3&gt;

&lt;p&gt;The safest secret is the one that never goes into a prompt. Keep API keys, passwords and customer records out of chatbot messages, and keep .env files out of the folders a coding assistant can read.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. See what a stranger can already reach
&lt;/h3&gt;

&lt;p&gt;A &lt;a href="https://brandingme.ai/app/r/_UEkx-I5vO7wE6A7EvGXGA" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see on your site from the internet, without logging in. It will not tell you which prompts were exposed, or which provider handled them. That check still has to be made by you.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Did my chatbot definitely leak my passwords?
&lt;/h3&gt;

&lt;p&gt;Only if those passwords, or other secrets, were in the prompts, and only if the requests for that model were routed to Phala in the window above. Phala asks affected users to treat any secrets in those request contents as compromised.&lt;/p&gt;

&lt;h3&gt;
  
  
  Was this a breach of DeepSeek itself?
&lt;/h3&gt;

&lt;p&gt;No. Privacy Affairs says DeepSeek's own API and the other companies serving this model through OpenRouter are not part of the report. The exposure Phala describes is limited to requests routed to Phala's deployment of that one model.&lt;/p&gt;

&lt;h3&gt;
  
  
  Why an 18-day window if only two days were confirmed?
&lt;/h3&gt;

&lt;p&gt;Phala says the 21 and 22 October dates come from canary evidence, but the tests sampled requests rather than monitoring every request, and it cannot yet limit the attacker's access to those two days. The wider window is precautionary.&lt;/p&gt;

&lt;h3&gt;
  
  
  Was this data sold on the dark web?
&lt;/h3&gt;

&lt;p&gt;No source we opened says the prompt data was sold or posted on the dark web. Phala does say that some of the researcher's canary credentials were later misused.&lt;/p&gt;

&lt;h3&gt;
  
  
  What should a small site owner do now?
&lt;/h3&gt;

&lt;p&gt;Check whether your chatbot or AI app used OpenRouter with that model in the window, rotate any secret that was in those prompts, review the related accounts, and keep secrets out of prompts from now on.&lt;/p&gt;

</description>
      <category>security</category>
      <category>webdev</category>
      <category>ai</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>A Stranger Could Reinstall Your Drupal Site and Take It Over. Is Yours Updated?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Thu, 19 Nov 2026 14:15:39 +0000</pubDate>
      <link>https://dev.to/hackerme/a-stranger-could-reinstall-your-drupal-site-and-take-it-over-is-yours-updated-500j</link>
      <guid>https://dev.to/hackerme/a-stranger-could-reinstall-your-drupal-site-and-take-it-over-is-yours-updated-500j</guid>
      <description>&lt;p&gt;Most site owners never think about the installer that set up their website. It ran once, years ago, and was forgotten. On 18 November 2026, the Drupal Security Team showed why that forgotten file matters: a critical flaw in Drupal core can let an anonymous attacker use the installer to try to reinstall the site. According to the advisory, that "can lead to a site takeover or remote code execution."&lt;/p&gt;

&lt;p&gt;If your website runs on Drupal, this Drupal security update should be on your developer's list today.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;Drupal core ships with a file called install.php, which sets up a new Drupal site. The Drupal Security Team's advisory, SA-CORE-2026-017, says that "under specific conditions, an attacker can use the installer to attempt to reinstall Drupal." No login is needed. The advisory rates the issue Critical (18 out of 25) and tracks it as CVE-2026-132165. It was found by Anthropic's OSS Scanner.&lt;/p&gt;

&lt;p&gt;The team had warned on Monday, two days before the release, that the issue was critical, that "at least one can be exploited by anonymous users," and that site owners should reserve time for updates "because exploits might be developed within hours or days," as BleepingComputer reported.&lt;/p&gt;

&lt;p&gt;The same release fixed a second, moderately critical issue (CVE-2026-132166, SA-CORE-2026-018), a "gadget chain" that could let someone who already has admin rights run code if another bug lets untrusted data be deserialized. That one needs an admin account, so the installer flaw is the urgent one.&lt;/p&gt;

&lt;p&gt;At the time of writing we found no report that the installer flaw is being exploited, and it was not in CISA's Known Exploited Vulnerabilities catalog. That can change quickly: CISA has listed several earlier Drupal flaws as exploited, and BleepingComputer notes that a Drupal SQL injection flaw was exploited earlier this year.&lt;/p&gt;

&lt;h2&gt;
  
  
  Who is affected
&lt;/h2&gt;

&lt;p&gt;According to the advisory, affected versions are:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Drupal 10 below 10.6.22&lt;/li&gt;
&lt;li&gt;Drupal 11.0 up to 11.3.21&lt;/li&gt;
&lt;li&gt;Drupal 11.4 below 11.4.13&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The fixed versions are 10.6.22, 11.3.22 and 11.4.13. Drupal 11.2.x and below, Drupal 10.5.x and below, and Drupal 8 and 9 are end-of-life and get no security fixes. If your site is on one of those, it will not get a patch at all.&lt;/p&gt;

&lt;p&gt;BleepingComputer describes Drupal as commonly used by government bodies, universities, research institutions and large media and enterprise sites. Smaller businesses often meet it through an agency that built their site years ago.&lt;/p&gt;

&lt;p&gt;Sites using the paid Drupal Steward firewall service are protected from the known attack paths, the advisory says, but should still update. Some exploit paths are also blocked if the web server cannot write to, or change permissions on, the settings.php file.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Update Drupal core
&lt;/h3&gt;

&lt;p&gt;Ask your developer or host to confirm the exact Drupal version and update to 10.6.22, 11.3.22 or 11.4.13 or later. Take a backup first. If you are on an end-of-life version, plan the upgrade now; there is no fix for you to apply.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Block the installer
&lt;/h3&gt;

&lt;p&gt;If you cannot update today, the Drupal Security Team recommends securing the installer. In practice, that means blocking access to /core/install.php in the web server configuration or in the .htaccess file in the site's web root, as BleepingComputer summarizes. It is a good idea to keep it blocked even after the update.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Lock down settings.php
&lt;/h3&gt;

&lt;p&gt;Make sure the web server cannot write to settings.php or change its permissions. The advisory says this mitigates some exploit paths.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. Look for signs of trouble
&lt;/h3&gt;

&lt;p&gt;After updating, check for new admin accounts, changed site settings, unexpected files, or a site that suddenly shows a setup screen. Our guide on &lt;a href="https://security.hackerme.ai/how-to-tell-if-your-website-has-been-hacked/" rel="noopener noreferrer"&gt;how to tell if your website has been hacked&lt;/a&gt; covers the common signs. For a real example of what a takeover looks like from the outside, see &lt;a href="https://security.hackerme.ai/hacked-homepage-hockey-federation/" rel="noopener noreferrer"&gt;how a hacked homepage forced the World Hockey Federation's site offline&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  5. Patch the layer underneath, too
&lt;/h3&gt;

&lt;p&gt;Drupal runs on PHP, which also shipped a critical security release this week. Our post on the &lt;a href="https://security.hackerme.ai/php-security-update-critical/" rel="noopener noreferrer"&gt;critical PHP security update&lt;/a&gt; explains what to ask your host.&lt;/p&gt;

&lt;h2&gt;
  
  
  The dark-web angle
&lt;/h2&gt;

&lt;p&gt;Site takeovers rarely end with the attacker. Access to a compromised website, especially one belonging to a university, government body or well-known brand, is often resold on criminal forums as "web access" or "shell access," used to host phishing pages, inject card skimmers, or spread malware to visitors. We have not seen listings tied to this flaw; this is how such access is generally used, and it is why a quick update is cheaper than a cleanup.&lt;/p&gt;

&lt;h2&gt;
  
  
  Know what your site shows the world
&lt;/h2&gt;

&lt;p&gt;A CMS update fixes one known hole. It does not tell you what else an outsider can see on your site. A &lt;a href="https://brandingme.ai/app/r/YUHPSce7zvwFOb28-A3FYA" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what is exposed from the internet, without logging in.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Is this Drupal flaw being exploited right now?
&lt;/h3&gt;

&lt;p&gt;At the time of writing we found no report of attacks, and it was not in CISA's Known Exploited Vulnerabilities catalog. The Drupal Security Team warned that exploits might be developed "within hours or days," so treat it as urgent.&lt;/p&gt;

&lt;h3&gt;
  
  
  Which Drupal versions fix the installer flaw?
&lt;/h3&gt;

&lt;p&gt;Drupal 10.6.22, 11.3.22 and 11.4.13. Older branches (11.2.x and below, 10.5.x and below, Drupal 8 and 9) are end-of-life and get no fix.&lt;/p&gt;

&lt;h3&gt;
  
  
  What can I do if I cannot update today?
&lt;/h3&gt;

&lt;p&gt;Block access to /core/install.php in your web server configuration or .htaccess file, and make sure the web server cannot write to settings.php. Then schedule the update as soon as possible.&lt;/p&gt;

&lt;p&gt;Sources: Drupal Security Team advisory SA-CORE-2026-017 (&lt;a href="https://www.drupal.org/sa-core-2026-017" rel="noopener noreferrer"&gt;https://www.drupal.org/sa-core-2026-017&lt;/a&gt;) and SA-CORE-2026-018 (&lt;a href="https://www.drupal.org/sa-core-2026-018" rel="noopener noreferrer"&gt;https://www.drupal.org/sa-core-2026-018&lt;/a&gt;); BleepingComputer, 19 Nov 2026 (&lt;a href="https://www.bleepingcomputer.com/news/security/drupal-patches-critical-rce-flaw-warns-exploits-may-surface-quickly/" rel="noopener noreferrer"&gt;https://www.bleepingcomputer.com/news/security/drupal-patches-critical-rce-flaw-warns-exploits-may-surface-quickly/&lt;/a&gt;).&lt;/p&gt;

&lt;p&gt;Disclosure: I'm with HackerMe.&lt;/p&gt;

</description>
      <category>security</category>
      <category>webdev</category>
      <category>drupal</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>A Plain Text File Could Let Attackers Run Code on Your Ruby Site. Have You Upgraded?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Thu, 19 Nov 2026 11:04:22 +0000</pubDate>
      <link>https://dev.to/hackerme/a-plain-text-file-could-let-attackers-run-code-on-your-ruby-site-have-you-upgraded-33cn</link>
      <guid>https://dev.to/hackerme/a-plain-text-file-could-let-attackers-run-code-on-your-ruby-site-have-you-upgraded-33cn</guid>
      <description>&lt;p&gt;If your website runs on Ruby, including many Rails apps, a security fix landed on 18 November 2026 that site owners should not treat as a routine patch. The Ruby team says it fixed seven memory-safety flaws in the regular-expression engine and in how Ruby changes the case of text, and that it knows of proof-of-concept code showing that some of these flaws can be used to run code remotely. Ruby Central adds an important detail for ordinary businesses: some of the bugs can be triggered by everyday text handling, such as a legacy-encoded email or an imported file, not only by an obviously hostile upload.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;On 18 November 2026 the Ruby project published a &lt;a href="https://www.ruby-lang.org/en/news/2026/11/18/multiple-vulnerabilities-in-regexp-and-string/" rel="noopener noreferrer"&gt;security advisory&lt;/a&gt; and released Ruby 4.0.10, 3.4.13 and 3.3.14. The same notice was shared on the &lt;a href="https://www.openwall.com/lists/oss-security/2026/11/19/1" rel="noopener noreferrer"&gt;oss-security mailing list&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;The advisory says: "Seven vulnerabilities have been discovered in Ruby's regular expression engine and in the case conversion of strings. We are aware of PoCs that prove some of them can be used to trigger remote code execution." It strongly recommends upgrading.&lt;/p&gt;

&lt;p&gt;Identifiers have been requested but had not been assigned when the advisory was published, so the Ruby team refers to each issue by a GitHub Security Advisory identifier. This article does not describe how the flaws work in practice.&lt;/p&gt;

&lt;h2&gt;
  
  
  Who is affected
&lt;/h2&gt;

&lt;p&gt;The official advisory says these versions are affected:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Ruby 4.0.0 through 4.0.9&lt;/li&gt;
&lt;li&gt;Ruby 3.4.0 through 3.4.12&lt;/li&gt;
&lt;li&gt;Ruby 3.3.13 and earlier&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The fixed versions are 4.0.10, 3.4.13 and 3.3.14.&lt;/p&gt;

&lt;p&gt;Ruby 3.3 is in the security maintenance phase. The advisory says Ruby 3.2 and earlier have reached end of life and will not receive these fixes. If a site is still on those older versions, the only path the Ruby team and Ruby Central describe is to move to a supported release that includes the fix.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://rubycentral.org/news/ruby-security-releases-upgrade-to-4-0-10-3-4-13-or-3-3-14/" rel="noopener noreferrer"&gt;Ruby Central&lt;/a&gt; says the official advisory is the source of truth and that it does not list a workaround: "Upgrading is the fix."&lt;/p&gt;

&lt;h2&gt;
  
  
  Why a plain text file is part of this story
&lt;/h2&gt;

&lt;p&gt;Four of the seven flaws involve regular expressions, which often matter most when an application builds patterns from untrusted input. Ruby Central says two others are different. They are memory-safety bugs in the case conversion of strings, and the advisory says that changing the case of a crafted string may allow remote code execution.&lt;/p&gt;

&lt;p&gt;Ruby Central adds that one of those case-conversion bugs applies to strings in some single-byte encodings, such as ISO-8859-1 and Windows-1252, and the other applies when the converted result exceeds 4 GiB. "If your application handles legacy-encoded text, such as email or imported files, an ordinary case conversion could be the trigger."&lt;/p&gt;

&lt;p&gt;That is the site-owner version of the problem: the risky input may look like a normal document or message, not like an attack tool. This article does not describe how to craft such a file.&lt;/p&gt;

&lt;p&gt;The seventh issue, the advisory says, can disclose memory contents or crash the process.&lt;/p&gt;

&lt;p&gt;None of the sources we read mention attacks using these flaws in the wild. That is not a reason to wait. Once a patch is public, the people who write attack tools can compare the old and new code.&lt;/p&gt;

&lt;h2&gt;
  
  
  What site owners should do
&lt;/h2&gt;

&lt;p&gt;This is not only a problem for teams that write Ruby themselves. It also applies to Rails apps, to hosts that install Ruby for you, and to container images that bundle a Ruby version.&lt;/p&gt;

&lt;p&gt;We explain the usual split in &lt;a href="https://security.hackerme.ai/website-security-what-your-host-handles-and-what-you-still-need-to-check/" rel="noopener noreferrer"&gt;what your host handles and what you still need to check&lt;/a&gt;. Today's &lt;a href="https://security.hackerme.ai/php-security-update-critical/" rel="noopener noreferrer"&gt;critical PHP security update&lt;/a&gt; is the same kind of question: who actually installs the new version on your server? If your Ruby comes from a container image, an outdated image can keep an old Ruby on a site that looks up to date in the control panel, a risk we also covered in a quieter way in &lt;a href="https://security.hackerme.ai/docker-hub-leaked-keys/" rel="noopener noreferrer"&gt;Docker Hub leaked keys&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  1. Ask which Ruby version is running
&lt;/h3&gt;

&lt;p&gt;Not the version in a document from last year. The version that is running on the live site.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Ask when it will be upgraded
&lt;/h3&gt;

&lt;p&gt;On a supported 4.0, 3.4 or 3.3 branch, that means 4.0.10, 3.4.13 or 3.3.14. On Ruby 3.2 or older, it means moving to a supported version that includes the fix.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Ask who will rebuild and redeploy
&lt;/h3&gt;

&lt;p&gt;If Ruby comes from a Linux package, a container image or a hosting platform, Ruby Central says to watch for an updated package or image, then rebuild and redeploy. A host that "will patch it" is not the same as a host that has already shipped a new image and restarted your app.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. See what strangers see
&lt;/h3&gt;

&lt;p&gt;An update fixes the language your site runs on. It does not tell you what else the site exposes. A &lt;a href="https://brandingme.ai/app/r/CGqrn-gbcqdpG_5fl2t9aA" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see from the internet, without logging in.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Does this affect WordPress sites?
&lt;/h3&gt;

&lt;p&gt;WordPress runs on PHP, not Ruby. This update matters if you also have a Ruby or Rails app, a tool built in Ruby, or a host or container that ships Ruby.&lt;/p&gt;

&lt;h3&gt;
  
  
  Is there a temporary setting I can turn off until we upgrade?
&lt;/h3&gt;

&lt;p&gt;The official advisory does not list a workaround. Ruby Central says upgrading is the fix.&lt;/p&gt;

&lt;h3&gt;
  
  
  Were these flaws already used in attacks?
&lt;/h3&gt;

&lt;p&gt;None of the sources we read mention attacks in the wild. The Ruby team does say it knows of proof-of-concept code showing remote code execution in some scenarios, which is why the upgrade is urgent.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;The original version of this article is on the &lt;a href="https://security.hackerme.ai/ruby-regexp-string-flaws/" rel="noopener noreferrer"&gt;HackerMe Security blog&lt;/a&gt;.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>security</category>
      <category>webdev</category>
      <category>ruby</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>London's Transport Site Leaked 265,000 People's Security Answers. Would You Spot the Warning Sign on Yours?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Thu, 19 Nov 2026 10:50:19 +0000</pubDate>
      <link>https://dev.to/hackerme/londons-transport-site-leaked-265000-peoples-security-answers-would-you-spot-the-warning-sign-5h44</link>
      <guid>https://dev.to/hackerme/londons-transport-site-leaked-265000-peoples-security-answers-would-you-spot-the-warning-sign-5h44</guid>
      <description>&lt;p&gt;The first warning was an email nobody had asked for. On 16 November 2026, Transport for London (TfL) says Oyster photocard customers started receiving messages from TfL's own system asking them to create an account they already had. TfL later said about 265,000 customers had personal details accessed, including the security questions and answers used to prove who they are. If you run a customer portal, that kind of unexpected system email is worth treating as an alarm, not as a glitch.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;According to The Independent, TfL was alerted on Monday, 16 November, after cardholders reported those unexpected "create an account" emails. Shashi Verma, TfL's director of technology strategy and revenue, said: "On 16 November 2026, we became aware that Oyster photocard customers were receiving emails asking them to create an account, when they already have a photocard account. We immediately took the TfL photocard website down while this was being investigated."&lt;/p&gt;

&lt;p&gt;In notices sent on Wednesday, 18 November, and in statements carried by the &lt;a href="https://www.bbc.co.uk/news/articles/c6n4kz3lgdz5o" rel="noopener noreferrer"&gt;BBC&lt;/a&gt;, &lt;a href="https://www.independent.co.uk/news/uk/home-news/oyster-card-tfl-apology-data-breach-b3087538.html" rel="noopener noreferrer"&gt;The Independent&lt;/a&gt; and &lt;a href="https://www.telegraph.co.uk/news/2026/11/18/oyster-cards-cyber-attack-london-underground/" rel="noopener noreferrer"&gt;The Telegraph&lt;/a&gt;, TfL said this activity led to unauthorised access to personal information in web accounts. The BBC reports that TfL puts the figure at about 265,000 Oyster photocard customers. Verma said TfL now believes "around 265,000 photocard customers may have had their account details compromised as part of this incident."&lt;/p&gt;

&lt;p&gt;TfL has not said how the data was accessed, and it said it is still investigating. We are not going to guess at the method.&lt;/p&gt;

&lt;h2&gt;
  
  
  What was accessed, and what was not
&lt;/h2&gt;

&lt;p&gt;TfL says the accessed details include names, email addresses, home addresses, phone numbers, dates of birth, and security questions and answers.&lt;/p&gt;

&lt;p&gt;TfL also says its current assessment is that passwords, bank or payment card details, and journey history were not compromised. It added that the names and dates of birth of children with a Zip Oyster card were not affected.&lt;/p&gt;

&lt;p&gt;The Telegraph reports that TfL has urged all photocard customers to change their passwords and security questions as a precaution, even though TfL says passwords were not compromised. The photocard website remains offline, so customers cannot apply for or renew a photocard online until it is restored. TfL says customers can still travel with their photocards, and The Independent reports they can still top up at ticket machines.&lt;/p&gt;

&lt;p&gt;The Independent says TfL has officially reported the incident to the Information Commissioner's Office (ICO), the UK privacy regulator. Verma said TfL has "now contained the incident" and has already contacted affected customers.&lt;/p&gt;

&lt;p&gt;The affected account types, according to TfL as reported by The Independent and The Telegraph, include Zip Oyster cards, 18+ Student, 18-25 Care Leaver, 60+ London, Apprentice and Veterans Oyster photocards.&lt;/p&gt;

&lt;h2&gt;
  
  
  This is not the first time this system has been taken down
&lt;/h2&gt;

&lt;p&gt;The BBC says this is the second time in just over two years that the photocard system was taken down because of a cyber incident. In September 2024, a cyber attack forced TfL to suspend new photocard applications for more than two months. That was a separate incident. TfL has not said the two events are connected, and we are not treating them as the same attack.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why this matters if you run a customer portal
&lt;/h2&gt;

&lt;p&gt;Most small businesses will never run a city transport site. They do run login pages, account portals, and "create an account" or "reset your password" emails. The useful lesson is the warning sign TfL says it saw first: a burst of system emails that customers did not request.&lt;/p&gt;

&lt;p&gt;Security questions and answers are a poor second lock once they leak. Unlike a password, a mother's maiden name or a first-school answer is often reused on other sites and cannot really be changed. If those answers are part of how customers log in or recover an account, a stronger method, such as a one-time code or an authenticator app, is the better choice.&lt;/p&gt;

&lt;p&gt;Unexpected account-creation or password-reset emails can also be the first visible sign that someone is abusing a customer portal, even when the emails come from your own system. That is why volume matters: a sudden jump in messages nobody asked for is worth investigating the same day.&lt;/p&gt;

&lt;p&gt;We have written before about how hard it is to &lt;a href="https://security.hackerme.ai/credit-union-breach-access-control-logs/" rel="noopener noreferrer"&gt;tell which customer records were accessed&lt;/a&gt; after an incident, and about what can go wrong when &lt;a href="https://security.hackerme.ai/password-reset-form-dooray/" rel="noopener noreferrer"&gt;a password-reset form is left weaker than the login page&lt;/a&gt;. The split between &lt;a href="https://security.hackerme.ai/website-security-what-your-host-handles-and-what-you-still-need-to-check/" rel="noopener noreferrer"&gt;what your host handles and what you still need to check&lt;/a&gt; is the same here: the portal is your system, even if a vendor helps run it.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check on your own site
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Watch the emails your own system sends
&lt;/h3&gt;

&lt;p&gt;Ask your developer or email provider for a daily count of account-creation, "verify your email" and password-reset messages. If that number jumps and customers start reporting emails they did not request, treat it as an incident, not as a mail-server glitch.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Look at how customers recover a lost login
&lt;/h3&gt;

&lt;p&gt;If your site still uses security questions, ask whether they can be replaced with a one-time code or an authenticator app. Once those answers leak, they are hard to take back.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Know what you would be able to say after a breach
&lt;/h3&gt;

&lt;p&gt;If someone accessed customer records tomorrow, could you say which accounts were touched, and which fields? If the answer is no, that is a logging and access-control problem to fix before you need it.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. See what strangers see
&lt;/h3&gt;

&lt;p&gt;An investigation can close a hole in a portal. It does not tell you what else the site exposes. A &lt;a href="https://brandingme.ai/app/r/k904yOip5d7jAld2Lpeppg" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see from the internet, without logging in.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Were bank cards or travel history taken?
&lt;/h3&gt;

&lt;p&gt;TfL says its current assessment is that passwords, bank or payment card details, and journey history were not compromised, and that children's names and dates of birth on Zip Oyster cards were not affected. It is still investigating.&lt;/p&gt;

&lt;h3&gt;
  
  
  How did this happen?
&lt;/h3&gt;

&lt;p&gt;TfL has not said. The public reports describe unauthorised access after suspicious activity, not a method. We are not filling in that gap.&lt;/p&gt;

&lt;h3&gt;
  
  
  What if I am not in London and I do not have an Oyster card?
&lt;/h3&gt;

&lt;p&gt;The useful part for a site owner is the warning sign and the leaked security answers. If your business sends account emails or stores security questions, the same checks apply.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;The original version of this article is on the &lt;a href="https://security.hackerme.ai/tfl-oyster-photocard-breach/" rel="noopener noreferrer"&gt;HackerMe Security blog&lt;/a&gt;.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>security</category>
      <category>webdev</category>
      <category>privacy</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>Hackers Broke Into the Spam Filter Your Web Host May Use. Do You Know Who Guards Your Email?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Thu, 19 Nov 2026 10:37:36 +0000</pubDate>
      <link>https://dev.to/hackerme/hackers-broke-into-the-spam-filter-your-web-host-may-use-do-you-know-who-guards-your-email-14o3</link>
      <guid>https://dev.to/hackerme/hackers-broke-into-the-spam-filter-your-web-host-may-use-do-you-know-who-guards-your-email-14o3</guid>
      <description>&lt;p&gt;If your business email is cleaned by a filter your web host or IT provider set up, you may never have seen the brand name behind it. On 19 November 2026, The Hacker News reported that N-able had confirmed unauthorized access to one of the servers in its hosted email-filtering cloud, the service behind products called Mail Assure and SpamExperts. The management dashboards have been offline in every region. Mail itself is still being delivered and filtered.&lt;/p&gt;

&lt;p&gt;According to N-able's own status page (incident 202672), the company first told customers about a disruption on 16 November 2026, and first described it as a problem with platform configuration and service dependencies. It later said the disruption was caused by a security incident.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;N-able's 18 November update, as recorded on its status page and quoted by The Hacker News, says: "We are aware of unauthorized access to one of the servers in the Mail Assure and SpamExperts Hosted Cloud." It also says: "The investigation is ongoing, but at this time we do not have evidence that customer data was compromised, including access to tenants' mailboxes."&lt;/p&gt;

&lt;p&gt;The same N-able update says this is "unrelated to the N-central incident." The Hacker News adds that the event is separate from the N-central vulnerabilities referred to as Able2Shell, 4EverAble2Shell, and Agent2Shell.&lt;/p&gt;

&lt;p&gt;UtiBox, writing on 18 November 2026, notes that N-able sells these hosted filters mostly through IT providers and hosting companies. A site owner can depend on one of them without ever logging into N-able's own dashboard.&lt;/p&gt;

&lt;h2&gt;
  
  
  What still works, and what does not
&lt;/h2&gt;

&lt;p&gt;N-able and The Hacker News both say inbound and outbound mail flow and filtering have continued to operate. Messages should still be arriving, and the filter should still be checking them.&lt;/p&gt;

&lt;p&gt;What has been offline, according to N-able's status updates, is the management side: the portal or dashboard, the APIs, quarantine, logging, and mailbox synchronization. Until those come back, N-able says customers may be unable to sign in, view or release quarantined mail, review logs, or add or remove domains. The Hacker News lists the portal, APIs, quarantine, and logging as still offline, and reports that as of 19 November 2026 at 8 a.m. UTC, restoration work was still under way. A matching 8 a.m. UTC status update, mirrored by the status tracker Pingoru, says portal and API access remain disabled for now, while core mail flow and filtering are still operating normally.&lt;/p&gt;

&lt;p&gt;The six regional services named on the status page and by The Hacker News are Mail Assure and SpamExperts in the Americas, APAC, and Europe — every region where those products are hosted.&lt;/p&gt;

&lt;p&gt;N-able says it has contacted Mail Assure and SpamExperts Hosted Cloud partners directly by email "with information relevant to their environment and whether action is required or not."&lt;/p&gt;

&lt;h2&gt;
  
  
  Why this matters if you run a site
&lt;/h2&gt;

&lt;p&gt;A filter that sits in front of your domain's email can be easy to forget, because it is often sold and managed by someone else. If your host or IT provider uses Mail Assure or SpamExperts, the locked dashboard may already be affecting you even if you have never heard those names.&lt;/p&gt;

&lt;p&gt;UtiBox's practical warning is worth taking seriously: while quarantine is down, some held messages may not be releasable, and incidents like this attract phishing. Fake "reset your password" or "release your quarantine" emails are a real risk. N-able has not reported that customer mailboxes were accessed, and this article does not claim that data was stolen. The investigation is still open.&lt;/p&gt;

&lt;p&gt;UtiBox also cautions against a rushed change to mail routing. N-able says mail flow is still working, and a hasty change to MX records can cause the outage you are trying to avoid.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check today
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Ask who runs your email filter
&lt;/h3&gt;

&lt;p&gt;Ask your web host or IT provider whether your domain's filtering runs on Mail Assure or SpamExperts, and whether N-able's partner notice calls for any action on your account. If you manage the filter yourself, watch N-able's status page and any email you have already received from N-able or your provider. Ignore unofficial "fix" instructions.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Plan for quarantined mail you cannot reach
&lt;/h3&gt;

&lt;p&gt;Until the dashboard and quarantine tools are restored, you may be unable to view or release held messages. Tell anyone who depends on that inbox that a legitimate message could be stuck, and agree how urgent mail will be handled in the meantime.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Be extra careful with phishing
&lt;/h3&gt;

&lt;p&gt;Treat unexpected emails about this incident with suspicion, including messages that ask you to sign in, reset a password, or release quarantine. Use only the sign-in address you already use, and only once N-able or your provider says access is back. Our earlier report on a &lt;a href="https://security.hackerme.ai/password-reset-form-dooray/" rel="noopener noreferrer"&gt;guessed password-reset form&lt;/a&gt; is a reminder that account recovery pages are a common target.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. See what strangers can see
&lt;/h3&gt;

&lt;p&gt;A third-party email filter is only one part of the setup in front of your site. A &lt;a href="https://brandingme.ai/app/r/rl2r4P_7GRX8_GyBAK7FDg" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see from the internet, without logging in. If you want a wider checklist, use &lt;a href="https://security.hackerme.ai/how-to-secure-a-wordpress-website/" rel="noopener noreferrer"&gt;how to secure a WordPress website&lt;/a&gt;, and our guide to &lt;a href="https://security.hackerme.ai/web-hosting-ransomware-backups/" rel="noopener noreferrer"&gt;what web-host incidents have taught site owners&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Did my email stop arriving?
&lt;/h3&gt;

&lt;p&gt;N-able says no. According to its status updates and to The Hacker News, mail flow and inbound and outbound filtering have continued to operate. The outage is on the management side, not on message delivery.&lt;/p&gt;

&lt;h3&gt;
  
  
  Was customer data stolen?
&lt;/h3&gt;

&lt;p&gt;N-able says it has "no evidence that customer data was compromised, including access to tenants' mailboxes." That is a statement about what it has found so far, not a final finding. The investigation is ongoing. This article does not say that data was stolen, and no customer counts have been published.&lt;/p&gt;

&lt;h3&gt;
  
  
  How do I know if I use this filter?
&lt;/h3&gt;

&lt;p&gt;The product names are Mail Assure and SpamExperts, sold by N-able, usually through a host or an IT provider. If you are not sure, ask the company that set up your domain's email. Do not assume you are unaffected just because you have never logged into an N-able dashboard.&lt;/p&gt;

&lt;h3&gt;
  
  
  Is this the same as the N-central vulnerabilities?
&lt;/h3&gt;

&lt;p&gt;No. N-able says the incident is unrelated to a separate N-central incident. The Hacker News reports that it is also separate from the N-central vulnerabilities referred to as Able2Shell, 4EverAble2Shell, and Agent2Shell.&lt;/p&gt;




&lt;p&gt;Disclosure: I'm with HackerMe.&lt;/p&gt;

</description>
      <category>security</category>
      <category>webdev</category>
      <category>email</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>A Checkout Plugin Can Hand a Stranger Your Store's Admin Login. Is It on Your Site?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Thu, 19 Nov 2026 10:19:25 +0000</pubDate>
      <link>https://dev.to/hackerme/a-checkout-plugin-can-hand-a-stranger-your-stores-admin-login-is-it-on-your-site-1knk</link>
      <guid>https://dev.to/hackerme/a-checkout-plugin-can-hand-a-stranger-your-stores-admin-login-is-it-on-your-site-1knk</guid>
      <description>&lt;p&gt;If your WooCommerce store takes payments through a plugin called Unzer Payments, a new security advisory is worth acting on today. According to a Wordfence advisory published on 18 November 2026, a checkout function in that plugin can log a visitor in as the matching WordPress user — including the store administrator — using only an email address. No password is required, and the advisory says the plugin does not check that a real payment was made. That would mean full control of the store.&lt;/p&gt;

&lt;p&gt;This report comes from a single advisory chain. Wordfence published the details, and public records such as the GitHub Advisory Database and NVD repeat the same wording. As of 19 November 2026, we have not found a second, independent technical investigation or coverage in a news outlet. The facts below are presented as Wordfence describes them.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;Wordfence says the flaw is in Unzer Payments for WooCommerce, in all versions up to and including 2.2.1, and tracks it as CVE-2026-110009, with a critical severity score of 9.8. According to the advisory, the plugin's Google Pay Express Checkout path accepts an email address and then signs the visitor in as the WordPress account that uses that email, administrators included.&lt;/p&gt;

&lt;p&gt;Wordfence lists the plugin as unpatched. Its advice is that, with no known fix, it may be best to uninstall the plugin and switch to a replacement. We checked the vendor's public GitHub releases for the same plugin: the latest release shown is still 2.2.1, dated 16 September 2026. We did not find a vendor security notice from Unzer.&lt;/p&gt;

&lt;p&gt;None of the official records we reviewed report that this flaw has been used in attacks. A CISA assessment attached to the NVD entry also lists exploitation as none. That can change once a flaw is public, so it is not a reason to wait.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why this matters if you run a store
&lt;/h2&gt;

&lt;p&gt;"They would need my email" is not much of a barrier. Administrator and shop-manager emails often appear on the public site, in author pages, or in contact details. If the advisory is right, that address alone could be enough to take over the store: change settings, reach customer and order data, or install other software.&lt;/p&gt;

&lt;p&gt;WordPress.org does not publish an active-install figure for this plugin (it shows N/A), so there is no reliable number of stores to cite.&lt;/p&gt;

&lt;p&gt;Separately, the plugin's page on WordPress.org says it was closed on 11 November 2026 and is not available for download, pending a full review. WordPress.org does not give a reason. That closure should not be treated as confirmation of this advisory, which was published a week later.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check today
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. See whether the plugin is installed
&lt;/h3&gt;

&lt;p&gt;In WordPress, open Plugins and look for "Unzer Payments". If you do not manage the site yourself, ask your developer or agency. Also check with anyone who set up checkout, because a payment plugin can be installed and then forgotten.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. If it is installed, plan a replacement
&lt;/h3&gt;

&lt;p&gt;According to Wordfence, there is no known patch. Do not wait for an update that may not exist yet. Arrange a replacement payment method first so checkout does not simply stop, then deactivate and uninstall Unzer Payments. Take a backup before you change payment plugins, and test checkout afterwards, as in our guide to &lt;a href="https://security.hackerme.ai/safe-wordpress-security-updates-how-to-rehearse-apply-and-harden-without-breaking-the-site/" rel="noopener noreferrer"&gt;safe WordPress security updates&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Review who can get into the store
&lt;/h3&gt;

&lt;p&gt;If this plugin is, or was, on the site, ask your developer to review administrator and shop-manager accounts, recent logins, and any unexpected users or plugins. Our guide to &lt;a href="https://security.hackerme.ai/how-to-tell-if-your-website-has-been-hacked/" rel="noopener noreferrer"&gt;how to tell if your website has been hacked&lt;/a&gt; covers the practical signs. Wordfence has not reported attacks, but a review is still reasonable.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. See what strangers see
&lt;/h3&gt;

&lt;p&gt;Removing one plugin does not tell you what else the site exposes. A &lt;a href="https://brandingme.ai/app/r/kC9tCuGP0LG-GIZy4fDb2g" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see from the internet, without logging in. You can also use the checklist in &lt;a href="https://security.hackerme.ai/how-to-secure-a-wordpress-website/" rel="noopener noreferrer"&gt;how to secure a WordPress website&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Does this affect every WooCommerce store?
&lt;/h3&gt;

&lt;p&gt;No. According to the advisory, it affects stores that have the Unzer Payments plugin installed, in versions up to 2.2.1. Other payment plugins are a separate question. If you are not sure which checkout plugin you use, check the Plugins screen or ask your developer.&lt;/p&gt;

&lt;h3&gt;
  
  
  Is there a patch I can install?
&lt;/h3&gt;

&lt;p&gt;Wordfence lists no known patch. The latest public vendor release we found is still 2.2.1. Until a fixed version is published and checked, Wordfence's suggested path is to uninstall the plugin and use a replacement.&lt;/p&gt;

&lt;h3&gt;
  
  
  If I turn off Google Pay, am I safe?
&lt;/h3&gt;

&lt;p&gt;The advisory does not say whether the affected checkout path is still reachable when that option is switched off in the plugin's settings. Do not assume that a disabled setting is enough. The safer step, according to Wordfence, is to remove the plugin.&lt;/p&gt;

&lt;h3&gt;
  
  
  Has this been used in attacks?
&lt;/h3&gt;

&lt;p&gt;None of the official records we reviewed report exploitation, and a CISA assessment on the NVD entry lists exploitation as none. That is not a guarantee. Treat it as urgent if the plugin is on your site.&lt;/p&gt;




&lt;p&gt;Disclosure: I'm with HackerMe.&lt;/p&gt;

</description>
      <category>security</category>
      <category>webdev</category>
      <category>wordpress</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>Attackers Are Already Knocking: 150+ Break-In Attempts in a Day. Is Your Site's Java Patched?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Thu, 19 Nov 2026 10:02:38 +0000</pubDate>
      <link>https://dev.to/hackerme/attackers-are-already-knocking-150-break-in-attempts-in-a-day-is-your-sites-java-patched-4ek9</link>
      <guid>https://dev.to/hackerme/attackers-are-already-knocking-150-break-in-attempts-in-a-day-is-your-sites-java-patched-4ek9</guid>
      <description>&lt;p&gt;If your site or internal tools run on Java — including self-hosted software your agency or host may have installed for you — attackers have started probing a serious flaw in Java itself. The attempts are real. A successful break-in has not been confirmed.&lt;/p&gt;

&lt;p&gt;On 19 November 2026, SecurityWeek reported that GreyNoise had recorded more than 150 requests shaped like this attack, coming from five IP addresses, as of early that day. GreyNoise said it had not seen any successful intrusions on its sensors. That 150-plus figure is GreyNoise's, as reported by SecurityWeek.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;Security firm Calif disclosed the flaw, often called Java4Shell (CVE-2026-114938), on 17 November 2026. Oracle and the OpenJDK project released patches the same day. CISA urged organizations to patch immediately and to check inventories for Java, including copies bundled inside third-party applications. Calif described a memory bug in how newer Java builds decode a common data format, and said that on some systems an attacker would not need a password. This article does not include exploit steps.&lt;/p&gt;

&lt;p&gt;What is new is the probing. Besides the GreyNoise count, SecurityWeek quoted watchTowr as saying attackers are already spraying the internet and hitting its honeypots, including blind attempts with correctly shaped requests. Jake Knott, head of threat intelligence at watchTowr, said the activity so far is not particularly functional, but that over the coming days attackers are expected to refine more reliable, product-specific exploits. SC Media published the same watchTowr quotes on 18 November 2026.&lt;/p&gt;

&lt;p&gt;SecurityWeek also reported that Cloudflare's Cloudforce One team saw a scanner probe a target with a user agent that clearly referenced the flaw, then return roughly 17 hours later with the exploit trigger while impersonating the Chrome browser.&lt;/p&gt;

&lt;p&gt;A separate firm, Previdian, told SecurityWeek its sensors logged nine exploitation attempts from six IP addresses on 18 and 19 November, and that this data cannot confirm whether any target was compromised. Those nine attempts are a separate sensor count. They should not be added to GreyNoise's 150-plus figure.&lt;/p&gt;

&lt;h2&gt;
  
  
  This is not a confirmed break-in
&lt;/h2&gt;

&lt;p&gt;The accurate words are "attempts" and "probing," not "confirmed exploited." Tenable's FAQ, updated on 18 November 2026, says the flaw is not listed in CISA's Known Exploited Vulnerabilities catalog. A direct check of that catalog on 19 November 2026 (catalog version 2026.11.18, released 18 November) also does not list CVE-2026-114938. GreyNoise and Previdian both say they have not confirmed a successful intrusion. Cisco told SecurityWeek it has no knowledge of malicious use.&lt;/p&gt;

&lt;p&gt;SecurityWeek and Tenable report that VulnCheck has a full exploit available only to its customers. SC Media quotes Adrian Culley of SafeBreach as saying unverified exploit code is appearing on GitHub. Tenable and SecurityWeek also note that Kevin Breen of Immersive has reported building a working exploit; those reports do not say it has been released to the public. None of that is the same as a confirmed intrusion.&lt;/p&gt;

&lt;h2&gt;
  
  
  What this means if you run Java tools
&lt;/h2&gt;

&lt;p&gt;For a typical small-business site, the practical question is whether any self-hosted Java-based tools are on a patched Java build. Examples of Java software businesses often run include Jira, Confluence, Jenkins and Keycloak. A confirmed break-in of those products has not been reported in the sources we reviewed. SecurityWeek does report that Atlassian, the Jenkins project and the Keycloak team have started issuing their own customer guidance. Atlassian, according to SecurityWeek, said cloud customers do not need to take action, while self-hosted and Data Center products are a separate question until patches are available.&lt;/p&gt;

&lt;p&gt;SecurityWeek also reports that Cisco has confirmed more than a dozen of its products are affected, with no fixes and no workarounds yet, and that Siemens says three of its Industrial Edge Management products are affected and recommends taking those instances offline while it works on patches. Those vendor notes matter most if you run those products. For most site owners, the urgent step is still the Java runtime itself.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check today
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Confirm the Java build is patched
&lt;/h3&gt;

&lt;p&gt;Ask your host, developer or agency whether every self-hosted Java tool is running a patched Java build from the November 2026 releases. Then make sure the service was restarted, because a patch only takes effect after the Java process restarts. If you use a hosted version of a tool rather than a server you manage, ask the vendor or host whether their Java runtime is already patched.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. If a patch cannot be applied today, use the vendor workaround
&lt;/h3&gt;

&lt;p&gt;Official advisories describe a temporary Java startup setting that turns off the affected fast path until the patch can be installed. Ask your host or agency to apply the vendor's instructions and restart the service. A workaround is not a substitute for the patch.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Keep admin consoles off the open internet until they are patched
&lt;/h3&gt;

&lt;p&gt;Researchers quoted by SC Media advise taking identity, build and collaboration consoles off the public internet, or putting them behind a VPN, until they are patched.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. See what strangers can see
&lt;/h3&gt;

&lt;p&gt;Patching one runtime does not tell you what else the site exposes from the internet. A &lt;a href="https://brandingme.ai/app/r/P32jFNw6V2jJDLfFiBEHXQ" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see, without logging in. Our guide to &lt;a href="https://security.hackerme.ai/how-to-tell-if-your-website-has-been-hacked/" rel="noopener noreferrer"&gt;how to tell if your website has been hacked&lt;/a&gt; covers practical signs to review if you are worried something already went wrong. You can also use the checklist in &lt;a href="https://security.hackerme.ai/how-to-secure-a-wordpress-website/" rel="noopener noreferrer"&gt;how to secure a WordPress website&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Does this affect every website?
&lt;/h3&gt;

&lt;p&gt;No. It depends on whether the site or its tools run an affected Java build, and on the kind of processor underneath. The OpenJDK advisory marks newer Java 17, 21, 25 and 27 builds for this flaw, and does not mark standard Java 8 or 11. ARM-based systems are not affected. If you are not sure what you run, ask your host or developer.&lt;/p&gt;

&lt;h3&gt;
  
  
  Has this been used in a successful attack?
&lt;/h3&gt;

&lt;p&gt;No successful intrusion has been confirmed in the sources we reviewed. GreyNoise and Previdian both report attempts without a confirmed compromise, and the flaw is not on CISA's known-exploited list. That is not a reason to delay a patch. Experts quoted by SecurityWeek expect more reliable, product-specific exploits within days.&lt;/p&gt;

&lt;h3&gt;
  
  
  I only use a hosted tool. Do I need to do anything?
&lt;/h3&gt;

&lt;p&gt;Ask the vendor or host whether their Java runtime is patched. According to SecurityWeek, Atlassian said its cloud customers do not need to take action. Self-hosted and Data Center installations are a different case: those need a patched Java build, or the vendor's temporary workaround, and a restart.&lt;/p&gt;

</description>
      <category>security</category>
      <category>webdev</category>
      <category>java</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>The Code Behind About 70% of Websites Gets a Critical Security Fix Today. Will Your Site Be Updated?</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Thu, 19 Nov 2026 09:19:00 +0000</pubDate>
      <link>https://dev.to/hackerme/the-code-behind-about-70-of-websites-gets-a-critical-security-fix-today-will-your-site-be-updated-jld</link>
      <guid>https://dev.to/hackerme/the-code-behind-about-70-of-websites-gets-a-critical-security-fix-today-will-your-site-be-updated-jld</guid>
      <description>&lt;p&gt;If your website runs on WordPress, WooCommerce, Magento or Laravel, it runs on PHP. Today, 19 November 2026, between 14:00 and 14:30 UTC (16:00–16:30 in Israel), the PHP project releases security fixes for every supported version, and it rates the most serious flaws as critical. The practical step for a site owner is simple: ask your hosting company or developer today when the new PHP version will be installed on your site, and who will check that the site still works afterwards.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;On 18 November 2026 the PHP security team posted an advance notice on the &lt;a href="https://www.php.net/archive/2026.php" rel="noopener noreferrer"&gt;official PHP website&lt;/a&gt;. It said the regular four-week release would "address security vulnerabilities affecting all supported PHP versions" and that "the highest severity is CRITICAL."&lt;/p&gt;

&lt;p&gt;The same notice adds an important limit: the critical issues "require either non-default configuration or uncommon setups." PHP said that the technical details and the fixed version numbers would be published together with the releases, not before. The notice was also shared on the oss-security mailing list.&lt;/p&gt;

&lt;p&gt;The Register reports that PHP runs on about 70 percent of websites. That figure is The Register's, not a count we made ourselves.&lt;/p&gt;

&lt;h2&gt;
  
  
  The one flaw that is already known
&lt;/h2&gt;

&lt;p&gt;&lt;a href="https://thehackernews.com/2026/11/php-warns-of-critical-security-flaws.html" rel="noopener noreferrer"&gt;The Hacker News&lt;/a&gt; and The Register both expect the release to include CVE-2026-115560, a remote code execution flaw in PHP's session upload progress feature. It only works when two things are true:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;the application's settings use the &lt;code&gt;php_binary&lt;/code&gt; or &lt;code&gt;php_serialize&lt;/code&gt; session handler, and its code switches the handler to &lt;code&gt;php&lt;/code&gt; at runtime before starting the session; and&lt;/li&gt;
&lt;li&gt;session upload progress is turned on.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The PHP Foundation told The Register in October that this combination is very unlikely in practice. Its interim advice was to set &lt;code&gt;session.upload_progress.enabled=0&lt;/code&gt; and restart PHP until the fix arrived. If your developer has not done that, today's update is the cleaner fix.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why site owners should still pay attention
&lt;/h2&gt;

&lt;p&gt;"Uncommon setup" is not the same as "no risk." Most small businesses never look at their PHP settings, so they cannot know whether their site is the uncommon one.&lt;/p&gt;

&lt;p&gt;There is also recent history. In October, PHP fixed two other critical flaws known as PHP2Shell (CVE-2026-109618 and CVE-2026-109619). According to The Register, CISA added them to its Known Exploited Vulnerabilities catalog on 26 October, and The Hacker News reports limited active exploitation against exposed management interfaces. Hunt.io found hundreds of exploit-related files, and people on underground forums were asking for working exploits. That is how a fixed bug becomes a product: once a patch is public, attackers compare old and new code, build tools, and trade or sell access to the servers they get into.&lt;/p&gt;

&lt;p&gt;None of the sources we read report attacks on the new flaws in today's release.&lt;/p&gt;

&lt;h2&gt;
  
  
  Who updates PHP on your site?
&lt;/h2&gt;

&lt;p&gt;This is the question most owners cannot answer. On shared or managed hosting, the host usually installs PHP updates, but often only for the PHP versions it still supports. On a VPS or your own server, it is your developer or nobody. We explain this split in more detail in &lt;a href="https://security.hackerme.ai/website-security-what-your-host-handles-and-what-you-still-need-to-check/" rel="noopener noreferrer"&gt;what your host handles and what you still need to check&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;A shared server is also a shared risk. When one layer of the hosting stack has a flaw, every site on it can be affected, as we saw in the &lt;a href="https://security.hackerme.ai/shared-hosting-vulnerability-litespeed/" rel="noopener noreferrer"&gt;shared hosting vulnerability in LiteSpeed&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check today
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Ask your host two questions
&lt;/h3&gt;

&lt;p&gt;When will the new PHP release be installed on my server? And is my site on a PHP version that still gets security fixes at all?&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Ask your developer one question
&lt;/h3&gt;

&lt;p&gt;Does our code change session settings at runtime, and is session upload progress turned on? If they are not sure, ask them to apply PHP's interim setting until the update is in place.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Test after the update
&lt;/h3&gt;

&lt;p&gt;PHP updates rarely break sites, but plugins and older themes sometimes misbehave. Follow a short routine like the one in our guide to &lt;a href="https://security.hackerme.ai/safe-wordpress-security-updates-how-to-rehearse-apply-and-harden-without-breaking-the-site/" rel="noopener noreferrer"&gt;safe WordPress security updates&lt;/a&gt;: take a backup, update, then check the home page, login, forms and checkout.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. See what strangers see
&lt;/h3&gt;

&lt;p&gt;An update fixes the server software. It does not tell you what else your site exposes. A &lt;a href="https://brandingme.ai/app/r/hl4DE78bpki378-qllJluw" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what an outsider can see from the internet, without logging in.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Is my WordPress site in danger today?
&lt;/h3&gt;

&lt;p&gt;PHP says the critical issues need non-default settings or uncommon setups, and no attacks on the new flaws had been reported when we wrote this. The safe move is still to get the update installed soon, because details become public with the release.&lt;/p&gt;

&lt;h3&gt;
  
  
  Do I need to do anything if I am on managed hosting?
&lt;/h3&gt;

&lt;p&gt;Ask. Many hosts update PHP quickly, but some only patch the versions they still support, and older sites are often left on old PHP versions. Get a date from your host.&lt;/p&gt;

&lt;h3&gt;
  
  
  What is CVE-2026-115560?
&lt;/h3&gt;

&lt;p&gt;It is a remote code execution flaw in PHP's session upload progress feature that only works with a rare mix of session settings and application code. PHP's interim fix is to turn off session upload progress until the update is installed.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;The original version of this article is on the &lt;a href="https://security.hackerme.ai/php-security-update-critical/" rel="noopener noreferrer"&gt;HackerMe Security blog&lt;/a&gt;.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>security</category>
      <category>webdev</category>
      <category>php</category>
      <category>cybersecurity</category>
    </item>
    <item>
      <title>Docker Hub Leaked Keys: Abused 7.86 Seconds After Upload</title>
      <dc:creator>HackerMe</dc:creator>
      <pubDate>Wed, 18 Nov 2026 06:16:18 +0000</pubDate>
      <link>https://dev.to/hackerme/docker-hub-leaked-keys-abused-786-seconds-after-upload-27id</link>
      <guid>https://dev.to/hackerme/docker-hub-leaked-keys-abused-786-seconds-after-upload-27id</guid>
      <description>&lt;p&gt;If someone builds or hosts your website or web app, ask them one question this week: do we publish any public container images, and are our passwords and keys kept out of them? New research shows that keys left inside public Docker Hub images can be found and used in seconds, not days.&lt;/p&gt;

&lt;h2&gt;
  
  
  What happened
&lt;/h2&gt;

&lt;p&gt;Two researchers at Stony Brook University, Billy Tsouvalas and Nick Nikiforakis, wanted to measure how fast attackers find secrets hidden inside public container images. A container image is a packaged copy of an app, with its code and settings, that developers upload to a "registry" such as Docker Hub so servers can download and run it.&lt;/p&gt;

&lt;p&gt;Over four months, in two rounds, they published 300 test images across eight public registries. Inside, they planted decoy secrets: AWS cloud keys, OpenAI and Anthropic (Claude) API keys, and fake SSH and database logins. They also added small "beacons" that would alert them if anyone built or ran the images. Their paper, "Gone in 8 Seconds: Uncovering Malicious Bots Operating on Public Container Registries," appears in the proceedings of ACM CCS 2026, and BleepingComputer reported on it on 17 November 2026.&lt;/p&gt;

&lt;p&gt;The results:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;On Docker Hub, the fastest use of a planted AWS key came &lt;strong&gt;7.86 seconds&lt;/strong&gt; after the image was published. The median time to first use was 51.19 seconds.&lt;/li&gt;
&lt;li&gt;BleepingComputer reports that all 100 AWS keys were used within 15 minutes.&lt;/li&gt;
&lt;li&gt;Across the study, the AWS keys received more than 140,000 API calls from more than 5,000 unique IP addresses, and the images were pulled more than 75,000 times.&lt;/li&gt;
&lt;li&gt;The OpenAI and Claude keys were drained of their full account balances in short bursts.&lt;/li&gt;
&lt;li&gt;Docker Hub drew almost all of the AWS activity. One other registry, Harbor, saw some.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The most important detail for business owners: the bots almost never ran the images. The researchers say the scanners "overwhelmingly" pulled secrets out through static analysis, by unpacking image layers or querying the registry, "without ever instantiating the container." In plain terms, nobody has to break into your server. If a key is inside a public image file, reading the file is enough.&lt;/p&gt;

&lt;p&gt;The researchers say they reported their findings to Docker in May. Tsouvalas told BleepingComputer they received an acknowledgment but no follow-up, and a small test in August showed no change. We did not find a public statement from Docker about the study.&lt;/p&gt;

&lt;h2&gt;
  
  
  Who is affected
&lt;/h2&gt;

&lt;p&gt;You are at risk if all three are true:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Your site or app is shipped as a container image (common for custom web apps, headless stores, APIs and AI features).&lt;/li&gt;
&lt;li&gt;That image was ever pushed to a public repository, even for a few minutes.&lt;/li&gt;
&lt;li&gt;A secret was inside it: the site's &lt;code&gt;.env&lt;/code&gt; file, a database password, a Stripe or other payment key, a cloud access key, or an AI provider key.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Many small businesses never see this layer. An agency or freelancer builds the app, pushes an image, and the hosting runs it. That is exactly why it is worth asking.&lt;/p&gt;

&lt;h3&gt;
  
  
  What a leaked key can cost you
&lt;/h3&gt;

&lt;p&gt;The study saw AWS keys tested, used to list the resources they could reach, and sometimes used to consume those resources. AI keys were spent down to zero. For a small business, the same pattern can mean a cloud or AI bill you did not run up, customer data read from a database, or an attacker changing your site. We saw the same "find the key, test it, spend it" pattern in our post on &lt;a href="https://security.hackerme.ai/exposed-api-keys-ai-agents/" rel="noopener noreferrer"&gt;how AI agents now hunt exposed API keys&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  The dark-web side
&lt;/h3&gt;

&lt;p&gt;This study did not track whether stolen keys were resold. In general, though, working credentials are a commodity: some attackers use them right away, others pass them on to people who will. That is why "the image was only public for a minute" is not a defense, and why a dark-web check is a lagging signal at best. Our guide on &lt;a href="https://security.hackerme.ai/what-a-dark-web-scan-checks-for-your-website-and-what-it-does-not/" rel="noopener noreferrer"&gt;what a dark web scan checks for your website, and what it does not&lt;/a&gt; explains the limits.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to check
&lt;/h2&gt;

&lt;p&gt;Ask your developer or host to confirm, in writing if possible:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Which images are public.&lt;/strong&gt; List every repository on Docker Hub or any other registry and whether it is public or private.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;No secrets inside images.&lt;/strong&gt; Docker's own documentation says build arguments and environment variables are "inappropriate for passing secrets" to a build "because they persist in the final image." Keys should be supplied at runtime or through a secrets manager.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Deleted is not gone.&lt;/strong&gt; A file removed in a later build step can still sit in an earlier image layer, and deleting an image does not un-leak a key that was already copied.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Rotate anything exposed.&lt;/strong&gt; If a key was ever in a public image, replace it with a new one and turn the old one off. Then check the account's logs and bills for activity you do not recognize. Our &lt;a href="https://security.hackerme.ai/apify-breach-api-token-rotation/" rel="noopener noreferrer"&gt;API token rotation checklist&lt;/a&gt; walks through the order.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Limit damage.&lt;/strong&gt; Use separate keys per project, with the smallest permissions that work and spending limits on AI and cloud accounts.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;A &lt;a href="https://brandingme.ai/app/r/aqD4FmX11egxawaG__yUzg" rel="noopener noreferrer"&gt;free website security scanner&lt;/a&gt; can show what your public website exposes from the outside, such as leftover public files or missing security headers. It cannot see inside your container images or your registry account. Those checks need your developer or host.&lt;/p&gt;

&lt;h2&gt;
  
  
  FAQ
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Does this mean Docker Hub was hacked?
&lt;/h3&gt;

&lt;p&gt;No. Nobody broke into Docker Hub. The researchers measured what happens when secrets are placed inside images that anyone can download. The bots simply read what was published.&lt;/p&gt;

&lt;h3&gt;
  
  
  My site is on WordPress with ordinary hosting. Does this apply to me?
&lt;/h3&gt;

&lt;p&gt;Probably less, but ask. If your site uses a custom app, an API, an AI feature or a headless store built by an agency, it may be shipped as a container. The question to ask is the same either way: where are our keys stored, and could any of them be public?&lt;/p&gt;

&lt;h3&gt;
  
  
  If we deleted the public image, are we safe?
&lt;/h3&gt;

&lt;p&gt;Not on its own. In this study, the first key was used within seconds, so assume a key that was public has already been copied. Replace it, then review the account's activity.&lt;/p&gt;

&lt;p&gt;The simple action for today: send your developer or host one message asking whether any of your images are public and whether secrets are kept out of them, and replace any key that was ever exposed.&lt;/p&gt;




&lt;p&gt;Sources: Tsouvalas and Nikiforakis, "Gone in 8 Seconds: Uncovering Malicious Bots Operating on Public Container Registries," ACM CCS 2026, &lt;a href="https://doi.org/10.1145/3830454.3846791" rel="noopener noreferrer"&gt;https://doi.org/10.1145/3830454.3846791&lt;/a&gt; (abstract via &lt;a href="https://containerbots.github.io/" rel="noopener noreferrer"&gt;https://containerbots.github.io/&lt;/a&gt;); BleepingComputer, 17 November 2026, &lt;a href="https://www.bleepingcomputer.com/news/security/cloud-keys-leaked-in-docker-hub-images-are-stolen-within-seconds/" rel="noopener noreferrer"&gt;https://www.bleepingcomputer.com/news/security/cloud-keys-leaked-in-docker-hub-images-are-stolen-within-seconds/&lt;/a&gt;; Docker documentation, Build secrets, &lt;a href="https://docs.docker.com/build/building/secrets/" rel="noopener noreferrer"&gt;https://docs.docker.com/build/building/secrets/&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>security</category>
      <category>docker</category>
      <category>devops</category>
      <category>webdev</category>
    </item>
  </channel>
</rss>
