OpenAI on Tuesday revealed the rogue artificial intelligence (AI) agent that escaped its sealed evaluation environment and broke into Hugging Face's production environment also hacked multiple third-party accounts and services as part of the attack. The latest disclosure shows that the security incident, which stemmed from an internal security test, was more extensive in scope than previously
Cybersecurity News and Vulnerability Aggregator
Cybersecurity news aggregator
Updated: 8:20 pm
J/K Navigate / Search S Save V Open Link
— or
treemd <(curl -sL https://allsec.sh/md) (as Markdown) Top Cybersecurity Stories Today
Ruby on Rails has released fixes for a critical Active Storage vulnerability that could let unauthenticated attackers read arbitrary files from application servers through crafted image uploads. Tracked as CVE-2026-66066 (CVSS score: 9.5), the flaw can expose the Rails process environment and secrets such as secret_key_base, the Rails master key, database passwords, cloud storage credentials,
Cybersecurity researchers have shared additional technical details about a recently patched critical security flaw impacting Check Point Security Management Server and Multi-Domain Security Management Server (MDS) that has come under active exploitation in the wild. The vulnerability, tracked as CVE-2026-16232 (CVSS score: 9.3), is an authentication bypass in the SmartConsole login process that
Latest
Wednesday, July 29
Details of parents and staff, including email addresses and phone numbers, are among data taken by cybercriminals The Department for Education and a police database have been targeted by a cyber-attack, exposing more than 740,000 pieces of data. Details of government officials, senior school leaders, university staff, police officers and members of the public have been taken by hackers. Continue reading...
Europe regulates AI through risk tiers and conformity checks. America accelerates AI through deregulation and infrastructure speed. Both models ask different questions but land on the same gap: neither tells you whether a live AI system actually holds up against real attacks. Enterprise AI security depends on evidence, not paperwork or policy. Continuous, human-validated offensive testing is the layer that both regulatory philosophies are missing and that every organization operating across borders actually needs. The post Europe Is Regulating AI. America Is Accelerating AI. Both Need Offensive Security Validation. appeared first on Synack .
Ruby on Rails has released fixes for a critical Active Storage vulnerability that could let unauthenticated attackers read arbitrary files from application servers through crafted image uploads. Tracked as CVE-2026-66066 (CVSS score: 9.5), the flaw can expose the Rails process environment and secrets such as secret_key_base, the Rails master key, database passwords, cloud storage credentials,
VU#293714: Arbitrary File Overwrite in Develar app-builder (zipx.Unzip) via Symlink Following on macOS (APFS)
Overview A vulnerability in the zipx.Unzip extraction routine of Develar’s app-builder allows an attacker to overwrite arbitrary files on macOS using Apple File System (APFS). The issue arises from a combination of Unicode normalization collisions and unsafe symlink-following behavior. APFS treats certain Unicode equivalent filenames as identical (e.g., ß ↔ ss), while app builder performs no canonical normalization before validating or writing paths. Description Develar’s app-builder is a command‑line build tool used heavily in the Electron ecosystem to package, sign, notarize, and produce distributable application bundles for macOS, Windows, and Linux. It is popular because it is a transitive dependency of electron-builder, one of the most widely used packaging tools for Electron apps. The vulnerability arises from how the zipx.Unzip routine handles Unicode‑equivalent filenames and symbolic links during ZIP extraction. APFS treats certain Unicode representations as identical (e.g., ß and ss), but app-builder does not perform canonical normalization before validating output paths. As a result, an attacker can craft ZIP archives that combine Unicode normalization collisions with malicious symlinks to redirect writes outside the intended extraction directory. The GitHub Pull Request addresses the symlink-following variant of this vulnerability (CWE‑22 and CWE‑59) by rejecting any symlink whose resolved target escapes the intended output directory and by adding O_NOFOLLOW to regular file writes. Without these protections, an attacker can craft ZIP archives containing malicious symlinks and Unicode-colliding filenames to overwrite arbitrary files on APFS-backed systems. Impact Exploitation of this vulnerabi
CISA's Binding Operational Directive 26-04, issued June 10, 2026, replaces the KEV directive with a four-variable risk model: asset exposure, KEV status, exploit automation, and technical impact. CISA publishes three of the four answers generically per CVE. The two that decide whether prioritization is defensible (real exposure and real impact in your environment) can only be proven by testing the asset the way an adversary would. Continuous, human-validated pentesting is built to produce that evidence. The post BOD 26-04 Makes Exploitability the Priority Signal: Scanners Can’t Measure It appeared first on Synack .
Broadcom has released security updates to address multiple security flaws impacting VMware ESX, vCenter, Workstation, and Fusion, three of which have been designated as critical in severity. The first of the three critical-rated flaws is CVE-2026-59309 (CVSS score: 9.8), which has been described as an authentication bypass in VMware vCenter. "A malicious actor with network access to vCenter
Overview A vulnerability has been discovered in the OPeNDAP Hyrax software solution. A remote attacker with the ability to submit crafted requests to an affected Hyrax instance could cause the application to communicate with unauthorized remote systems. Under certain conditions, the vulnerability may also result in the unintended disclosure of user authentication tokens to unauthorized destinations. Description CVE-2026-16637 OPeNDAP Hyrax is vulnerable to Server Side Request Forgery (SSRF) and credential disclosure via unvalidated HTTP redirects that bypass the AllowedHosts allowlist and leak Earthdata headers (User-Id, Echo-Token) to attacker-controlled endpoints. OPeNDAP Hyrax is an open-source data server software that enables remote access to scientific datasets over the internet using the OPeNDAP protocol. It allows users to query, subset, and retrieve data in various formats (such as NetCDF, HDF, or GrADS) without downloading entire files. OPeNDAP Hyrax uses a list of allowed hosts, specified through a regular expression, to limit where it can fetch data. When a requested server responds with a redirect, HTTP 3xx, the software follows the redirect without checking whether the new destination is still on the allowed list. This behavior enables an attacker to redirect the system to an untrusted or internal target that would otherwise be blocked. Additionally, when the system follows such a redirect, it may carry certain user identification headers, including a legacy credential called Echo-Token, to the new destination, even though the main authorization token is correctly stripped by the underlying library. Impact Successful exploitation could allow an unauthenticated remote attacker to access internal services that are not intended to be reachable fr
This is Hugging Face’s write up from the supposedly rouge OpenAI agent. 17k actions from four agents across 4 days. How did it make that much noise and not get sniffed out sooner… Also this just smells like the agents were trying every combination they knew of vs trying to creare anything new (like finding a new vuln). Visual is pretty cool at least.
A coordinated cyberattack targeted operational technology at more than 30 Minnesota community water systems on July 26 and 27, triggering a statewide cybersecurity response. Braham, Plymouth, South St. Paul and Maple Plain have publicly described a plant outage, communications failures or affected automated controls. Braham's water plant went offline, and the city asked residents to minimize
Cybersecurity researchers have disclosed details of a large-scale fraud campaign that involves creating lookalike websites of major Russian companies with an aim to siphon funds from international firms for more than nine years. According to Russian cybersecurity vendor F6, the threat actors have set up clone websites of Russian companies across fertilizer manufacturers, petrochemical companies
Cloudflare's Authenticated Origin Pulls and Custom Origin Trust Store now support post-quantum authentication. Here we’ll explain how you can configure fully post-quantum secure mutually authenticated TLS connections to your origin server, dive into the engineering details of how we built it, make a shameful confession, and finally explain how this work fits into our overall post-quantum migration roadmap. Reaching a major milestone Our focus for the past several years has been in deploying post-quantum encryption to protect against harvest-now/decrypt-later attacks, where an attacker quietly stockpiles your encrypted data with the hope of decrypting it in the future with a quantum computer. However, recent breakthroughs in quantum computing and cryptanalysis pulled the timelines for upgrading to post-quantum cryptography forward across industry and government and have caused us to shift our attention to deploying post-quantum authentication , to protect against attackers who will soon be able to use quantum computers to break classical credentials and carry out impersonation attacks. In a previous post, we announced that Cloudflare is
AI is compressing exploit timelines. The real question isn't whether your vulnerability management playbook needs to change, it's which part of it you've been getting wrong all along. The conversation happening in security circles right now goes something like this: Mythos is here. Exploit timelines are collapsing. Does the vulnerability management playbook need to change? The honest answer is
Nebula Security says a patched Firefox JIT flaw could be triggered by simply visiting a malicious webpage and was also used to compromise Tor Browser. Tracked as CVE-2026-10702, the bug provides arbitrary code execution inside the browser's renderer process. Mozilla rated it High and fixed it in the Firefox 151.0.3 update. "No settings or additional user interaction are required," Eten Zou,
Hi everyone, I've released APKX-Hunter v2.5.0, an open-source Android Static Analysis Framework written in C.
Only 1% of AI-discovered vulnerabilities have actually been exploited in the wild - a rate that matches standard, human-found bugs.
Is the 'Cyber-AI Industrial Complex' creating a dangerous distraction for CISOs? Out of thousands of AI-discovered vulnerabilities, only 1% end up being exploited in the wild. the same exploitation rate as human-found bugs. This raises a huge question about priority: If AI isn't unleashing a wave of hyper-dangerous, novel zero-days, why are security budgets being steered away from core hygiene and toward edge-case AI threat vectors, all while ransomware & data breaches hit new records? Curious to hear from defenders and CISOs: 1. Are AI security tools actually giving your team ROI, or just adding noise to your backlog? 2. How are you balancing the pressure to fund "next-gen AI defense" with patching fundamental exposure?
Most organizations have incident response plans, security tools, and technical teams in place. Yet new research suggests that many still lack the coordination, visibility, and executive alignment needed to withstand a serious cyberattack. According to The State of Incident Response Readiness 2026, based on a survey of 600 senior IT security decision makers conducted by Vanson Bourne in January
The Federal Security Service of the Russian Federation (FSB) on Wednesday said it charged Telegram founder Pavel Durov for allegedly facilitating terrorist activities and for failing to remove prohibited information in violation of Russian law. The principal security agency said the instant messaging platform "failed to remove numerous channels, chats, and bots on the platform that are
One day after a federal judge ordered an ICE detention center opened to state health inspectors, the agency posted new contract terms that would void state oversight at four facilities.
Cybersecurity researchers have shared additional technical details about a recently patched critical security flaw impacting Check Point Security Management Server and Multi-Domain Security Management Server (MDS) that has come under active exploitation in the wild. The vulnerability, tracked as CVE-2026-16232 (CVSS score: 9.3), is an authentication bypass in the SmartConsole login process that
OpenAI on Tuesday revealed the rogue artificial intelligence (AI) agent that escaped its sealed evaluation environment and broke into Hugging Face's production environment also hacked multiple third-party accounts and services as part of the attack. The latest disclosure shows that the security incident, which stemmed from an internal security test, was more extensive in scope than previously
Gitea, the self-hosted Git platform, has patched a critical remote code execution vulnerability. A user with ordinary repository write access can turn attacker-controlled patch content into a live Git hook and run shell commands as the Gitea service account. Tracked as CVE-2026-60004 (CVSS score: 9.8), the flaw affects Gitea versions 1.17 and later before 1.27.1 and is fixed in 1.27.1. The
Source code for the Flying Eagle Android remote access trojan (RAT) framework is circulating through criminal Telegram channels. Hunt.io and independent researcher NetAskari traced matching control panels and certificates to 170 internet servers. They linked the framework to a fake "公安一网通办" Public Security service application targeting Android users in China. The kit supports payment-password
Beta release versions of two npm packages in the @joyfill namespace have been compromised to deliver a remote access trojan (RAT) associated with the DEV#POPPER malware family. The list of affected packages is as follows - @joyfill/layouts@0.1.2-2773.beta.0 @joyfill/components@4.0.0-rc24-2773-beta.4 The two packages "contain an import-time JavaScript implant that resolves encrypted code
Tuesday, July 28
How much could a single underscore in a username really matter? Just ask Brandon Klayme, who served 18 months in prison before realizing how authorities arrested, charged, and convicted the wrong man.
Anthropic says Claude Mythos Preview helped derive an end-to-end key-recovery attack against HAWK-256 and a 200- to 800-fold speedup for an attack on seven-round AES-128. The HAWK attack exploits a previously unused symmetry in the lattice behind the signature scheme. Anthropic's released implementation gives an expected end-to-end runtime of about three hours and 42 minutes on a 96-core server
A new Mirai-derived botnet called Tengu can use a compromised Linux device's hardware watchdog to trigger a reboot when defenders kill its main process. If that happens, Tengu's other persistence mechanisms get another chance to relaunch it. Nozomi Networks Labs observed the dropper reaching its honeypots through Telnet credential brute force. Tengu supports 25 distributed denial-of-service (
“Make everyone one percent better and it compounds across the team.” That’s the line Joe Head wrote about his own job and when I read it back to him, he didn’t hesitate. “That is 100% the objective,” he said. Joe runs AI adoption for an 80-person team as the only AI automation specialist in the […] The post Tools Change. Teach People How to Keep Up appeared first on Heimdal Security Blog .
Cybersecurity researchers have sounded an alert after finding more than 36,000 Baseboard Management Controller (BMC) management interfaces exposing Intelligent Platform Management Interface (IPMI) protocol to the public internet. Of the 36,872 internet-exposed server-management interfaces running IPMI, 24,650 have been found to disclose password-derived authentication hashes before login due to
TL;DR: We identified 36,872 internet-exposed BMCs, and 24,650 of them disclosed password-derived authentication hashes before login because of CVE-2013-4786. More than 30% of the returned hashes were linked to passwords that could be recovered using common wordlists or predictable factory password formats. The exposure affected modern Supermicro and HPE servers, including systems operated by GPU providers. The bigger risk is that a compromised BMC gives an attacker highly privileged access below the operating system. Because BMC management networks are often poorly segmented and lightly monitored, one exposed interface can become a foothold into broader data center infrastructure. We also created an interactive map where you can explore the exposed systems: https://lavahq.io/bmcradar
Like most infrastructure, the Internet's fragility is easy to overlook — as long as it's working. When it fails, its complexity comes into full view. Cloudflare is in a unique position to detect and document the moments when one of the interrelated systems the Internet depends on breaks down and connectivity suffers as a result. Each quarter, we summarize the disruptions we detect and annotate on Cloudflare Radar . In Q2 2026, Super Typhoon Sinlaku just north of Guam caused the longest outage, while government-mandated shutdowns during exam periods in Sudan were the most frequent. Iran restored national Internet access, reconnecting its citizens to the global network after an 88-day blackout, even as damage from drone strikes continued to disrupt AWS infrastructure elsewhere in the region. Finally, a cable cut in Saint Lucia and the distribution of faulty DNSSEC signatures in Germany underscored the fragility of Internet infrastructure, but also the remarkable stability these regional and global systems maintain when operating normally. Here we will walk through the most significant Internet disruptions we observed in the second quarter of 2026, drawing on traffic data from Cloudflare Radar to show how each unfolded and what it meant for users on the ground. As always, this is a summary of notable, confirmed disruptions rather than an exhaustive list; a fuller view of detected traffic anomalies is available in the Cloudflare Radar Outage Center . Natural disasters and electricity cause disruptions in Guam, Venezuela, and Tanzania Super Typhoon Sinlaku, the strongest storm of the 2026 Pacific typhoon season so far, tracked through the Mariana Islands in mid-April, passing just north of Guam. Tho
Great note-taking and reporting start with clear goals and planning. Organization of your findings matters just as much as what you include. If documenting your #DFIR investigations feels like a chore, this post might spark some ideas and give you a few quick tips.
The Iranian state-backed hacking group tracked as Nimbus Manticore (aka GalaxyGato, Mirage Kitten, Smoke Sandstorm, Subtle Snail, and UNC1549) has been attributed to a fresh set of attacks targeting entities across the Middle East, Africa, and South Asia. The intrusions involve the use of a previously undocumented Windows backdoor called NightLedger and two custom WebSocket tunnelers,
Reverse-engineered the BLE protocol of a discontinued Fisher-Price toy Lumalou after its app was discontinued
Codex’s /goal feature amplifies bug hunting, but getting good results requires the right prompt, the right scope, and the right number of outcomes per run. For Patch the Planet , our joint initiative with OpenAI to find and fix bugs in open-source software, we pointed Codex at some of the most widely used, heavily audited codebases in the world, like Rust, curl, and zlib. One tool came up again and again in our internal bug-report channels: /goal , which hands Codex an open-ended objective and lets it work independently toward a success condition. Here are a few highlights: /goal found every Rust bug we submitted, including a soundness hole and a miscompilation now patched in Rust 1.98, from a single variant-analysis pipeline. It turned every project’s past CVEs into Semgrep rules that had to fire on the vulnerable version and stay silent on the patched one, then flagged 11 variant hits across multiple projects. It uncovered two potential high-severity privilege-escalation bugs in Keycloak’s SAML component during a discovery run. Over the first few weeks of Patch the Planet, our engineers independently converged on three techniques for using /goal . We found that getting the most out of /goal means treating the prompt as a set of specific success criteria, not a set of instructions. (Note that this blog post uses /goal to refer to goal-based prompting in general. Codex can also set goals for itself through a tool call, and that’s how we recommend everyone use it; we rarely type the slash command ourselves.) 1. Let Codex write the goal The art o
BrainDrain: A Chrome extension that collects your AI prompts without you ever opening it and has 100k users, 9 AI platforms
"Prompt Optimizer - SecondBrain" https://chromewebstore.google.com/detail/prompt-optimizer-secondbr/aajjgdpofhhcjmjoombjdfepplndhgcp, v2.3.1. The prompt rewriting works fine. Alongside it a capture engine runs at `document_start` on 9 AI sites and POSTs prompts and replies to the vendor's ingest endpoint. No interaction with the extension required. Reproduced on a clean profile, with the service worker devtools open: 1. Installed the extension. Never opened it. 2. Browsed to an unrelated site. The extension pulled its configuration from the server and wrote a userId and credentials into extension storage. 3. Opened ChatGPT and asked a question. Once the reply finished, a POST to `/context` went out carrying both the prompt and the response, encrypted with the credentials issued in step 2. **At no point was the extension opened or clicked.** Store privacy declaration: "The developer has disclosed that it will not collect or use your data." Write-up, IOCs : https://malext.io/reports/BrainDrain/ Happy to provide the decryption for anyone wanting to test the extension in a sandbox
GitHub - memues/mstar-monitor-firmware-dumper: Dump SPI flash firmware from MStar scaler-based monitors over I2C/DDC
Monday, July 27
Microsoft is introducing new AI tools designed to help customers continuously streamline and automate the process of identifying and reducing their exposure to security risks. The new tools come less than a week after OpenAI lost control of two of its security models when they infiltrated the servers of startup Hugging Face. The hack, Hugging Face added, involved “a swarm of tens of thousands of automated actions” that stole internal Hugging Face credentials. The OpenAI models achieved this feat by exploiting a zero-day flaw in Hugging Face’s data-processing pipeline to run malicious code that escalated the models’ access to the company’s high-value cloud and server clusters. Microsoft’s announcements on Monday made no reference to the event, which OpenAI said was “unprecedented.” The company also didn’t say what would prevent the new tools from similarly going rogue. Read full article Comments ]]>
The outgoing executive director of the Office of Homeland Security Statistics is one of very few federal officials to speak out against the Trump administration’s immigration crackdown.
The screwup shows how tricky it can be to stop web crawlers from making ostensibly private conversations with AI chatbots entirely too public.
CVE-2026-61511 - a critical vulnerability in vBulletin that allows an unauthenticated attacker to execute arbitrary code on a remote server.
Debugging privacy-preserving protocols is hard. Oblivious HTTP has several different steps across four different parties, not to mention binary HTTP encoding and details spread across many draft RFCs. We've taken what we've learned operating protocols like Oblivious HTTP at the scale of millions of requests per second, and wrapped it up in a nice, clean CLI tool — that we are open-sourcing today. We call it our privacy-client , or pvcli . We’re releasing it under the Apache-2.0 License, and it is open for contributions . Here’s a single line of code that executes a full Oblivious HTTP request with a relay, gateway and origin. Don't worry if you don't know what that means, we'll cover it below. We’ll explain why we built this tool, and show just how handy it can be. Why privacy protocols can be hard to debug Let’s take a closer look at our motivation for creating pvcli. Over time, the Privacy team’s product suite and customer base grew. We added products like Privacy Proxy and Privacy Gateway, which power Apple’s Private Relay , Microsoft’s Edge Secure Network VPN , Flo Health’s Anonymous Mode , and more. With it came an increasing amount of special customer requirements, domain knowledge, and complexity. As a result, we saw increased friction in development and incident response. To see this in action, let’s look at how one of our products implements
To reduce the amount of noise from questions, we have disabled self-posts in favor of a unified questions thread every week. Feel free to ask any question about reverse engineering here. If your question is about how to use a specific tool, or is specific to some particular target, you will have better luck on the [Reverse Engineering StackExchange](http://reverseengineering.stackexchange.com/). See also /r/AskReverseEngineering.
We analyze how VoidLink, LinkPro, and Atomic Arch abuse eBPF helpers to hide from defenders, and show how to detect them at load time, before they can act.
Sunday, July 26
Presently sponsored by: Report URI: Guarding you from rogue JavaScript! Don’t get pwned; get real-time alerts & prevent breaches #SecureYourSite The Origin Energy breach down here in Aus is all over the news this week, and as with many breaches, it's multi-faceted. You've got them leading with "don't worry, your credit card is fine", the hacker leading with "they didn't respond when I tried to report it", and now news that the two parties have "come to an agreement". Maybe money was paid, or maybe Origin made some promises to restrain the hounds if commitments about data deletion were made. But both outcomes, of course, provide no guarantee that data has been nuked, so now they get to spend perpetuity waiting for the data that maybe - just maybe - it leaks. And we all should be working on precisely that assumption, just like we did with Optus and Medibank and Latitude and Ticketek and Qantas...
Saturday, July 25
Friday, July 24
Border Gateway Protocol ( BGP ) is the de facto routing protocol of the Internet. It offers built-in mechanisms to allow entities, represented by Autonomous Systems (ASes), to express how they want to send and receive traffic on the Internet. One such mechanism is path attributes , which carry essential routing information and metadata for their associated route. The path selection algorithm processes some of these path attributes in a deterministic sequence to calculate the best path for this specific prefix. Using our unique position on the Internet, we took an investigative look at one of the well-known mandatory attributes in BGP, the ORIGIN attribute. ORIGIN must be present in every BGP prefix announcement and should not be modified by any router after being set by the originating one. What we found through our own experiments was a dramatic ~70% of observed paths in numerous vantage points have a different ORIGIN value compared to what was set by the originating Autonomous System. This ORIGIN attribute manipulation has a significant impact on the way traffic is forwarded on the Internet, as we’ll explore in this post. BGP ORIGIN and its operational history The ORIGIN attribute indicates how a route was injected into BGP — not to be confused with the origin AS that ind
Iberia Cards CISO José Manuel Rivera García explains why he's stuck with Synack's PTaaS model across multiple organizations, and how running Sara AI Pentest alongside human researchers helps him balance regulatory compliance with real risk reduction. He also shares candid advice for other CISOs on avoiding the false sense of security that comes from infrequent testing and over-reliance on perimeter controls. The post How Iberia Cards Uses Sara AI Pentesting to Stay Ahead of Modern Threats appeared first on Synack .
Analysis of satellite images of Myanmar shows dozens of alleged scam compounds have appeared in recent months, despite a purported crackdown on the criminal organizations.
Thursday, July 23
During an internal evaluation with its safety guardrails switched off, an OpenAI model escaped its test environment and breached Hugging Face's production systems, again, this time to steal answers to its own benchmark. No one told it to. It decided that on its own. The post How an OpenAI Model Escaped its Guardrails appeared first on Synack .
Getting locked out of an account is no fun. Google has a few ways to help you regain access if you happen to forget your password or lose an authenticator, including recovery contacts and backup codes. Now, Google has a completely new option: your face . You can now give Google a video record of your face and sign into your account with a selfie, which sounds like something people are going to just love. You will have to set this feature up ahead of time if you want the option of regaining account access with a selfie later on. To get started, verify your account type is supported. You won't be able to configure selfie sign-ins for Workspace accounts, child accounts, or any account enrolled in Google's Advanced Protection Program. Configuring selfie sign-in requires you to record a video, which Google will store on its servers. Google's selfie sign-in landing page includes the typical disclaimers about privacy and data access, promising that the company will keep the video encrypted and won't use it for any other purposes unless you opt in. Read full article Comments ]]>
Today we’re excited to announce Cache Response Rules. These are a new rule type that runs after an origin server replies but before Cloudflare caches the content. If you've ever been irked watching something that should easily sail out of cache get dragged back to the origin by a stray Set-Cookie or wrong Cache-Control , headers that are sometimes hard or impossible to strip or change on the origin itself, then Cache Response Rules is that fix, applied at exactly the right moment. When and how caching decisions are made A CDN cache and an origin server work as a pair. Their goal is to answer from the cache whenever possible and only go back to the origin when the edge can’t respond. Every point of cache hit ratio comes from getting that division of labor right. Check the cache when we shouldn't, and we waste a lookup that was always going to miss. Check it too rarely, the origin serves traffic the edge should have absorbed, and the performance win evaporates. Importantly, the origin guides the cache. When it returns a cacheable asset, its response headers tell Cloudflare how long it’s OK to serve it, when and how to revalidate, and even whether to cache it at all. The cache is only ever as efficient as the origin allows. If the origin gets it wrong, cache becomes decoration while the origin infrastructure costs skyrocket. Most cache eligibility problems are not decided at request time. They manifest after the origin replies. A visitor asks for /static/app.js . Cloudflare checks cache, it misses, and forwards the request to the origin. The origin returns the file. Somewhere in those response headers, quietly, is a Set-Cookie header. The asset that should have been cached at every Cloudflare data center is now uncacheable. Multiply that by every visitor, on every site, with the same accidental header, and you have a
Caught this in three open directories on a Hong Kong server, exposed July 9 to 13. The agent is Hermes, open source, and the recovered logs show it running LinPEAS and walking a ministry web root without a human in the loop. Target was Thailand's Ministry of Finance.
MSG’s sprawling surveillance system can monitor guests down to the second. Its owners made an exception for the pop star’s rehearsal dinner.
VU#492466: Logto Identity Platform has authentication and authorization failures in core protocol handling
Overview The Logto platform contains multiple vulnerabilities affecting the identity‑processing pipeline. These flaws reduce the reliability of authentication and authorization decisions and may allow attackers to bypass account‑ownership checks, skip MFA, replay externally issued SSO responses, or submit identity assertions without proper cryptographic or validity checks. Collectively, the issues create several paths for unauthorized access across both local and federated sign‑in flows. Description Developed by Silverhand Inc., Logto is an identity and access management system for software as a service (SaaS) and AI applications. It provides multi‑tenant authentication, single sign-on (SSO), role-based access control (RBAC), support for openId connect (OIDC), open authorization (OAuth) 2.1, and Security Assertion Markup Language (SAML) authentication. Several vulnerabilities have been identified in Logto's authentication and SSO components that weaken core identity‑security protections across these protocols. CVE-2026-15611 Logto allows unverified email-based SSO account linking. If a new SSO login arrives and no existing (issuer, identityId) account is found, Logto searches for a local user by the email address supplied by the identity provider (IdP) and links the SSO identity to that account. Logto does not require the IdP to confirm email_verified before linking, so an attacker who registers the victim's email at a permissive upstream IdP can take over the victim's existing account on first login. CVE-2026-15612 Logto stores a nonce in the connector session at authorization request time and expects the returned id_token to echo it back. The valida
Full report is available at [https://any.run/cybersecurity-blog/banana-rat-evolution-analysis/](https://any.run/cybersecurity-blog/banana-rat-evolution-analysis/) The exposed server at 198\[.\]245\[.\]53\[.\]26 gave a rare opportunity to compare two related Banana RAT branches through live infrastructure, sandbox telemetry, and recovered payloads. The older branch used ETW-themed paths, static Microsoft-looking names, and a typo-based pseudo-Microsoft C2 identity. The newer branch kept the same staging concept but moved to randomized install identifiers, better-structured SYSTEM persistence, and a WebSocket channel built around a hashed `testewin.com` subdomain. IoC: * 198\[.\]245\[.\]53\[.\]26 * [https://app.any.run/tasks/96796146-688f-4b12-894c-236dadab8413](https://app.any.run/tasks/96796146-688f-4b12-894c-236dadab8413)
There are several cybersecurity companies that offer managed EDR services in 2026. Here’s what actually separates them, and who each one suits. Most successful cyber attacks begin with a breached laptop, a smartphone or a server. Over the past 15 years, endpoint detection and response (EDR) has gone from niche to a mainstream security solution. […] The post The 4 best managed EDR service suppliers (and how to choose) appeared first on Heimdal Security Blog .
Hackers access Australian customers’ names, addresses, dates of birth, phone numbers and some bank account details, company says Follow our Australia news live blog for latest updates Get our breaking news email , free app or daily news podcast Origin Energy customers’ addresses, phone numbers and partial bank account data have been accessed in a hack, the company has confirmed. The firm has 4.8m customer accounts in Australia, providing electricity, fossil gas, LPG and internet services to homes and businesses. Continue reading...
Wednesday, July 22