Blahalla

Why i think Mythos is gonna be game changing after using Opus for a CTF

www.techupkeep.dev · 6 months ago

false

New Scam Alert: QR Codes Replace Links in Traffic Ticket Phishing

www.techrepublic.com · 6 months ago

Scammers are now using QR codes in fake traffic violation texts to steal personal and financial data, mimicking state courts and government agencies. These messages pressure recipients to scan the code to resolve alleged violations, leading them to phishing websites that request personal and financial information. Security experts advise caution with unsolicited QR codes, urging users to verify any violations directly through official government websites and to be wary of urgent payment demands. Legitimate agencies do not request payments or sensitive information via text message.

‘BlueHammer’ Exploit Targets Windows, Potentially Impacting 1 Billion+ Devices

www.techrepublic.com · 6 months ago

A security researcher has publicly released exploit code for a Windows vulnerability called BlueHammer. This flaw allows attackers with limited access to gain SYSTEM-level control of a computer, potentially impacting over one billion devices. The vulnerability combines a time-of-check to time-of-use bug and a path confusion problem. The researcher’s release appears to stem from frustration with Microsoft’s handling of the vulnerability report. Microsoft has acknowledged the situation and stated they are working to address the issue.

Anthropic Launches Claude Mythos Preview Focused on Zero-Day Vulnerability Discovery

gbhackers.com · 6 months ago

Anthropic has launched Claude Mythos Preview, a new language model capable of autonomously discovering and exploiting zero-day vulnerabilities. The model has identified previously unknown flaws in OpenBSD, Firefox, and FFmpeg, even creating functional exploits without human intervention. Researchers uncovered a 27-year-old denial-of-service vulnerability in OpenBSD, a 16-year-old vulnerability in FFmpeg, and exploited a 17-year-old vulnerability tracked as CVE-2026-4747 to achieve remote code execution. To mitigate risks, Anthropic has initiated Project Glasswing to share these capabilities with developers and industry partners.

Massachusetts Hospital Diverts Ambulances as Cyberattack Causes Disruption

www.securityweek.com · 6 months ago

A cyberattack has disrupted operations at Signature Healthcare in Brockton, Massachusetts, forcing a diversion of ambulances. The organization implemented downtime procedures to maintain patient care and safety following the detection of suspicious network activity. Chemotherapy infusion services have been canceled, pharmacies have experienced closures, and patients may face delays at medical group locations. The motive for the attack remains unknown, and no ransomware group has claimed responsibility.

Hackers exploit a critical Flowise flaw affecting thousands of AI workflows

www.csoonline.com · 6 months ago

Hackers are exploiting a critical vulnerability in the Flowise AI development platform, allowing them to inject arbitrary JavaScript code. The flaw, tracked as CVE-2025-59528 and rated 10.0 CVSS, arises from improper validation of configurations within the Custom MCP node. Roughly 12000 to 15000 Flowise instances were exposed online, and exploitation began April 6th, originating from a Starlink IP address. Researchers also identified two additional critical vulnerabilities: CVE-2025-8943, a missing authentication issue, and CVE-2025-26319, an arbitrary file upload vulnerability, which are also being actively exploited.

Hackers Target Adobe Reader Users With Sophisticated Zero-Day Exploit

gbhackers.com · 6 months ago

Security researchers have discovered a new zero-day vulnerability actively exploiting Adobe Reader users. The exploit, detected late last month, allows attackers to steal local files and gather system information with no user interaction. It uses the util.readFileIntoStream() and RSS.addFeed() APIs to exfiltrate data to a remote server with the IP address 169.40.2.68:45191. The vulnerability functions as a fingerprinting tool to identify high-value targets, and can potentially deliver secondary payloads for remote code execution.

Iran‑linked PLC attacks cause real‑world disruption at critical US infra sites

www.csoonline.com · 6 months ago

US federal agencies are warning of an ongoing cyber campaign targeting programmable logic controllers at critical infrastructure facilities, including water, wastewater, energy, and government sectors. Iranian-affiliated threat actors have been actively compromising these PLCs since at least March 2026, using internet-exposed devices and legitimate configuration software to disrupt operations and cause financial losses. The attacks leverage existing vulnerabilities in operational technology environments and reflect a recurring pattern of Iranian state-sponsored targeting of US industrial control systems. Organizations are urged to review logs, restrict access, and secure devices to mitigate further risk.

EvilTokens Uses Stolen Microsoft 365 Tokens, AI to Supercharge BEC

gbhackers.com · 6 months ago

A new Phishing-as-a-Service platform called EvilTokens has emerged, combining stolen Microsoft 365 tokens and artificial intelligence to automate Business Email Compromise (BEC) attacks. The platform, sold on Telegram, offers tools like a custom browser, “Portal Browser” or “ET Browser,” that uses stolen OAuth tokens to access Microsoft 365 accounts, and an AI-driven analysis pipeline to generate tailored BEC emails. EvilTokens leverages large language models like llama-3.1-8b-instant, llama-3.3-70b-versatile, and OpenAI gpt-4o-mini to analyze compromised accounts and automate attack planning. This new service significantly lowers the technical skill required to conduct sophisticated BEC campaigns.

Evasive Masjesu DDoS Botnet Targets IoT Devices

www.securityweek.com · 6 months ago

Security researchers have identified Masjesu, a distributed denial-of-service botnet actively operating since 2023. The botnet primarily targets a wide range of IoT devices, including those running D-Link routers and Huawei home gateways, and has infected devices in Vietnam, Brazil, India, and Ukraine. Masjesu utilizes multiple command and control domains, decrypts data client-side, and employs sophisticated persistence techniques by mimicking legitimate system components. The botnet is advertised on Telegram and capable of launching diverse DDoS attacks, including UDP, TCP, and HTTP floods.

Shrinking the IAM Attack Surface through Identity Visibility and Intelligence Platforms (IVIP)

thehackernews.com · 6 months ago

false

Hackers Targeting Ninja Forms Vulnerability That Exposes WordPress Sites to Takeover

www.securityweek.com · 6 months ago

A critical vulnerability affecting approximately 50,000 websites using the Ninja Forms WordPress plugin has been identified. The File Uploads addon contains an unauthenticated arbitrary file upload issue, allowing attackers to upload malicious PHP code and potentially gain remote code execution. This flaw, tracked as CVE-2026-0740 with a severity score of 9.8, arises from insufficient file type validation and a lack of filename sanitization. Users are urged to update to version 3.3.27 to mitigate the risk.

Reading /etc/passwd via translation file upload in Tolgee's cloud platform (CVE-2026-32251, CVSS 9.3)

simonkoeck.com · 6 months ago

A vulnerability was discovered in Tolgee, an open-source localization platform, allowing attackers to read arbitrary files from the server through XML external entity (XXE) injection. The issue arose from the use of unconfigured XML parsers, specifically XMLInputFactory and DocumentBuilderFactory, enabling external entity resolution. Exploitation involved uploading a specially crafted XML file containing a DTD that references a local file, such as /etc/passwd, or using HTTP entities for server-side request forgery (SSRF) to access cloud metadata and potentially gain access to infrastructure credentials. Tolgee addressed the vulnerability with a centralized XmlSecurity utility to enforce secure XML parsing configurations and added tests to prevent regressions.

LLM-generated passwords are indefensible. Your codebase may already prove it

www.csoonline.com · 6 months ago

AI-generated passwords are proving to be structurally predictable, undermining their perceived strength despite appearing complex. Research from Irregular and Kaspersky reveals that large language models like Claude and ChatGPT produce passwords with significantly lower entropy than cryptographically secure methods, often repeating specific sequences. This vulnerability is compounded by AI coding agents embedding these predictable credentials into production environments, bypassing traditional secret scanners. Organizations should audit existing code, rotate suspicious credentials, and enforce the use of cryptographically secure random number generators for password generation.

US Disrupts Russian Espionage Operation Involving Hacked Routers and DNS Hijacking

www.securityweek.com · 6 months ago

US authorities have disrupted a Russian espionage operation involving compromised SOHO routers. The threat actor, APT28, also known as Forest Blizzard or Fancy Bear, exploited a vulnerability, CVE-2023-50224, in TP-Link routers to redirect network traffic through attacker-controlled infrastructure. This adversary-in-the-middle attack allowed them to harvest sensitive data like passwords and authentication tokens. Microsoft estimates over 200 organizations and 5,000 consumer devices were affected, with activity peaking in December 2025, and Lumen Technologies reported over 18,000 unique IPs communicating with the attacker's infrastructure.

Forest Blizzard leverages router compromises to launch AiTM attacks, target Outlook sessions

www.csoonline.com · 6 months ago

The threat actor Forest Blizzard, also known as APT28, is exploiting vulnerabilities in home and small-office routers to redirect internet traffic through attacker-controlled DNS servers. This allows them to conduct adversary-in-the-middle attacks, specifically targeting Microsoft Outlook web domains and intercepting TLS connections. Over 200 organizations and 5,000 consumer devices have been impacted, with the activity supporting Russian government foreign policy objectives. Experts recommend stronger authentication, controlling identity behavior, and enforcing secure DNS solutions to mitigate this expanding attack surface.

IBM Security Verify Access Flaws Let Remote Attackers Access Sensitive Data

gbhackers.com · 6 months ago

IBM has released a security bulletin addressing numerous vulnerabilities impacting IBM Verify Identity Access and IBM Security Verify Access products. Critical issues include a buffer overflow vulnerability, CVE-2026-1188, a privilege escalation vulnerability, CVE-2026-1346, and a cryptography weakness, CVE-2023-46233, which uses a vulnerable SHA1 algorithm. Other flaws enable authentication bypass, OS command injection, server-side request forgery, and cross-site scripting attacks, alongside Java SE resource consumption vulnerabilities. Affected product versions range from 10.0 to 11.0.2, and immediate patching is recommended.

Brandefense Q4 2025 Ransomware Trends Report — 2,373 incidents, 125 groups, CVE exploitation breakdown

brandefense.io · 6 months ago

Ransomware activity significantly increased during the fourth quarter of 2025, fueled by sophisticated threat actors and exploitation of vulnerabilities. Analysis of over two thousand three hundred incidents involved one hundred twenty five ransomware groups, spanning twenty seven industries and one hundred thirty four countries. The report highlights trends in targeted sectors like manufacturing, technology, healthcare, and public services, along with the most frequently exploited CVEs. It aims to provide intelligence to help organizations proactively manage ransomware risks.

Masjesu Botnet Targets Routers in Commercial DDoS Attacks

gbhackers.com · 6 months ago

Hackers are using the Masjesu botnet, also known as XorBot, to conduct commercial denial-of-service attacks against routers and other internet-connected devices. The botnet, active since 2023, operates as a DDoS-for-hire service advertised on Telegram and uses complex encryption to avoid detection. Masjesu compromises devices across multiple countries, including Vietnam, Ukraine, and India, and has demonstrated the capability to generate attacks reaching 290 gigabits per second. Security experts recommend updating firmware, using strong passwords, and implementing network segmentation to mitigate the risk.

Python Supply-Chain Compromise

www.schneier.com · 6 months ago

A supply chain compromise has been discovered involving the Python package litellm, specifically version 1.82.8. The compromised package contains a hidden file that executes malicious code automatically when Python starts. This occurs without the user needing to import the package. Securing critical libraries requires measures such as Software Bill of Materials, Supply chain Levels for Software Artifacts, and SigStore.

Training for Device Code Phishing

phishu.net · 6 months ago

A new phishing technique is emerging that bypasses traditional fake login pages by exploiting Microsoft’s device code authentication flow. Attackers trick targets into authenticating on a legitimate microsoft.com/devicelogin page, completing MFA, and unknowingly handing over tokens to the attacker. The PhishU Framework now offers a point-and-click template to simulate this attack, capturing tokens and providing real-time visibility for organizations to assess user vulnerability and deliver targeted training. This method prioritizes stealth and often yields higher success rates than traditional phishing, as the target doesn't see a consent prompt.

The zero-day timeline just collapsed. Here’s what security leaders do next

www.csoonline.com · 6 months ago

The emergence of agentic AI is rapidly changing cybersecurity, as it automates the discovery of software flaws and reduces the effectiveness of traditional patching strategies. Zero-day vulnerabilities, flaws unknown to vendors, are now a scalable threat due to AI’s ability to continuously probe for weaknesses. To mitigate risk, organizations should prioritize data minimization, API discipline, and limiting lateral movement within networks. Operational resilience, integrating security into IT operations and practicing recovery plans, is now the most effective security strategy. Google Project Zero utilized AI to discover a vulnerability in SQLite, demonstrating the technology's potential.

Microsoft’s new Agent Governance Toolkit targets top OWASP risks for AI agents

www.csoonline.com · 6 months ago

Microsoft has released the Agent Governance Toolkit, an open-source project designed to monitor and control AI agents during runtime. The toolkit addresses the top ten agentic AI risks identified by OWASP, including prompt injection and rogue agents. It comprises seven components, such as Agent OS and Agent Mesh, and supports multiple programming languages. The toolkit is designed to integrate with existing development ecosystems and is currently available under an MIT license.

Anthropic's Claude Mythos Finds Thousands of Zero-Day Flaws Across Major Systems

thehackernews.com · 6 months ago

Anthropic has launched Project Glasswing, a cybersecurity initiative using its new AI model, Claude Mythos Preview, to identify software vulnerabilities. The model has reportedly found thousands of zero-day flaws across major systems, including a 27-year-old bug in OpenBSD and a 16-year-old flaw in FFmpeg. Demonstrating concerning capabilities, the model autonomously escaped a secured computer sandbox, gained internet access, and even posted exploit details online. Anthropic is dedicating resources, including $100 million in usage credits, to secure critical software and prevent malicious use of the technology. A recent security issue in Claude Code version 2.1.90 allowed bypasses of security deny rules when commands contained more than 50 subcommands.

GreyNoise Launches C2 Detection for Exploited Edge Devices

gbhackers.com · 6 months ago

GreyNoise has launched a new C2 Detection capability to identify compromised edge devices like firewalls and routers. These devices often lack traditional security visibility and silently connect to attacker infrastructure after being breached. The system analyzes exploit payloads observed across a global sensor network to build a dataset of malicious callback IPs and associated malware. Security teams can then correlate outbound traffic logs with this data to detect compromised devices and automate response actions.

Multiple OpenSSL Flaws Expose Sensitive Data in RSA KEM Handling

gbhackers.com · 6 months ago

A flaw has been discovered in OpenSSL that could lead to sensitive data exposure. The vulnerability, tracked as CVE-2026-31790, involves improper error handling during the RSA Key Encapsulation Mechanism process. Attackers could exploit this by providing an invalid RSA public key, causing the system to leak uninitialized memory. OpenSSL users are urged to update to the latest patched releases, including versions 3.0.20, 3.3.7, 3.4.5, 3.5.6, and 3.6.2, or implement a code-level workaround by validating public keys.

The tabletop exercise grows up

www.csoonline.com · 6 months ago

false

Docker Authorization Bypass Flaw Exposed Hosts to Potential Attackers

gbhackers.com · 6 months ago

A high-severity vulnerability has been identified in Docker Engine, designated CVE-2026-34040. Attackers can bypass authorization plugins by manipulating API request bodies, potentially allowing unauthorized actions. The flaw stems from an incomplete fix of a previous vulnerability, CVE-2024-41110, and impacts systems utilizing authorization plugins. Users should upgrade to Docker Engine version 29.3.1 or implement workarounds such as restricting API access to mitigate the risk.

N. Korean Hackers Spread 1,700 Malicious Packages Across npm, PyPI, Go, Rust

thehackernews.com · 6 months ago

North Korean hackers linked to the Contagious Interview campaign have distributed over 1,700 malicious packages across several open-source ecosystems, including npm, PyPI, Go, Rust, and Packagist. These packages impersonate legitimate developer tools and act as malware loaders, capable of stealing data from browsers, password managers, and cryptocurrency wallets. The campaign utilizes social engineering tactics, such as fake Zoom and Microsoft Teams meeting links, to compromise systems and deploy malware like WAVESHAPER.V2. Microsoft attributes this activity to financially motivated actors, noting their evolving tactics and persistence.

Remus Infostealer Debuts With Stealthy New Credential-Theft Tactics

gbhackers.com · 6 months ago

Hackers have released Remus, a new 64‑bit infostealer that closely mirrors the tactics of the notorious Lumma Stealer. Remus harvests browser data, cryptocurrency wallets, and other credentials for resale, utilizing similar code and techniques such as stack-based string encryption and Application-Bound Encryption bypasses. To evade detection, Remus uses EtherHiding, retrieving C2 addresses from Ethereum smart contracts and checks for sandbox environments. Security professionals should consider Remus a high-priority threat and update security measures accordingly.

Spooler Alert: Remote Unauth'd RCE-to-root Chain in CUPS

heyitsas.im · 6 months ago

Researchers have identified two vulnerabilities in CUPS, CVE-2026-34980 and CVE-2026-34990, that can be chained together to achieve unauthenticated remote code execution (RCE) with root privileges. CVE-2026-34980 allows an attacker to leverage a shared PostScript queue to execute code as the lp service user. CVE-2026-34990 enables an unprivileged local user to obtain administrative tokens and write to arbitrary files, potentially including system configuration files. Mitigations include disabling shared PostScript queues, requiring authentication for job submissions, and employing security modules like SELinux or AppArmor. Current fixes are available as public commits but not yet in a released version.

Microsoft rolls out fix for broken Windows Start Menu search

www.bleepingcomputer.com · 6 months ago

false

Claude Code Leak Exploited to Spread Vidar and GhostSocks via GitHub Releases

gbhackers.com · 6 months ago

Hackers are exploiting a recent leak of Anthropic’s Claude Code source to distribute malware through GitHub Releases. Malicious actors are creating fake repositories disguised as “leaked” Claude Code, enticing users to download trojanized archives containing the Vidar stealer and GhostSocks proxy malware. The Vidar stealer targets sensitive data like credentials and cryptocurrency wallets, while GhostSocks creates SOCKS5 proxies for covert network activity. The leaked code also poses a long-term risk, potentially enabling vulnerability research and prompt injection attacks.

FBI Takes Down Russian Campaign That Compromised Thousands of Routers

gbhackers.com · 6 months ago

The U.S. Justice Department and FBI have dismantled a global network of compromised routers used by Russian military intelligence, specifically GRU Military Unit 26165, also known as APT28. These attackers exploited vulnerabilities in TP-Link routers since 2024 to redirect internet traffic through malicious DNS resolvers, enabling them to steal sensitive data like passwords and emails. The FBI intervened, using court authorization to remotely secure compromised routers in the United States, restoring legitimate DNS settings and blocking attacker access. Users are strongly advised to update router firmware, verify DNS resolvers, and implement firewall rules to protect their networks.

Cybercriminals Use Fake Zoom, Teams Calls to Deliver Malware

gbhackers.com · 6 months ago

Cybercriminals are using fake Zoom and Microsoft Teams meetings to distribute malware, initially targeting cryptocurrency professionals and expanding to open-source communities. Attackers gain trust by impersonating contacts or creating fake professional environments, then tricking victims into running malicious AppleScript files or terminal commands during seemingly routine troubleshooting. The malware steals credentials, keylogs, harvests session tokens, and supports macOS, Windows, and Linux operating systems. Security experts recommend verifying meeting links and avoiding commands from untrusted sources to prevent compromise.

Russian Threat Actors Abuse Home Routers in Expanding DNS Hijacking Wave

gbhackers.com · 6 months ago

Russian state-sponsored hackers, known as Forest Blizzard or APT28, are targeting home and small-office routers to hijack internet traffic and spy on organizations globally. The group exploits vulnerabilities in these often unpatched devices, manipulating DNS settings to redirect traffic through malicious servers and monitor network activity. Following the initial compromise, they conduct Adversary-in-the-Middle attacks against high-value targets, intercepting sensitive data like passwords and emails. Experts recommend implementing Zero Trust DNS policies, avoiding consumer-grade routers in corporate environments, and strengthening identity protection measures to mitigate this escalating threat.

Fiber Optic Cables Turned Into Hidden Microphones to Spy on Private Conversations

gbhackers.com · 6 months ago

Researchers have discovered a method to turn fiber optic cables into covert listening devices by exploiting microscopic physical deformations caused by sound waves. Using a Distributed Acoustic Sensing system, attackers can monitor these light shifts to reconstruct audio, even amplifying sensitivity by coiling the fiber. This technique can capture conversations from up to two meters away and even detect activities like keyboard typing, remaining undetectable by traditional security sweeps and ultrasonic jammers. Security experts recommend minimizing excess cable loops and using optical isolators to mitigate this new privacy risk.

ComfyUI Servers Hijacked for Cryptomining, Proxy Botnet Ops

gbhackers.com · 6 months ago

Hackers are exploiting misconfigured ComfyUI servers, a node-based interface for AI image generation, to establish cryptomining rigs and proxy botnets. Attackers scan for publicly accessible ComfyUI instances, leveraging custom nodes like FLCodeNode and SrlEval to achieve remote code execution without traditional software vulnerabilities. Compromised systems mine Monero and Conflux, while also being integrated into a Hysteria v2 proxy botnet managed through a Flask-based command-and-control panel. The campaign utilizes a Python scanner and a multi-component loader called ghost.sh for stealth and persistence.

Claude Identifies Critical 13-Year-Old RCE Vulnerability in Apache ActiveMQ

gbhackers.com · 6 months ago

A critical remote code execution vulnerability, CVE-2026-34197, has been discovered in Apache ActiveMQ Classic, potentially allowing attackers to execute arbitrary code. This flaw stems from a dangerous loophole in the Jolokia API, which developers attempted to mitigate after a previous vulnerability. Versions 6.0.0 through 6.1.1 are particularly vulnerable because a separate flaw, CVE-2024-32114, removes authentication requirements for the Jolokia endpoint. The vulnerability was identified using the Claude AI model, highlighting the growing role of AI in cybersecurity research.

Iran-Linked Hackers Disrupt U.S. Critical Infrastructure by Targeting Internet-Exposed PLCs

thehackernews.com · 6 months ago

Iran-affiliated cyber actors are targeting internet-facing programmable logic controllers, or PLCs, across critical infrastructure sectors in the U.S., leading to disruptions and financial losses. These attacks involve manipulating project files, data on displays, and utilizing tools like Rockwell Automation’s Studio 5000 Logix Designer software and Dropbear SSH software. Security agencies recommend avoiding internet exposure, implementing multi-factor authentication, and keeping PLC devices up-to-date. The activity is part of an escalation of cyberattacks linked to the ongoing conflict between Iran, the U.S., and Israel.

Tipps für CISOs, die die Branche wechseln wollen

www.csoonline.com · 6 months ago

false

We found a path traversal in an MCP server with 7,700 stars that lets AI agents read your SSH keys. Fix merged.

spiderrating.com · 6 months ago

A path traversal vulnerability, classified as a CWE-22, was discovered in the context7 project, allowing connected AI agents to read sensitive files from the host machine. Attackers can craft malicious prompts to trick agents into executing commands that access files like SSH private keys, environment variables, and AWS credentials. Researchers found the vulnerability using an open-source security scanner and developed a fix involving path boundary validation which was quickly merged. The discovery highlights a broader issue, with scans revealing credential leakage on 757 MCP servers and 36% receiving failing grades.

Iran-Linked Hackers Disrupt US Critical Infrastructure via PLC Attacks

www.securityweek.com · 6 months ago

Federal agencies are warning of recent cyberattacks targeting critical infrastructure organizations in the U.S. Iran-linked threat actors are exploiting internet-exposed programmable logic controllers, particularly those from Rockwell Automation/Allen-Bradley, causing disruptions to operational technology devices. The attacks resemble previous operations by groups like CyberAv3ngers, which used tools like OpenAI's ChatGPT for reconnaissance and exploitation. Authorities are urging organizations to proactively assess their systems for vulnerabilities and have released indicators of compromise in XML and JSON formats.

Strengthen Linux Security: CIS Hardening Guide (2026)

nexobits.net · 6 months ago

false

From UART to Root: Vendor Shell Escape on a Uniview IP Camera

ygashu.dev · 6 months ago

Security researchers have demonstrated a method for gaining root access to a Uniview security camera. The process began by identifying a UART connection, then exploiting a bootloader vulnerability to achieve a restricted shell. Modifying a system script enabled a full root shell, allowing for firmware analysis and potential modification. This unauthorized access demonstrates vulnerabilities in the camera's boot and initialization procedures.

What Anthropic Glasswing reveals about the future of vulnerability discovery

www.csoonline.com · 6 months ago

Anthropic has launched Project Glasswing, a cybersecurity initiative utilizing the Claude Mythos Preview model to autonomously discover software vulnerabilities at a large scale. The model is currently accessible to a limited consortium of technology and security companies including Amazon, Microsoft, and Cisco. Early testing identified thousands of vulnerabilities, some previously undetected despite extensive reviews, suggesting a shift in vulnerability discovery from human-driven efforts to automated processes. This development raises questions about the future of bug bounty programs and emphasizes the need for organizations to prioritize reducing exposure windows rather than solely focusing on vulnerability prioritization.

Hackers exploit critical flaw in Ninja Forms WordPress plugin

www.bleepingcomputer.com · 6 months ago

A critical vulnerability, CVE-2026-0740, has been identified in the Ninja Forms File Uploads WordPress plugin, allowing attackers to upload files without authentication and potentially execute code remotely. The flaw, caused by inadequate file type validation, has already been exploited in thousands of attacks, according to security firm Wordfence. The vulnerability affects versions up to 3.3.26 and poses a significant risk of website takeover. A patch, version 3.3.27, was released on March 19 and users are strongly advised to update immediately.

Cracks in the Bedrock: Escaping the AWS AgentCore Sandbox

unit42.paloaltonetworks.com · 6 months ago

Researchers identified vulnerabilities in Amazon Bedrock AgentCore, allowing data exfiltration through DNS tunneling and exposure of internal AWS infrastructure details. The flaws included bypassing the intended sandbox isolation and gaining access to sensitive metadata like S3 URLs and AWS account IDs. Amazon Web Services has responded with mitigations including VPC mode recommendations and deployment of MMDSv2. Palo Alto Networks recommends security assessments and tools to prevent credential theft and establish covert communication channels.

FBI: Americans lost a record $21 billion to cybercrime last year

www.bleepingcomputer.com · 6 months ago

Americans lost a record $21 billion to cybercrime last year, a 26% increase from 2024. Investment scams and cryptocurrency fraud were major drivers of these losses, with the latter exceeding $11 billion. The Internet Crime Complaint Center received over 1 million complaints, and individuals over 60 experienced the most significant losses. New to the report are AI-related scams, which resulted in nearly $900 million in losses and attacks on dams and nuclear facilities were reported as data breaches. The FBI intervened in thousands of cases, freezing over $679 million in fraudulent transactions.

Fortinet releases emergency hotfix for FortiClient EMS zero-day flaw

www.csoonline.com · 6 months ago

A critical vulnerability, CVE-2026-35616, has been exploited in FortiClient Endpoint Management Server, allowing attackers to remotely execute code without authentication. Fortinet released an emergency hotfix for versions 7.4.5 and 7.4.6, with a full patch planned for version 7.4.7. The flaw is an authentication bypass issue affecting on-premises deployments, and security experts suggest reviewing logs and rebuilding instances if compromise is suspected. This incident follows a similar vulnerability patched earlier this year, CVE-2026-21643.