Stories cluster related articles into a single narrative, linked to the underlying CVEs and affected products. 3,785 stories synthesized.
Interpol, with support from Group-IB, has dismantled the SniperDz phishing-as-a-service platform, which operated for over a decade enabling large-scale credential theft.
The Khmer Shadow campaign targets Cambodian government agencies using a VMware-signed binary to DLL-sideload the NIGHTFORGE loader, which deploys Havoc Demon for espionage.
Splunk and Palo Alto Networks released patches for severe vulnerabilities that could let attackers create or modify arbitrary files and access protected resources.
The FBI seized 13 domains allegedly used by Chinese intelligence to target current and former U.S. government and military personnel holding security clearances.
Unit 42 researchers unveil Behavioral Integrity Verification (BIV), an audit framework that detects hidden threats in third-party AI agent skills by comparing declared vs. actual behavior.
A new zero-day exploit named GreatXML abuses Microsoft Defender's offline scan to bypass BitLocker encryption and spawn a SYSTEM shell during Windows Recovery Mode boot.
Key findings • 30 malicious npm packages disclosed in a 26-minute window on June 11, 2026 • All packages were registered just 17 hours before disclosure — a fresh, automated typosquatting c…
OceanLotus (APT32) targeted Vietnamese infrastructure firms and stock investors in two campaigns using the SPECTRALVIPER backdoor, marking a shift toward domestic espionage.
Tenet Security researchers have disclosed 'agentjacking,' a novel attack that exploits AI coding agents' trust in Sentry error data to execute arbitrary code on developer machines.
ESET has tied the state-aligned threat actor OceanLotus (APT32) to a supply-chain attack on Vietnam's FireAnt financial platform and a prolonged intrusion into a transport infrastructure company, indicating a pivot toward domestic espionage.
Microsoft has resolved a known issue causing some Windows Server 2025 devices to boot into BitLocker recovery after installing the April 2026 security update.
ShinyHunters leaked over 450,000 email addresses from the University of Nottingham, which confirmed the breach.
A malicious backdoor in the LiteLLM PyPI package, downloaded nearly 47,000 times in three hours, delivered an autonomous attack bot and highlights OWASP's finding that prompt injection remains the top agentic AI security failure.
A directory traversal vulnerability in Allegra, tracked as CVE-2026-11442, allows authenticated remote attackers to read sensitive files on affected systems.
A heap-based buffer overflow in MATE Desktop's Atril document viewer (CVE-2026-52849) allows remote code execution via malicious EPUB files.
A low-severity out-of-bounds read vulnerability in Apache HTTP Server's mod_proxy_ajp module could allow attackers with prior AJP backend access to leak sensitive information.
A cross-site scripting vulnerability in Allegra's downloadAttachment functionality, tracked as CVE-2026-11443, allows remote attackers to execute arbitrary script in the application's context.
A numeric truncation vulnerability in Samsung's rlottie library, tracked as CVE-2026-8916, allows remote attackers to execute arbitrary code on affected systems.
A high-severity vulnerability (CVE-2026-6973) in Ivanti EPMM allows authenticated attackers to achieve remote code execution by injecting malicious Apache configuration directives.
Researcher 'Pliny the Liberator' jailbroken Anthropic's Claude Fable 5 within days of release, using multi-agent decomposition and Unicode tricks to bypass safety classifiers and generate exploit code.
State-linked cybercrime groups from North Korea and China are increasingly targeting financial firms and cryptocurrency assets in the Asia-Pacific region, with their illicit gains directly contributing to national GDP.
Key findings • CISA added CVE-2026-10520 to the KEV catalog on June 11, 2026, confirming active exploitation. • The flaw affects an Ivanti product; no ransomware association has been identifi…
Anthropic's latest AI model, Claude Mythos 5, can discover vulnerabilities and construct complex exploit chains, but remains incapable of orchestrating large-scale, autonomous cyber operations.
CISA has updated its Binding Operational Directive (BOD) 26-04, implementing a risk-based approach to federal vulnerability remediation, mandating a three-day patch window for critical threats and acknowledging the accelerating pace of AI-driven attacks.