New WhatsApp Video Call Trick Bypasses Android Lock Screens to Expose Your Photos
A newly disclosed WhatsApp flaw lets anyone answer a video call on a locked Android phone and, through the in-call background editor, browse the device's entire photo gallery...
High-Severity Cleo Harmony Bug Lets Attackers Forge Their Way to Admin Access
A high-severity flaw in Cleo Harmony's JWT refresh-token handling, tracked as CVE-2026-84115, lets remote attackers escalate privileges to admin level with a working exploit already public. Cleo has...
Attackers Exploit Critical Langflow and Rails Flaws to Hunt Cloud Secrets
Attackers are actively exploiting critical flaws in Langflow and Ruby on Rails, with observed activity focused on credentials, application secrets and paths to remote code execution. Defenders should...
Critical Artifactory Bypass Is Being Used to Mint Administrator Tokens
A critical JFrog Artifactory authentication bypass is under active exploitation, with attackers reportedly creating administrator tokens on vulnerable servers. Self-hosted customers should upgrade immediately, revoke suspicious credentials and...
BGP Hijack Poisoned Virtualizor Updates and Put Hosting Servers at Root-Level Risk
A routing hijack diverted Softaculous infrastructure and enabled a malicious Virtualizor update to reach a small number of hosting servers. Because update packages lacked cryptographic verification, valid TLS...
D-Link Fixes Router Flaws That Exposed Admin and Wi-Fi Credentials on Local Networks
D-Link has patched access-control failures in the DIR-X1860Z that could let an unauthenticated local attacker reset the administrator password and retrieve wireless credentials. Owners should install the corrected...
Public HardBreacher Code Claims Kaspersky Privilege Escalation, but Evidence Remains Unverified
A public proof of concept called HardBreacher claims a local privilege-escalation weakness in Kaspersky Endpoint Security on Windows 11. The report remains unconfirmed, has no CVE, and is...
AI Gateways Under Fire: Attackers Chain LiteLLM and MCP Flaws for Remote Code Execution
A 90-day honeypot study shows attackers systematically probing AI proxies, MCP servers, and agent frameworks like LangChain and Langflow, chaining authentication bypass and command-injection bugs into remote code...