Live

Intelligence Feed

Latest threat intelligence articles from trusted security sources, auto-processed to extract entities, IoCs, and TTPs.

Four Spy Groups Used the Same Chrome and Windows Exploit Kit Within a Week

2mo ago · hacker-news

Multiple espionage-motivated threat actors, including APT31 and several China-aligned clusters, have leveraged a previously undocumented exploit kit named BlueMoon to exploit unpatched vulnerabilities in Google Chrome and Microsoft Windows. The exploit chain combines CVE-2026-85046, CVE-2026-87491, and CVE-2026-85880 to achieve sandbox escape and local privilege escalation, delivered via phishing emails leading to malicious URLs. The kit downloads payloads such as the GemStone browser backdoor, ShadowPad, and Rust-based malware using DLL sideloading and in-memory execution techniques. The U.S. CISA has added all three CVEs to its KEV catalog, mandating federal agencies to patch them by mid-September 2026.

11 IoCs 1 Actors 1 Malware 1 CVEs
Nearly 1 in 10 Exposed LiteLLM Gateways Accepted the Example "sk-1234" Admin Key

2mo ago · hacker-news

Wiz Research identified that nearly 1 in 10 internet-facing LiteLLM gateways accepted the default example admin key 'sk-1234', exposing critical cloud IAM credentials and API keys. Attackers with access to the master key can read all model provider API keys, execute code via misconfigured endpoints, and exploit pass-through routing to harvest cloud instance metadata. Multiple CVEs have been documented, including authentication bypasses and post-authentication code execution flaws, some of which are already being exploited in the wild to deploy cryptocurrency miners and extract sensitive data. Despite patches, many deployments remain vulnerable due to misconfigurations and unrotated default keys.

4 CVEs
Trezor warns users of email provider breach, phishing attacks

2mo ago · bleeping-computer

Trezor has warned customers about a phishing campaign exploiting a breach of its third-party email provider. Users received fake 'critical security alert' emails from [email protected], claiming a vulnerability in STM32 microcontrollers used in Trezor wallets could expose recovery seeds. The emails were phishing attempts containing malicious links, and Trezor confirmed the domain used in the attack has been taken down. This incident follows prior breaches involving ShipMonk and a support ticketing portal, with attackers leveraging stolen data for further targeting.

1 Actors
CISA: WatchGuard RCE flaw now exploited in ransomware attacks

2mo ago · bleeping-computer

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has confirmed that ransomware gangs are actively exploiting a critical remote code execution vulnerability, CVE-2025-14733, in WatchGuard Firebox firewalls. The flaw allows unauthenticated attackers to execute arbitrary code remotely due to an out-of-bounds write issue, particularly affecting systems with IKEv2 VPN configurations. Despite patches being available since December 2025, nearly 9,000 vulnerable devices remain exposed online, posing ongoing risks to organizations. CISA added the vulnerability to its Known Exploited Vulnerabilities catalog and mandated federal agencies to patch it within one week.

Malicious Chrome and Firefox Extensions Steal Crypto Traders’ Session and Wallet Data

2mo ago · socket-dev

A coordinated campaign involving malicious Chrome and Firefox extensions has been targeting cryptocurrency traders using Axiom Trade and Padre (now Terminal) platforms. The extensions, including J7Tracker, VREO, and Orbit Tracker, steal authenticated session tokens, wallet data, and browser state by injecting malicious JavaScript modules into active trading sessions. Data is exfiltrated via browser navigation to attacker-controlled domains hosted on Vercel and bonto.run infrastructure, bypassing CORS restrictions. The threat actor uses repackaged extensions with cloned functionality and maintains persistence through rotating C2 infrastructure and new publisher accounts, posing a direct risk of account compromise and cryptocurrency theft.

10 IoCs
Off Guard: Breaking LiteLLM from authentication bypass to cloud compromise

2mo ago · wiz

Multiple critical vulnerabilities were discovered in LiteLLM, a popular open-source LLM gateway, enabling authentication bypass, remote code execution, and cloud credential theft. CVE-2026-59822 allows unauthenticated access to MCP endpoints via a Bearer token bypass, while CVE-2026-59821 enables post-authentication root-level RCE through unsandboxed custom code guardrails. A default master key (sk-1234) is accepted by 9.6% of public instances, allowing attackers to achieve admin access and exploit these vulnerabilities. Additionally, pass-through endpoints can be abused to exfiltrate cloud metadata and IAM credentials, especially when combined with default or missing authentication.

4 IoCs 3 CVEs
Veradigm warns of patient data breach after ransomware gang claims attack

2mo ago · bleeping-computer

Veradigm, a healthcare technology company, disclosed a data breach resulting from a cybersecurity incident at a third-party vendor, which led to unauthorized access to patient data via compromised API credentials. The Gentlemen ransomware group claimed responsibility, stating they exfiltrated 3.5 million patient records containing personally identifiable information, including names, addresses, Social Security numbers, and email addresses. The group threatens to leak the data if a ransom is not paid, with a deadline of September 11, 2026. While clinical data was not accessed, the breach involved sensitive personal information, and Veradigm has initiated incident response procedures and customer notifications.

1 IoCs
Skullcandy Dime 3 earbuds expose users to Bluetooth hijacking

2mo ago · bleeping-computer

The Skullcandy Dime 3 wireless earbuds (model S2DCW) running firmware version 1.0.0.28 are vulnerable to Bluetooth hijacking due to a missing authentication flaw in the Airoha Bluetooth Audio SDK. This high-severity vulnerability, tracked as CVE-2025-20701, allows nearby attackers to pair with the device without user interaction or PIN confirmation. Once paired, the attacker's device becomes trusted and can automatically reconnect, enabling audio hijacking, microphone access, and connection disruption. Although Skullcandy released a fix in firmware version 1.0.0.30, there is no user-accessible method to update existing vulnerable units, leaving them permanently exposed.

2 IoCs
AdaptHealth confirms 4.1 million people exposed in July cyberattack

2mo ago · bleeping-computer

AdaptHealth, a healthcare provider, confirmed a cyberattack in July 2026 that exposed the personal and health information of 4.1 million individuals. The breach resulted from a social engineering attack that compromised a third-party contractor's privileged account, allowing attackers to exfiltrate data from cloud-based systems. The ShinyHunters threat group claimed responsibility for the attack, which involved the theft of full names, contact details, demographic data, health insurance, and health information. The company detected the intrusion on June 5 and received a ransom demand on June 15, but has found no evidence of data misuse to date.

1 Actors
Cisco confirms CVE-2026-20079 Secure FMC flaw exploited in attacks

2mo ago · bleeping-computer

Cisco has confirmed active exploitation of a critical authentication bypass vulnerability, CVE-2026-20079, in its Secure Firewall Management Center (FMC) software. The flaw, rated CVSS 10.0, allows unauthenticated remote attackers to execute commands as root by sending crafted HTTP requests. Indicators of compromise, including a specific log entry referencing '/var/tmp/license.tmp', were observed as early as July 23, 2026, suggesting exploitation began before Cisco's public awareness in August. The U.S. CISA has added the vulnerability to its Known Exploited Vulnerabilities catalog, mandating federal agencies to patch by September 12, 2026.

1 IoCs
Alby Hub Critical Flaw Could Let Attackers Take Over Internet-Exposed Bitcoin Wallets

2mo ago · hacker-news

A critical vulnerability in Alby Hub, a self-hosted Lightning wallet, affects versions v1.7.0 through v1.18.5 and could allow attackers to take over internet-exposed wallets and potentially steal funds. The flaw exists when users expose the Hub's management interface to the public internet, contrary to intended deployment on private networks. Alby has confirmed one user was affected, though it is unclear if funds were lost. The company recommends immediately restricting external access and upgrading to version v1.24.0, while also changing the wallet's unlock password if exposed.

DeepSeek Harness Flaw Let AI Agents Disable Their Own File Sandbox Without Approval

2mo ago · hacker-news

A critical vulnerability in DeepSeek Harness, an open-source tool for running AI coding agents, allowed sandboxed agents to disable their own file sandbox by invoking a local web interface without authentication. The flaw, tracked as CVE-2026-82533, enabled an agent to switch its session to 'danger-full-access' mode via a single command, escaping file restrictions and potentially executing commands outside its workspace. The vulnerability affected versions 0.1.1-rc.2 and earlier, with the fix introduced in version 0.1.2-alpha.2, which added one-time token authentication for the local interface. Researchers from OX Research reported the issue, and community members had previously observed the same behavior before official disclosure.

1 CVEs
Infostealer Logs Expose Replayable AI Tokens That Can Bypass MFA

2mo ago · hacker-news

Information stealer malware such as Lumma Stealer and Vidar are harvesting session tokens and API keys from compromised systems, enabling threat actors to replay these credentials and gain unauthorized access to AI services like Google Gemini, OpenAI, Anthropic, and others, bypassing multi-factor authentication. A 7 GB infostealer dump analyzed by Okta contained 44,791 unique JSON Web Tokens (JWTs), 555 of which were likely tied to AI service authentication, along with 2,937 encrypted JWE structures, primarily from OpenAI. Attackers are using stolen tokens to access premium AI models and sell access on underground forums, leveraging anti-detect browsers like Camoufox to avoid detection. The abuse of valid, unexpired API keys and tokens—termed 'LLMjacking'—allows attackers to conduct espionage, resource theft, or run up AI service bills on victims.

3 IoCs 2 Malware
Google warns of new Chrome zero-day bug exploited in attacks

2mo ago · bleeping-computer

Google has patched a new Chrome zero-day vulnerability, CVE-2026-87491, which has been actively exploited in the wild. The vulnerability is a high-severity out-of-bounds write issue in the V8 JavaScript and WebAssembly engine, allowing remote attackers to execute arbitrary code within the browser sandbox by luring users to malicious HTML pages. Exploitation can lead to heap corruption, data exposure, or browser crashes. Google has not disclosed further details about the ongoing attacks to avoid exposing technical details before most users are patched.

4 CVEs
New Microsoft Defender 'ShieldCrash' zero-day grants SYSTEM access

2mo ago · bleeping-computer

An anonymous security researcher known as Nightmare Eclipse disclosed a new zero-day exploit named 'ShieldCrash' that exploits an incomplete patch for the previously patched Microsoft Defender privilege escalation vulnerability CVE-2026-69414 (ShieldBreak). The exploit allows attackers to achieve arbitrary file read with SYSTEM privileges on fully patched Windows 10, Windows 11, and Windows Server systems. This disclosure is part of an ongoing dispute between the researcher and Microsoft over bug bounty and vulnerability disclosure practices. Microsoft has not yet patched this newly revealed bypass.

Over 36,000 exposed Plex servers vulnerable to recent flaws

2mo ago · bleeping-computer

Over 36,000 internet-exposed Plex Media Server instances remain unpatched against multiple security vulnerabilities disclosed by Plex in September 2026. Affected versions include v1.43.2 and earlier. Plex has released version 1.43.3 to address these issues, but CVE identifiers have not yet been assigned, limiting visibility and response within the security community. Attackers could reverse-engineer the patches to develop exploits, especially given the history of prior vulnerabilities in Plex being exploited for credential theft and remote code execution.

N-able N-central Pre-Auth RCE Flaw Exploited in the Wild

2mo ago · hacker-news

A critical pre-authentication remote code execution vulnerability, CVE-2026-86218 (CVSS score: 10.0), in N-able N-central is being actively exploited in the wild. The flaw allows unauthenticated attackers to execute arbitrary code remotely, enabling full compromise of the platform used by MSPs and large IT organizations to manage downstream systems. CISA has added the vulnerability to its Known Exploited Vulnerabilities catalog, requiring federal agencies to patch by September 11, 2026. Although N-able released a fix in version 2026.3 Hotfix 4, evidence suggests prior exploitation of internet-facing instances, with potential links to ransomware threat actors.

Microsoft Patches Record 974 Flaws, Including Two Exploited Windows Zero-Days

2mo ago · hacker-news

Microsoft's September 2026 Patch Tuesday addressed a record 974 vulnerabilities across its software ecosystem, including two actively exploited Windows zero-day flaws. CVE-2026-85880 is a heap-based buffer overflow in Windows ALPC that enables local privilege escalation to SYSTEM privileges. CVE-2026-81963 is an improper link resolution vulnerability in the Windows Update Stack, also allowing local privilege escalation. Both vulnerabilities have been added to CISA's Known Exploited Vulnerabilities catalog, requiring federal agencies to patch by September 22, 2026.

SAP Patches CVSS 10.0 Kernel Flaw Enabling Unauthenticated Remote Code Execution

2mo ago · hacker-news

SAP has patched a critical CVSS 10.0 vulnerability, CVE-2026-44756 (OVERPASS), in the SAP kernel's Extended Passport (EPP) processing, enabling unauthenticated remote code execution with administrative privileges. The flaw arises from missing boundary validation during EPP deserialization, allowing attackers to execute arbitrary OS commands, compromise business data, and move laterally across SAP systems. The vulnerability is reachable via multiple internet-facing protocols without authentication, including web, SAP GUI, and RFC layers. Another critical flaw, CVE-2026-58240 (S4GET), in SAP NetWeaver Message Server also allows unauthenticated remote code execution on SAP S/4HANA systems.

Researcher Drops New Microsoft Defender PoC Showing ShieldBreak Patch Can Be Bypassed

2mo ago · hacker-news

Security researcher Chaotic Eclipse has released a proof-of-concept (PoC) exploit demonstrating a patch bypass for CVE-2026-69414, a vulnerability in Microsoft Defender previously known as ShieldBreak. The newly disclosed bypass, dubbed ShieldCrash, allows arbitrary file read with SYSTEM privileges on all supported versions of Windows despite Microsoft's patch in Malware Protection Engine version 1.1.26080.3. The researcher claims Microsoft failed to fully address the original vulnerability, leaving a code path still exploitable under specific conditions.

F5 BIG-IP APM Malware Injects a PHP Web Shell Into Memory, Evading Disk Scans

2mo ago · hacker-news

Malware exploiting CVE-2025-53521 on F5 BIG-IP APM appliances injects a PHP web shell directly into memory, evading traditional disk-based detection. The malware modifies the Apache binary at /usr/sbin/httpd and hooks into the Apache Portable Runtime to intercept loading of PHP modules, dynamically injecting the web shell when specific PHP scripts are loaded. Attackers can gain remote code execution without authentication, and the malware establishes persistence through modified system binaries and hidden sockets, with indicators including unexpected file modifications, SELinux disablement, and anomalous HTTP responses.

8 IoCs
New cPanel Flaw Lets a Hosting Account With Mail Privileges Run Code as Root

2mo ago · hacker-news

cPanel has patched a critical vulnerability, tracked as CVE-2026-67401, that allows an authenticated hosting account with mail privileges to execute arbitrary code as root due to an SQL injection flaw in the EmailTrack component. The vulnerability affects all supported versions of cPanel and WHM, enabling a local privilege escalation path from a regular user to full server compromise. While no public exploit or active attacks have been observed as of September 9, 2026, similar recent cPanel flaws have been exploited in the wild, including in ransomware campaigns. The CVE record for this flaw had not yet been published at the time of reporting, and no mitigations are provided for systems unable to update immediately.

Chrome V8 Zero-Day Exploited in the Wild Enables Code Execution Inside Sandbox

2mo ago · hacker-news

A zero-day vulnerability in Chrome's V8 JavaScript and WebAssembly engine, tracked as CVE-2026-87491, is being actively exploited in the wild. The flaw is an out-of-bounds write that allows a remote attacker to execute arbitrary code inside the browser sandbox via a crafted HTML page. Google has patched the vulnerability in Chrome versions 153.0.8010.36 and later, but has not disclosed specific details about the exploitation or the threat actor involved, likely to prevent further abuse before widespread patching.

4 CVEs
WeChat Zero-Click Worm Took Over Accounts on iPhone and Android via Incoming Calls

2mo ago · hacker-news

Researchers at Calif demonstrated a zero-click worm exploiting a vulnerability in WeChat that allows full takeover of a user's account via an incoming call, without requiring any interaction from the target. The exploit spreads laterally by leveraging compromised accounts to call other contacts, enabling rapid propagation across devices. The attack works even if the call is not answered, though declining the call stops that particular attempt. Tencent mitigated the exploit server-side and released client updates (Android 8.0.77, iOS 8.0.76) in August 2026, but has not issued a formal advisory or CVE. No real-world attacks have been observed, and technical details have been withheld pending a future conference presentation.

ShinyHunters hackers claim breach of Florida "DAVID" DMV database

2mo ago · bleeping-computer

The ShinyHunters extortion gang claims to have breached the Florida Department of Motor Vehicles' 'DAVID' (Driver and Vehicle Information Database) system via a password-reset vulnerability, stealing over 200,000 driver records. As proof, they published a screenshot of Jeffrey Epstein's DMV record containing sensitive personal information. The attackers claim they accessed the system using compromised accounts belonging to DMV employees and an FBI agent, then exfiltrated data by iterating through driver IDs. They state they have since lost access and that the vulnerability is being patched, while also indicating plans to disclose breaches of other state DMVs in the near future.

1 Actors
Microsoft September 2026 Patch Tuesday fixes 966 flaws, 2 zero-days

2mo ago · bleeping-computer

Microsoft's September 2026 Patch Tuesday addresses 966 security vulnerabilities, including two actively exploited zero-day flaws. The first, CVE-2026-81963, is an elevation of privilege vulnerability in the Windows Update Stack that allows attackers to gain SYSTEM privileges through improper link resolution. The second, CVE-2026-85880, is a heap-based buffer overflow in Windows Advanced Local Procedure Call (ALPC) that enables local privilege escalation to SYSTEM. Both vulnerabilities were actively exploited in the wild before patches were released, though technical details on exploitation are not disclosed. The update marks Microsoft's largest Patch Tuesday to date, attributed to the use of AI-powered vulnerability discovery.

Autonomous AI Agents Compromise Thousands of Credentials in Under Six Hours

2mo ago · hacker-news

A financially motivated threat actor known as TeamPCP (aka Altered Spider, UNC6780) has conducted large-scale software supply chain attacks targeting PyPI, npm, and Docker Hub, deploying credential stealers SANDCLOCK and DUSTMAKER. SANDCLOCK, used in March and April 2026, is a Python-based tool targeting Linux and Kubernetes environments with container escape capabilities, while DUSTMAKER is a cross-platform JavaScript payload focused on credential theft in CI/CD pipelines and includes AI-targeting techniques like prompt injection. The group exfiltrates API credentials and targets AI coding assistants, monetizing access through ransomware and data theft extortion networks.

2 IoCs 1 Actors
Slim Spider Steals Crypto Custody Secrets From Brazilian Financial Institution

2mo ago · hacker-news

Slim Spider is a financially motivated threat actor targeting Brazilian financial institutions since at least March 2026. The group conducts multi-stage intrusions into cloud environments to steal cryptocurrency custody secrets and access instant payment systems like Pix. They use custom Bash scripts to extract cloud credentials, implement cloud-native cryptographic signing via OpenSSL, and deploy backdoors such as MikeDor. Slim Spider also leverages malicious DevOps pipelines and maintains web-based panels for automating attacks, including endpoint scanning, email reconnaissance, and unauthorized Pix transactions.

1 IoCs
Adobe fixes critical Magento zero-day exploited to backdoor servers

2mo ago · bleeping-computer

Adobe has patched a critical zero-day vulnerability, CVE-2026-75650 (StyleSmuggler), actively exploited in the wild to backdoor Magento and Adobe Commerce servers. The flaw allows arbitrary code execution, and attackers have deployed PHP web shells to exfiltrate server data. A second threat actor using different tooling has also been observed exploiting the same vulnerability. Sansec discovered the attacks, which began at least on September 4, 2026, and recommend immediate application of Adobe's hotfix and credential rotation.

1 IoCs
ChatGPT Flaw Let a Planted Prompt Send a Victim's Gmail Data to Another Account

2mo ago · hacker-news

Check Point Research discovered a vulnerability in ChatGPT that allowed a maliciously crafted prompt to enable covert data exfiltration from a user's session. By exploiting a shared internal JFrog Artifactory service used for package caching, an attacker could establish a hidden communication channel between isolated ChatGPT containers across different accounts. This allowed the silent reading of Gmail data, chat history, and files from a victim's session and transmitting them to an attacker-controlled account without user consent or awareness. The vulnerability stemmed from improper isolation in the internal service and excessive container permissions, which allowed write access to shared metadata properties. OpenAI confirmed the issue and took the internal service offline, but no user-facing patch was required or released.

1 IoCs
← Previous Next →