Opera on DanSec Daily

Cybersecurity Brief – 2026-09-20

Sun, 20 Sep 2026 00:00:00 +0000

Major Incidents or Breaches

Three researchers at Hacktron used Anthropic’s Claude Opus 5 to chain two vulnerabilities, allowing them to take over ChatGPT and Codex accounts belonging to several OpenAI employees, ultimately gaining access to internal OpenAI systems. The attack leveraged Claude Opus 5 to facilitate the exploitation process ^[1].
Researchers successfully escaped the OpenAI Codex sandbox using two separate methods, one of which enabled them to execute commands on a developer’s machine even in the most restricted mode. OpenAI has since patched both vulnerabilities ^[4].
The ShinyHunters extortion group breached the Clop ransomware operation’s data leak site, defaced its Tor site, and reportedly stole server data and private keys for the onion service, threatening to extort the Clop gang ^[7].
The North Korean WaterPlum threat group compromised at least 30,000 devices globally between December 2025 and July 2026, transferring over $10.7 million in illicit funds. The campaign was disclosed in a joint law enforcement advisory ^[6].

Newly Discovered Vulnerabilities

Security researchers demonstrated that AI browser agents including those in Chrome, Edge, Opera Neon, Perplexity Comet, and Claude in Chrome can be hijacked through a malicious extension using the BragJack proof-of-concept attack. The attack allows full control over the AI assistant within the browser ^[5].

Notable Threat Actor Activity

The WaterPlum group, attributed to North Korea, conducted widespread infections and financial theft via a global campaign affecting tens of thousands of devices ^[6].
ShinyHunters, typically known for data extortion, targeted another cybercriminal group by breaching the Clop ransomware gang’s infrastructure and threatening further extortion ^[7].

Trends, Tools, or Tactics of Interest

Chaining of vulnerabilities involving AI models and browser assistants has enabled new attack paths, as demonstrated by both the Claude Opus 5 incident and the BragJack attack ^[1]^[5].
Exploitation of sandbox escape techniques remains a significant risk in AI development environments, as evidenced by the Codex sandbox escape ^[4].
Use of malicious browser extensions as a vector for hijacking AI assistants is an emerging threat highlighted by the BragJack proof-of-concept ^[5].

Sources

Claude Opus 5 Helped Researchers Take Over OpenAI Staff Accounts via Chained Flaws - thehackernews.com
Researchers escape OpenAI Codex sandbox to run commands on host - bleepingcomputer.com
ShinyHunters hacks Clop leak site, threatens to extort ransomware gang - bleepingcomputer.com
North Korean WaterPlum hackers infected 30,000 devices worldwide - bleepingcomputer.com
BragJack attacks hijack AI browser agents through malicious extensions - bleepingcomputer.com

Cybersecurity Brief – 2026-07-06

Mon, 06 Jul 2026 00:00:00 +0000

Major Incidents or Breaches

Researchers demonstrated a new attack, TrojPix, which enables data exfiltration from air-gapped systems by manipulating on-screen pixels to emit electromagnetic signals via video cables, allowing nearby receivers to capture sensitive information^[1].
A vulnerability in the Opera GX browser allowed malicious websites to silently install add-ons which could then be used to steal data from pages visited by the user^[3].

Newly Discovered Vulnerabilities

A flaw in Opera GX enabled automatic installation of browser mods from malicious sites, facilitating data theft from visited web pages without user interaction^[3].

Notable Threat Actor Activity

Security researchers identified a new Java-based remote access trojan, QuimaRAT, offered as malware-as-a-service and capable of targeting Windows, Linux, and macOS systems^[2].

Trends, Tools, or Tactics of Interest

The TrojPix attack highlights ongoing research into side-channel data exfiltration methods targeting air-gapped environments, specifically leveraging electromagnetic emissions from video cables^[1].
The QuimaRAT trojan demonstrates continued cross-platform targeting and the prevalence of malware-as-a-service models in the threat landscape^[2].
Researchers described SkillCloak, a technique that allows malicious AI agent skills to evade static code scanners by using self-extracting packing methods, enabling malware to bypass existing detection tools^[4].
Development of the Flipper Zero firmware is set to continue with increased reliance on community contributions, following a reduction in the internal development team^[5].

Sources

New TrojPix Attack Leaks Data From Air-Gapped Systems via Video Cable Emissions - thehackernews.com
Opera GX Flaw Let Malicious Sites Auto-Install Mods to Steal Data From Visited Pages - thehackernews.com
New Java-Based QuimaRAT MaaS Built to Run on Windows, Linux, and macOS - thehackernews.com
SkillCloak Lets Malicious AI Agent Skills Evade Static Scanners with Self-Extracting Packing - thehackernews.com
Flipper Zero firmware development continues with community help - bleepingcomputer.com

Cybersecurity Brief – 2026-07-03

Fri, 03 Jul 2026 00:00:00 +0000

Major Incidents or Breaches

Google, in coordination with the FBI and industry partners, disrupted the NetNut residential proxy network, which relied on millions of compromised home devices to anonymize cybercriminal and nation-state activity. Hundreds of domains associated with NetNut were seized, significantly degrading the network’s operations ^[4]^[25]^[26].
A dual US-Estonian citizen alleged to be a member of the Scattered Spider hacking collective was extradited to the United States to face charges ^[18].
European Parliament members investigating Pegasus spyware were themselves targeted, with Citizen Lab confirming Pegasus infection on one investigator’s phone ^[24].

Newly Discovered Vulnerabilities

A critical vulnerability in the Cursor AI code editor (DuneSlide) allows zero-click prompt injection attacks that escape the editor’s sandbox and can execute arbitrary code at the operating system level ^[27].
A new WinRAR security flaw has been patched that could allow attackers to take control of a user’s computer. The lack of automatic updates in WinRAR increases the risk of unpatched systems ^[37].
A vulnerability in Apple’s Hide My Email feature, first reported a year ago, remains unpatched and exposes user email addresses ^[35].

Notable Threat Actor Activity

Ransomware groups, including those associated with Anubis, are exploiting the Citrix Bleed 2 (CVE-2025-5777) vulnerability for initial access and are also leveraging bring-your-own-vulnerable-driver (BYOVD) and supply chain credential theft tactics ^[5]^[28].
The threat actor ToddyCat is deploying new Umbrij malware that abuses OAuth to access Gmail accounts via the Google API, enabling covert email access ^[7].
Attackers linked to the FortiBleed campaign have harvested credentials from hundreds of thousands of FortiGate firewalls and are collaborating with the INC and Lynx ransomware gangs, also exploiting a Nextcloud zero-day vulnerability ^[21]^[30].
Ransomware operators are running campaigns that impersonate Interpol, targeting small businesses globally using basic social engineering techniques ^[22].
KnowBe4 ThreatLabs tracked phishing campaigns targeting the FIFA World Cup 2026, with cybercriminals leveraging the event for credential theft and malware distribution ^[3].
ClickFix and ConsentFix attacks are hijacking Microsoft 365 accounts in seconds by exploiting OAuth flows and fake prompts to bypass MFA protections ^[13]^[36].
Fake Google and Cloudflare verification pages are distributing multiple malware families, including infostealers and a new malware loader, through ClickFix-style social engineering ^[36].

Trends, Tools, or Tactics of Interest

Immediate exploitation of newly disclosed vulnerabilities, such as CitrixBleed, is increasing, with attackers using public proof-of-concept code to target NetScaler appliances ^[28].
The first fully autonomous ransomware attack orchestrated by an AI agent (JADEPUFFER) was observed, exploiting a Langflow RCE vulnerability to automate a database ransomware campaign from start to finish ^[9].
ConsentFix and ClickFix attacks demonstrate a trend of using rapid OAuth token theft and social engineering to bypass MFA and compromise cloud accounts ^[13]^[36].
Opera has introduced a Paste Protect feature to mitigate ClickFix-style attacks by blocking malicious command execution via social engineering ^[17].
Context manipulation attacks, such as BioShocking, can trick agentic browsers into exfiltrating sensitive credentials by bypassing their safety guardrails ^[33].
Compromise assessment projects highlight persistent threats, missed incidents, and gaps in incident response across organizations, emphasizing the need for improved detection and response practices ^[1].
Identity lifecycle management models are being challenged by the proliferation of AI agents, which lack traditional employment records and departure dates, complicating access governance ^[8].
IBM and Red Hat have launched Project Lightwell, assigning 20,000 engineers to address open-source software supply chain risks identified by Anthropic’s Mythos AI bug-finding technology ^[23].
There is a shift in cybercrime pressure toward small and medium-sized businesses in Australia as larger institutions improve safeguards and regulatory compliance ^[19].

Regulatory or Policy Developments Affecting the Security Industry

The Court of Justice of the European Union dismissed Google’s final appeal against a €4.1 billion antitrust fine related to Android’s promotion of Chrome and search, upholding significant regulatory action against the company ^[12].
The Trump administration lifted restrictions on Anthropic’s Claude AI models following cybersecurity concerns, making Claude Fable 5 widely available again ^[31].
Apple has revised its long-standing patch policy to accelerate patch cycles in response to attackers leveraging AI to reduce time to exploit, indicating a move toward more rapid security updates ^[20].

Sources

Google Disrupts NetNut Residential Proxy Network Spanning 2 Million Home Devices - thehackernews.com
FBI Seizes NetNut Proxy Platform, Popa Botnet - krebsonsecurity.com
Google, FBI Disrupt NetNut Residential Proxy Network Powered by Millions of Devices - securityweek.com
Alleged Scattered Spider hacker extradited to the United States - bleepingcomputer.com
EU Politicians Investigated Pegasus Spyware. Then It Ended Up on One of Their Phones - wired.com
Critical Cursor AI Code Editor Flaws Could Lead to OS-Level Remote Code Execution - securityweek.com
WinRAR flaw could allow attackers to take control of your computer - malwarebytes.com
Apple’s Hide My Email doesn’t hide it very well - malwarebytes.com
Ransomware Groups Turn to Citrix Bleed 2, BYOVD, and Supply Chain Credentials - thehackernews.com
New CitrixBleed Vulnerability Exploited Immediately After Public Disclosure - securityweek.com
ToddyCat-Linked Umbrij Malware Abuses OAuth to Access Gmail via Google API - thehackernews.com
FortiBleed Actors Collaborating With Inc, Lynx Ransomware Gangs - darkreading.com
FortiBleed Campaign Linked to INC, Lynx Ransomware Attacks - securityweek.com
Ransomware Thugs Masquerade as Interpol to Entice Small Biz - darkreading.com
Cybercriminals Are Targeting the FIFA World Cup 2026 - blog.knowbe4.com
ConsentFix and ClickFix: How Microsoft 365 Accounts are Hijacked in 3 Seconds - bleepingcomputer.com
Fake Google and Cloudflare verification pages spread multiple malware families - malwarebytes.com
AI Agent Exploits Langflow RCE to Automate Database Ransomware Attack - thehackernews.com
Opera rolls out Paste Protect feature to fight ClickFix attacks - bleepingcomputer.com
‘BioShocking’ Attack Tricks AI Browsers Into Stealing Credentials - securityweek.com
Missed incidents, persistent threats, and response gaps: Insights from compromise assessment projects - securelist.com
Identity Lifecycle Management Wasn't Built for AI Agents - thehackernews.com
Anthropic's AI Finds Bugs. IBM Bets $5B It Can Fix Them. - darkreading.com
Aussies Face Reduced Cybercrime Risk, as Pressure Shifts to SMBs - darkreading.com
Google loses final appeal to overturn €4.1 billion EU fine - bleepingcomputer.com
Trump Administration Lifts Restrictions on Anthropic’s Claude Models After Cybersecurity Alarm - securityweek.com
Apple Reverses Age-Old Patch Policy to Keep Up With AI - darkreading.com