Application Security | Breaking Cybersecurity News | The Hacker News

Hidden text on a web page was enough to make Kiro , AWS's agentic coding IDE, rewrite its own configuration file and run an attacker's code on a developer's machine, with no approval step able to stop it. Intezer, in research with Kodem Security, found that a request as ordinary as asking Kiro to summarize a page could end in remote code execution. AWS has patched the issue, and no CVE has been assigned to it. Kiro's safety model rests on a human clicking "allow." The agent can run shell commands, fetch URLs, and edit files, and the design assumes a developer reviews anything risky before it happens. That approval step is the security boundary, and the flaw let an attacker slip past it without the developer ever being offered a choice. The weak point was the file that tells Kiro which external tools to load. Kiro reads its list of Model Context Protocol servers, and the exact command used to start each one, from ~/.kiro/settings/mcp.json. When that fi...

Google's DeepMind on Tuesday announced the release of Gemini 3.5 Flash Cyber , a specialized artificial intelligence (AI) model built atop 3.5 Flash that's designed to discover, validate, and patch vulnerabilities quickly and efficiently. According to the tech giant, the model will be exclusively available to governments and trusted partners via CodeMender as part of a limited-access pilot program. CodeMender is an AI-powered agent for vulnerability discovery and patching that was unveiled by the company in October 2025. A Google DeepMind spokesperson told The Hacker News that there are plans to extend the model's capabilities to include red-teaming features and end-to-end enterprise defense. The lightweight model, per DeepMind, is both cost-efficient and highly capable alternative to large, costly cybersecurity-focused models. CodeMender can call upon 3.5 Flash Cyber "multiple times at high speed and low cost," allowing the AI agent to scan more code path...

Zimbra has rolled out fixes to address multiple critical security issues, including a command injection flaw in the Simple Network Management Protocol (SNMP) monitoring component. As many as nine security vulnerabilities have been patched in Zimbra 10.1.20 . Topping the list is a command injection vulnerability in the SNMP monitoring component when SNMP notifications are enabled. Also patched are four cross-site scripting (XSS) flaws in the Classic Web Client - A stored cross-site scripting (XSS) vulnerability that could allow malicious attachment filenames to execute script under specific conditions. An XSS vulnerability where crafted fields could execute a malicious script under specific conditions. An XSS vulnerability where a crafted field could execute a malicious script when rendered. An XSS vulnerability where crafted attachments could execute a malicious script when rendered. Separately, fixes have been released for a mail forwarding restriction bypass (CV...

An Android app that can draw over other windows and write to shared storage can slip instructions to the AI agent driving that phone, in text no human eye will ever see. Two more steps, and the same app is running commands on the PC driving the agent. Researchers demonstrated that chain, plus six other attacks, against five open-source mobile agent frameworks: AppAgent, AppAgentX, Mobile-Agent-v3, Open-AutoGLM, and MobA. Everyone fell to at least six of the seven. The paper went up on arXiv on July 1 and was revised on July 14. The authors are at Simon Fraser University, the Chinese University of Hong Kong, Shandong University, and the Xingtu Lab at Chinese security firm QAX. Nothing here has a CVE, and first author Zidong Zhang told The Hacker News the team has no evidence of the techniques being used outside a controlled setting. The Hacker News checked all five frameworks and found the screenshot paths, the shell call, and the broadcast fallback that the paper describes st...

Attackers have begun to exploit two critical vulnerabilities in WordPress that, when combined together, enable unauthenticated remote code execution (RCE) and complete compromise of vulnerable websites. The two security flaws, tracked as CVE-2026-63030 and CVE-2026-60137 , have been codenamed wp2shell . "By the early hours of Saturday morning (UTC), successful exploitation was already well underway, initially using public exploit code to exfiltrate hashed credentials, with remote code execution following once additional details were made public," Jake Knott, principal security researcher at watchTowr, told The Hacker News in a statement. "From our vantage point across a global client base, we are seeing widespread impact of this vulnerability across organizations of every size and every vertical." Telemetry data captured by KEVIntel shows that 13 unique IP addresses from Switzerland, Germany, the U.K., Indonesia, Lithuania, the Netherlands, and Singapore h...

Researchers at Sysdig have linked a second attack on the same Langflow server to JADEPUFFER , the AI-agent-driven operator it first documented earlier this month. The same operator has now been spotted deploying ENCFORGE , a new compiled Go ransomware designed to encrypt model weights, vector indexes, training datasets, and other AI infrastructure files across the host filesystem. The entry point did not change. Langflow versions before 1.3.0 expose the /api/v1/validate/code endpoint without authentication, allowing any remote attacker to execute arbitrary Python on the server. The flaw, CVE-2025-3248 , carries a CVSS score of 9.8 and has been in CISA's Known Exploited Vulnerabilities catalog since May 5, 2025. As The Hacker News reported earlier this month, the prior operation used throwaway Python code and MySQL's AES_ENCRYPT() function to encrypt and destroy data in Nacos (Alibaba's configuration server) and production databases. The new ENCFORGE payload re...

Threat actors are now exploiting a recently disclosed critical security flaw impacting ServiceNow AI Platform, according to Defused Cyber . In a post shared on X, the threat intelligence firm said it's observing in-the-wild exploitation of CVE-2026-6875 (CVSS score: 9.5), a sandbox escape vulnerability that could allow an unauthenticated user to run arbitrary code. Patches for the flaw were released by ServiceNow throughout June in the following versions - Brazil EA and Brazil GA Australia Patch 2 Zurich Patch 7b and Zurich Patch 9 Yokohama Patch 12 Hot Fix 1b and Yokohama Patch 13 Searchlight Cyber, which disclosed additional technical specifics, said it reported the issue on April 1, 2026, adding it allows a complete compromise of the ServiceNow instance as well as all connected proxy servers. Besides rolling out a fix, ServiceNow is "enhancing instance security by severely restricting the type of code that can run in sandbox contexts," securit...

The industry spent the initial months after Anthropic's April 7 Mythos reveal focused on volume . How many new CVEs would Mythos add to an already overloaded pipeline? How quickly would the flood of AI-driven discovery overwhelm triage capabilities? How long would it take adversaries to weaponize Mythos findings at scale? Those questions were and remain valid. Yet they all stop short of addressing the single metric that determines whether any of those vulnerabilities actually lead to a breach: the exposure window. The exposure window - the gap between the moment a vulnerability becomes exploitable and the moment your team fixes it - is the time an attacker has to do actual damage. That window is currently open far too wide. In 2025, the average eCrime breakout time dropped to 29 minutes . Even PCI DSS - the strictest compliance framework in the industry - allows 30 days to remediate a critical vulnerability. That's a 1,000-to-1 gap between how fast attackers move and how f...

Opening a crafted XZ archive in 7-Zip could let an attacker run code on the machine. The flaw, CVE-2026-14266 , is a heap-based buffer overflow in how the archiver processes XZ chunked data, and Trend Micro's Zero Day Initiative (ZDI) detailed it on July 15. A fix shipped on June 25 in 7-Zip 26.02 . The overflow lets an attacker "execute code in the context of the current process," per the advisory. The code runs with the token 7-Zip itself holds and gains no privileges of its own. On Windows, a normally launched 7-Zip runs under a filtered standard-user token even on an administrator account, so the attacker inherits those limited rights unless the program was started elevated. The bug came in from Landon Peng of Lunbun LLC, who reported it to 7-Zip on June 5. ZDI rates the flaw 7.0, or High, not the Critical several write-ups reached for. The full CVSS 3.0 vector is AV:L/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H . The AV:L makes it a local attack vector, not a network-re...

Cybersecurity researchers have flagged a new software supply chain attack codenamed SleeperGem targeting the Ruby ecosystem after three malicious gems were published to RubyGems with the end goal of serving additional payloads. The rogue gems are listed below - git_credential_manager (versions 2.8.0, 2.8.1, 2.8.2, 2.8.3) - Published on July 18, 2026 Dendreo (versions 1.1.3, 1.1.4) - Published on October 14, 2017 fastlane-plugin-run_tests_firebase_testlab (version 0.3.2) - Published on February 06, 2018 "Each malicious release is a loader," StepSecurity said in an analysis. "It fetches a second stage from an attacker controlled Forgejo host, checks whether it is running in a build system and skips if it is, and on a developer machine it drops a native daemon and installs persistence." One aspect of the attack that immediately stands out is that "git_credential_manager" impersonates the official Microsoft Git Credential Manager, while t...

F5 has shipped fixes for a critical nginx flaw that lets a remote, unauthenticated attacker trigger a heap buffer overflow in the worker process with crafted HTTP requests. CVE-2026-42533 was patched on July 15 in nginx 1.30.4 (stable) and 1.31.3 (mainline) , and in NGINX Plus 37.0.3.1; anyone on an earlier build should upgrade. Triggering it can crash or restart the worker, causing a denial of service; where ASLR is disabled or can be bypassed, F5 says it may also allow remote code execution. The overflow lives in nginx's script engine, the code that assembles strings from directives at request time. It only surfaces under a specific configuration: a regex-based map whose output variable is referenced in a string expression after a capture from an earlier regex match. Under that pattern the engine's two-pass evaluation comes apart. The first pass measures how many bytes the result needs and allocates a buffer to fit; the second pass writes the bytes in. Both read the ...

Updated July 18, 2026: the two flaws now carry CVE IDs, the full mechanism has been published, a persistent-object-cache condition has surfaced, and a working proof-of-concept is public. The story below reflects all of it. An anonymous HTTP request can run code on a WordPress site. The bug is in core, so a bare install with zero plugins is exploitable. Every 6.9 and 7.0 site was in range until Friday, when WordPress shipped 6.9.5 and 7.0.2 and enabled what it calls forced updates through its auto-update system. wp2shell is two bugs, not one, and both now carry CVE IDs. CVE-2026-63030 is the REST API batch-route confusion; CVE-2026-60137 is a SQL injection in WordPress core. Chained, they take an anonymous request all the way to code execution. Since Friday, the full mechanism has been published, and a working proof-of-concept has gone up on GitHub. Adam Kues at Assetnote, Searchlight Cyber's attack surface management arm, found the batch-route bug and reported it throug...

The European Commission on Thursday ordered Google to give rival AI assistants the same reach into Android that Gemini already has: the camera, the microphone, whatever is on screen, a wake word that fires with the display off, and the ability to drive other apps in the background by imitating taps and typing. Google has to ship it in the next major release, Android 18, and by 1 August 2027 at the latest. That is one of two binding specification decisions adopted on 16 July under the Digital Markets Act, six months after the Commission opened proceedings on 27 January. The second makes Google hand anonymised Search query, click, and ranking data to rival search engines, and to AI chatbots that do search, for a cost-based fee. Neither is a fine. Specification proceedings only say what a gatekeeper has to build; the Commission's separate power to open a non-compliance case , fines included, is untouched. Android carries around 60% of European mobile users. Five features g...

n8n , the workflow automation platform, handed out the wrong accounts at login. On Enterprise instances configured to trust more than one external token issuer, it matched an incoming JWT to a local user on the  sub  claim alone and ignored  iss . A valid token from issuer A carrying a  sub  that belongs to someone under issuer B logged you in as them. Their password never came into it. n8n shipped the fix on June 24. The flaw is tracked as  CVE-2026-59208 . The CVE record did not go public until July 9. n8n  credits the report  to the GitHub account bearsyankees , whose profile lists Strix, which makes an AI penetration testing agent. Strix  says  it pointed out that the agent at the token-exchange flow and found the identity-binding bug there. Two issuers, one account Token exchange is n8n's Enterprise route for  OEM partners who embed the product , an  RFC 8693 implementation  that spares their users a second...

Ask an AI agent to summarize the reviews on a product page, and a single planted review can make it click "Buy Now" instead. Ask a coding assistant to apply a maintainer's fix from a GitHub thread, and a fake comment can make it run a stranger's command on your computer. Neither trick hijacks the agent's task. Each one just corrupts the facts it trusts and lets it carry on with the job you asked for. That is the shape of a new class of attack laid out in a  paper posted July 6  by researchers from Seoul National University, the University of Illinois Urbana-Champaign, and Largosoft. They call it agent data injection , or ADI. The attacker's input gets dressed up as data the agent already trusts, like a sender's name or a button's ID, so it slips past most of the defenses built to stop prompt injection. The gap comes from how an agent reads. It takes in two kinds of things: instructions, meaning what you and the app's developer tell it to d...

Artificial intelligence (AI) is changing offensive security, but it has not changed the standard that matters most: a finding has to be proven before it becomes useful. AI-assisted tools can read code quickly, generate payloads, summarize attack surfaces, explain unfamiliar APIs, and run repetitive testing workflows at impressive speed. That is a real advantage for security teams. It also creates a new kind of pressure, because the industry can now produce more vulnerability-looking output than ever before. The problem is that output is not the same as evidence. A generated report can sound polished, include a severity rating, and even contain a proof-of-concept that looks reasonable at first glance. None of that proves the bug exists in the deployed environment. None of it proves exploitability, impact, or risk. In offensive testing, the hard part has never been writing something that sounds like a vulnerability report. The hard part is demonstrating what is actually true. That d...

OpenAI has disclosed details of GPT-Red , an internal automated red-teaming model that scales prompt injection vulnerability discovery with an aim to fix issues before the tools are deployed widely. "GPT‑Red is a strong red-teamer, and our previous models are highly vulnerable to its prompt injection attacks," the artificial intelligence (AI) company said . "We use GPT‑Red to adversarially train GPT‑5.6 , making it much more robust to prompt injections." The model works just like a human red-teamer. It sends a prompt, monitors how a GPT model responds, and iterates its way towards a malicious goal, such as uploading sensitive data to an external server. The development comes as adversarial prompt injections continue to be a persistent thorn in the flesh of large language models, which can be tricked into executing a carefully crafted instruction⁠ that can produce undesirable consequences. As agentic systems continue to be hooked to third-party data sources ...

Zoom has released security updates for a critical security flaw impacting Zoom Workplace for Windows that could facilitate account takeover. The vulnerability, tracked as CVE-2026-53412 (CVSS score: 9.8), affects Zoom Workplace for Windows before version 7.0.0 and Zoom Workplace VDI Client for Windows before version 7.0.10, 6.6.15, and 6.5.18 in their respective branches. "Improper Input Validation in Zoom Desktop Client for Windows and Zoom VDI Client for Windows may allow an unauthenticated user to conduct an account takeover via network access," Zoom said in an advisory released this week. The latest security fixes also address three high-severity flaws - CVE-2026-53411 (CVSS score: 7.8) - An improper input validation vulnerability in the Zoom Workplace VDI Plugin for Windows before version 6.6.14 that may allow an authenticated user to conduct an escalation of privilege via local access. CVE-2026-53410 (CVSS score: 7.0) - A time-of-check to time-of-use ...

Mozilla has released updates to address two critical flaws in Firefox for which it warned that exploit code has been published. The vulnerabilities are listed below - CVE-2026-15718 , an invalid pointer in the JavaScript: WebAssembly component CVE-2026-15719 , a site isolation in the DOM: Navigation component "We are aware that exploit code for this is public, however we are not aware of any attacks in the wild abusing this flaw," Mozilla said in an advisory. Both vulnerabilities have been addressed in Firefox version 152.0.6. The release comes as Google shipped fixes for 15 security flaws, including two critical use-after-free bugs in Ozone ( CVE-2026-15764 and CVE-2026-15765 ), a cross-platform abstraction layer that allows the browser to interact natively with various display servers and windowing systems. It supports Linux, ChromeOS, and Fuchsia. "Use after free in Ozone in Google Chrome on Linux prior to 150.0.7871.125 allowed a remote attacker...

A single approved marketing tag can quietly load fourth-party code your security team has never seen, granting full access to your forms, customer data, and checkout pages. This on-demand webinar reveals how this Approval Gap forms, and gives your team the blueprint to close it before an auditor, regulator, or attacker finds it first. The Reality of the Approval Gap It's a pattern every security and IT team recognizes: You ran the security review. You approved the vendor. You moved on.