Collecting Cyber-News from over 60 sources

Tag: injection

Azure DevOps Prompt Injection Targets AI Coding Agents

Jul 22, 2026 6:09 PM

Tags: ai, data, injection

Hidden prompt injections in Azure DevOps can manipulate AI coding agents into accessing sensitive data using a developer’s permissions. First seen on esecurityplanet.com Jump to article: www.esecurityplanet.com/threats/azure-devops-prompt-injection-targets-ai-coding-agents/
CISA Warns WordPress Core SQL Injection Vulnerability Is Actively Exploited in Attacks

Jul 22, 2026 2:10 PM

Tags: attack, cisa, cve, cyber, cybersecurity, exploit, infrastructure, injection, kev, sql, vulnerability, wordpress

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has classified a critical SQL injection vulnerability in WordPress Core, tracked as CVE-2026-60137, as one of its Known Exploited Vulnerabilities (KEV) due to its active exploitation in real-world attacks. This vulnerability affects the core functionality of WordPress when themes or plugins fail to properly validate untrusted input…
Zimbra 10.1.20 Fixes Critical SNMP Command Injection and Multiple XSS Flaws

Jul 22, 2026 1:09 PM

Tags: cyber, flaw, injection, monitoring, update, xss

Zimbra has released version 10.1.20 of its Collaboration Suite (ZCS) to address multiple high-severity security vulnerabilities. This release includes a critical command injection flaw in the SNMP monitoring component and several cross-site scripting (XSS) issues affecting the Classic Web Client. The update, published on July 20, 2026, provides a permanent fix for a previously disclosed…
Microsoft Azure DevOps MCP Flaw Lets Hidden PR Comments Hijack AI Review Agents

Jul 22, 2026 9:09 AM

Tags: ai, flaw, injection, microsoft, tool

A single invisible comment in an Azure DevOps pull request can turn a reviewer’s own AI coding agent against them, driving it into projects the attacker has no rights to reach and quietly leaking what it finds.The flaw is in Microsoft’s official Azure DevOps MCP server, and it works because one of its tools returns…
Zimbra 10.1.20 patches multiple security issues, including a critical command injection bug

Jul 21, 2026 9:12 PM

Tags: flaw, injection, monitoring, vulnerability

Zimbra patched nine flaws in version 10.1.20, including a critical SNMP monitoring command injection issue enabling arbitrary command execution. Zimbra released version 10.1.20 to fix nine security vulnerabilities, including a critical command injection flaw in the SNMP monitoring component. The vulnerability affects systems with SNMP notifications enabled and could allow attackers to execute arbitrary commands.…
Zimbra Patches Critical SNMP Command Injection and Four XSS Vulnerabilities

Jul 21, 2026 5:11 PM

Tags: flaw, injection, monitoring, network, vulnerability, xss

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…
SSRF und Code Injection – SonicWall warnt vor Zero-Day-Angriffen auf SMA1000-Serie

Jul 21, 2026 8:09 AM

Tags: injection, zero-day

First seen on security-insider.de Jump to article: www.security-insider.de/sonicwall-sma1000-zero-day-ssrf-code-injection-aktiv-ausgenutzt-a-8f7983e5666a6c63dc3fb556e9e6c379/
GPT-5.6 Sol Ultra Discovers WordPress Pre-Auth SQL Injection Leading to RCE

Jul 20, 2026 4:09 PM

Tags: authentication, cyber, exploit, flaw, injection, rce, remote-code-execution, sql, vulnerability, wordpress

A critical vulnerability chain in WordPress, called wp2shell, that allegedly allows unauthenticated attackers to exploit a pre-authentication SQL injection flaw to achieve remote code execution (RCE) on typical WordPress installations running MySQL. Security researcher Adam Kues discovered this vulnerability chain using GPT-5.6 Sol Ultra during a multi-agent audit of the WordPress source code. GPT-5.6 Sol…
Prompt Injection Attacks Are Thwarting AI Hacking Agents

Jul 18, 2026 11:31 AM

Tags: ai, attack, hacking, injection, malicious

“Context bombing” tricks malicious AI agents into shutting down before they can do harm. First seen on wired.com Jump to article: www.wired.com/story/prompt-injection-attacks-are-thwarting-ai-hacking-agents/
CISA Adds FortiSandbox Bugs to KEV Catalog

Jul 17, 2026 10:31 PM

Tags: cisa, exploit, injection, kev, malware, remote-code-execution, update, vulnerability

Agencies Have Until Sunday to Patch Two Critical Command Injection Flaws. CISA added two critical FortiSandbox command injection vulnerabilities to its Known Exploited Vulnerabilities catalog after evidence of active attacks. Experts warn that unauthenticated remote code execution could let attackers compromise malware analysis systems and pivot deeper into enterprise networks. First seen on govinfosecurity.com Jump…
CISA Warns of Two Fortinet FortiSandbox Flaws Exploited to Execute Commands

Jul 17, 2026 12:31 PM

Tags: cisa, cve, cyber, cybersecurity, exploit, flaw, fortinet, infrastructure, injection, kev, unauthorized, vulnerability

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has added two critical vulnerabilities in Fortinet FortiSandbox to its Known Exploited Vulnerabilities (KEV) catalog. These flaws are actively being exploited in the wild to execute unauthorized commands on affected systems. The vulnerabilities, tracked as CVE-2026-39808 and CVE-2026-25089, involve OS command injection weaknesses (CWE-78) and impact FortiSandbox…
Prompt injection is becoming the XSS of the web agent era

Jul 17, 2026 8:31 AM

Tags: injection, xss

Autonomous web agents read whatever a page displays, and much of that content comes from strangers. Product reviews, seller listings, and advertisements sit beside trusted … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/07/17/xss-web-agent-prompt-injection/
Hackers Exploit SonicWall SMA1000 Zero-Days to Execute Commands as Root

Jul 16, 2026 3:31 PM

Tags: access, attack, cyber, detection, exploit, flaw, hacker, injection, vulnerability, zero-day

Hackers are actively exploiting two zero-day vulnerabilities in the SonicWall SMA 1000 Series remote access appliances. They are chaining a critical server-side request forgery flaw with a local code injection bug to execute commands with root privileges. Rapid7’s Managed Detection and Response team detected targeted attacks before SonicWall publicly disclosed these vulnerabilities on July 14,…
New Framework Redefines AI Penetration Testing Around Prompt Injection and Behavioral Objective Violations

Jul 16, 2026 3:31 PM

Tags: access, ai, cyber, data, framework, infrastructure, injection, penetration-testing, service, theft, unauthorized

A newly proposed framework argues that AI penetration testing must move beyond conventional infrastructure compromise and assess whether an adversary can make an AI-enabled system act against its intended operational purpose. Traditional penetration testing typically measures compromise through outcomes such as unauthorized access, privilege escalation, data theft, service disruption, or persistence. Those outcomes remain critical…
New Agent Data Injection Attack Can Make AI Agents Misclick or Run Attacker Commands

Jul 16, 2026 3:31 PM

Tags: ai, attack, data, github, injection, jobs

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…
OpenAI Unveils GPT-Red AI Model That Automatically Finds Prompt Injection Vulnerabilities

Jul 16, 2026 2:31 PM

Tags: ai, cyber, data, email, exploit, injection, malicious, openai, RedTeam, theft, tool, unauthorized, vulnerability

OpenAI has introduced GPT-Red, an automated safety red-teaming model trained to identify and exploit prompt injection weaknesses in AI agents. Prompt injection occurs when malicious instructions hidden in webpages, emails, local files, code repositories, or tool outputs manipulate an AI system into ignoring its intended task, potentially causing data theft, unauthorized actions, or policy bypasses.…
OpenAI’s GPT-Red Automates Prompt Injection Testing to Harden GPT-5.6 Sol

Jul 16, 2026 11:31 AM

Tags: ai, attack, injection, intelligence, openai, RedTeam, tool, vulnerability

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…
Now, defenders are embracing the prompt injection, too

Jul 13, 2026 5:35 PM

Tags: hacking, injection

“Context bombing” tricks hacking agents into shutting down before they can do harm. First seen on arstechnica.com Jump to article: arstechnica.com/security/2026/07/now-defenders-are-embracing-the-prompt-injection-too/
Operation Capsule Vault Uses Malicious ISO Files and Process Injection to Deliver RokRAT

Jul 13, 2026 8:33 AM

Tags: conference, cyber, email, injection, malicious, phishing, spear-phishing

Operation Capsule Vault began with spear-phishing emails sent on June 22, 2026, posing as notices distributing materials from a legitimate academic event. The lures referenced the “Why Wonsan-Kalma Tourism Now?” conference, held at Seoul COEX on June 12, and incorporated publicly available event details, including its subject matter and host organizations. By reusing real-world conference…
New GhostCommit Technique Hides Exploits in Images to Evade AI Code Reviewers

Jul 13, 2026 6:33 AM

Tags: ai, cyber, exploit, group, injection, malicious

Researchers have revealed a technique called >>GhostCommit,<< which involves prompt injection by hiding malicious instructions within images included in pull requests. This technique has the potential to bypass text-only AI code reviewers and later manipulate coding agents into exposing repository secrets. The ASSET Research Group explained that this technique leverages the widening gap between automated…
Claude AI Prompt Injection Attack Turns Chatbot Into Stealthy C2 Agent to Achieve Remote Code Execution

Jul 9, 2026 7:16 AM

Tags: access, ai, attack, control, cyber, email, exploit, injection, malware, phishing, remote-code-execution

Claude Desktop’s synced Personal Preferences feature can be exploited as a covert prompt-injection vector, transforming the AI assistant into a de facto command-and-control (C2) agent. This method allows for remote code execution on a compromised user workstation without the need for phishing emails or traditional malware delivery. In this attack chain, the initial access is…
Ubiquiti Patches Critical UniFi OS Flaws Allowing Command Injection and Privilege Escalation

Jul 9, 2026 12:17 AM

Tags: cve, flaw, injection, update, vulnerability

Ubiquiti patched seven UniFi OS flaws, including critical CVE-2026-50746, which allows command injection in UniFi Connect Application. Ubiquiti released security updates for seven critical UniFi OS vulnerabilities, including a maximum-severity flaw, tracked as CVE-2026-50746 (CVSS score of 10.0), enabling command injection attacks. The issue affects UniFi Connect Application versions 3.4.16 and earlier, a platform used…
CrowdStrike Uncovers 5 New Prompt Injection Techniques Targeting AI Agents

Jul 8, 2026 1:17 PM

Tags: ai, attack, crowdstrike, cyber, injection

CrowdStrike has identified five new techniques for prompt injection targeting AI agents, emphasizing the rapid evolution of adversarial methods as enterprises increasingly deploy autonomous AI systems. Detailed in a report published on July 7, 2026, by CrowdStrike’s AI security research team, these techniques expand the company’s prompt injection taxonomy to over 200 documented attack methods.…
Ubiquiti warns of new max severity UniFi OS vulnerability

Jul 8, 2026 11:17 AM

Tags: exploit, flaw, injection, update, vulnerability

Ubiquiti has released security updates to patch seven critical vulnerabilities in UniFi OS, including a maximum-severity flaw that can be exploited in command injection attacks. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/security/ubiquiti-warns-of-new-max-severity-unifi-os-vulnerability/
GitLost: GitHub’s AI Agent Tricked Into Leaking Private Repository Data

Jul 7, 2026 4:20 PM

Tags: ai, data, flaw, github, injection

Noma Labs details GitLost, a prompt injection flaw that made GitHub’s AI agent expose private repo data through a crafted public issue and guardrail failures. First seen on hackread.com Jump to article: hackread.com/gitlost-github-ai-agent-leaking-repository-data/
Thousands of MCP Servers Found Vulnerable to File Access and Injection Attacks

Jul 7, 2026 1:16 PM

Tags: access, ai, attack, cyber, flaw, injection, sql, supply-chain, vulnerability

Thousands of Model Context Protocol (MCP) servers, widely used to connect large language models (LLMs) to external systems, have been found vulnerable to critical security flaws, including arbitrary file access, command injection, server-side request forgery (SSRF), and SQL injection, raising significant concerns about AI supply chain security. A large-scale analysis of 9,695 MCP servers across…
Indirect Prompt Injection in Web Content Targets AI Agents

Jul 6, 2026 6:16 PM

Tags: ai, crypto, injection

Zscaler found sites hiding prompt-injection text to manipulate AI agents into crypto payments First seen on infosecurity-magazine.com Jump to article: www.infosecurity-magazine.com/news/indirect-prompt-injection-web/
Hidden Web Prompts Trick AI Agents Into Sending Money

Jul 6, 2026 4:18 PM

Tags: ai, injection, malicious, risk

Hidden prompts on malicious websites trick AI agents into making payments or trusting fake sites, exposing new risks for autonomous AI workflows. Zscaler ThreatLabz documented two active campaigns that embed hidden instructions in web pages to manipulate AI agents, not human users, though those get caught too. The technique is called indirect prompt injection: malicious…
Gaslight Stealer Embeds Fake System Messages to Mislead AI Malware Analysts

Jul 6, 2026 8:16 AM

Tags: ai, blockchain, cyber, injection, jobs, macOS, malware, north-korea, social-engineering, software

A new macOS stealer, tracked as Gaslight and attributed to North Korean operators, demonstrates a worrying evolution in malware design: deliberate prompt-injection to mislead AI-driven security tools. Gaslight arrives as a standalone Mach-O executable commonly luring macOS users with faux meeting software, fake job recruitment materials, blockchain or gaming downloads, and developer test packages social-engineering…
KI-Exploit durch Prompt Injection – Cursor-Schwachstellen hebeln Sandbox aus

Jul 6, 2026 8:16 AM

Tags: ai, exploit, injection, vulnerability

First seen on security-insider.de Jump to article: www.security-insider.de/cursor-ide-duneslide-sicherheitsluecken-prompt-injection-sandbox-bypass-a-92a2a3d25ac1c9726ae9afde76020cd1/