57 matches found
-SCTT-2026-33-0002-DWM-Visual-Field-Singularity
SCTT-2026-33-0002: تفرد المجال البصري لـ DWM 📡 التصنيف النظري المعرف: SCTT-2026-33-0002 الباحث: Americo Simoes SimoesCTT الفيزياء: النظرية 4.2 - الانتقال الطوري المضطرب TPT الثابت: α = 0.0302011 الهدف: مدير سطح المكتب dwm.exe / مكوّن الرسومات في Windows يُلغي: CVE-2026-20805 و CVE-2026-20871...
ASM-Obfuscator
ASM-Obfuscator مُعقِّد متقدم لأكواد تجميع x86-64. الوصف تُعد أداة تعقيد أكواد التجميع Assembly Code Obfuscator أداة قوية ومتعددة الاستخدامات صُممت لتعزيز أمان أكواد تجميع x86-64 لديك. وهي تحمي مشاريعك من الهندسة العكسية والتحليل من خلال توظيف مجموعة واسعة من تقنيات التعقيد وإجراءات مكافحة التحليل...
AETHER-PROTOCOL-P
⬡ بروتوكول إيثر — شركة Aether AI LLC. 2026 بنية تحتية للهويات قصيرة العمر مُتحقق منها تشفيريًا للأنظمة الذكية المستقلة براءة اختراع قيد الإيداع — طلب أمريكي 64/006,746 — مُقدَّم في 16 مارس 2026 "يمكن اختراق كل نظام ذكاء اصطناعي في العالم عن طريق حقن تعليمات زائفة في خط معالجته. بروتوكول إيثر يجعل...
CVE-2026-95985
The file write tool in Amazon Kiro IDE versions before 1.0.242 might allow remote unauthenticated actors to inject crafted instructions into the agent's context. When a user runs the agent in a crafted repository as an untrusted workspace, sending any message can cause agent modifications to...
EUVD-2026-86308
The file write tool in Amazon Kiro IDE versions before 1.0.242 might allow remote unauthenticated actors to inject crafted instructions into the agent's context. When a user runs the agent in a crafted repository as an untrusted workspace, sending any message can cause agent modifications to...
CVE-2026-95985: Inclusion of Functionality from Untrusted Control Sphere
The file write tool in Amazon Kiro IDE versions before 1.0.242 might allow remote unauthenticated actors to inject crafted instructions into the agent's context. When a user runs the agent in a crafted repository as an untrusted workspace, sending any message can cause agent modifications to...
PT-2026-98205
Name of the Vulnerable Software and Affected Versions Amazon Kiro IDE versions prior to 1.0.242 Description The file write tool allows remote unauthenticated actors to inject crafted instructions into the agent's context. This occurs when a user runs the agent within a crafted repository as an...
Trustworthy RAG: An Evaluation Agent for Detecting Misinformation and Knowledge Poisoning in Generative AI Systems
Retrieval-Augmented Generation RAG grounds Large Language Model LLM outputs in external knowledge, but RAG systems usually trust whatever they retrieve, creating a Security-Reliability Gap: high semantic relevance does not guarantee factual truth. Adversaries exploit this through knowledge...
Piggybacking on Perception: Stealthy Concurrent Audio Prompt Injections against Multimodal LLM Agents
Large Language Model LLM-driven multimodal agents are increasingly deployed to execute autonomous tasks via continuous audio interaction. While this paradigm enhances interaction naturalness, it introduces a critical yet under-explored attack surface, as audio inputs inevitably contain...
Agent Data Injection Attacks Are Realistic Threats to AI Agents
AI agents act on behalf of user prompts, consuming external data and taking actions based on the agent context. Prior research on AI agent security has primarily focused on indirect prompt injection IPI. Its most well-studied category is instruction injection, where attacker-controlled untrusted...
CVE-2026-49185 Instruction Injection via FieldX MDM
The FieldX MDM adb messaging topic passes unverified payloads directly into Runtime.exec, allowing command/instruction injection...
CVE-2026-49185
The CVE-2026-49185 entry concerns FieldX MDM where the adb messaging topic passes unverified payloads directly into Runtime.exec(), enabling command/instruction injection. Affected component: adb messaging topic within FieldX MDM; root cause is unverified payloads executed via Runtime.exec(). Imp...
CVE-2026-46082 KVM: SVM: Inject #UD for INVLPGA if EFER.SVME=0
In the Linux kernel, the following vulnerability has been resolved: KVM: SVM: Inject UD for INVLPGA if EFER.SVME=0 INVLPGA should cause a UD when EFER.SVME is not set. Add a check to properly inject UD when EFER.SVME=0. sean: tag for stable@...
CVE-2026-46082
In the Linux kernel, the following vulnerability has been resolved: KVM: SVM: Inject UD for INVLPGA if EFER.SVME=0 INVLPGA should cause a UD when EFER.SVME is not set. Add a check to properly inject UD when EFER.SVME=0. sean: tag for stable@...
Linux Distros Unpatched Vulnerability : CVE-2026-41675
The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - xmldom is a pure JavaScript W3C standard-based XML DOM Level 2 Core DOMParser and XMLSerializer module. In @xmldom/xmldom prior to versions 0.9.10 and 0.8.13 an...
perl: corruption of intermediate language state of compiled regular expression due to integer overflow leads to DoS
Perl before 5.30.3 has an integer overflow related to mishandling of a "PLregkindOPn == NOTHING" situation. A crafted regular expression could lead to malformed bytecode with a possibility of instruction injection...
PT-2026-30532
Summary Android Canvas WebView pages from untrusted origins could invoke the JavascriptInterface bridge and inject instructions into the app. Affected Packages / Versions - Package: openclaw npm - Affected: = 2026.3.22 - Latest released tag checked: v2026.3.23-2...
Reverse CAPTCHA: Evaluating LLM Susceptibility to Invisible Unicode Instruction Injection
We introduce Reverse CAPTCHA, an evaluation framework that tests whether large language models follow invisible Unicode-encoded instructions embedded in otherwise normal-looking text. Unlike traditional CAPTCHAs that distinguish humans from machines, our benchmark exploits a capability gap: model...
Siemens SIMATIC S7-1500 Integer Overflow or Wraparound (CVE-2020-10878)
Perl before 5.30.3 has an integer overflow related to mishandling of a PLregkindOPn == NOTHING situation. A crafted regular expression could lead to malformed bytecode with a possibility of instruction injection. This plugin only works with Tenable.ot. Please visit...