743 matches found
CVE-2026-5241
A flaw was found in python-transformers. An attacker can exploit this vulnerability by providing a malicious model repository. During model initialization, the trustremotecode parameter, intended to prevent remote code execution, is overridden by untrusted configuration data. This allows the...
MAL-2026-5175 Malicious code in webpack-json (npm)
--- -= Per source details. Do not edit below this line.=- Source: ghsa-malware abd3559fc62e362d5e4d5068126317096f7e2e483d97bba9f59e192a9d49a363 Any computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be...
Backdoor Unlearning Generalization: A Path toward the Removal of Unknown Triggers in LLMs
Backdoor attacks in Large Language Models LLMs are a growing security concern, where models can generate adversary-chosen content. Existing defenses target backdoors one at a time and typically require knowledge of the trigger, leaving the defender at a structural disadvantage when unknown...
Malicious code in @redhat-cloud-services/rule-components (npm)
Part of the "Mini Shai-Hulud" supply chain worm campaign that compromised the GitHub Actions OIDC trusted publisher shared by Red Hat Cloud Services npm packages. The attacker injected a preinstall hook into this and 31 other packages in the @redhat-cloud-services scope. The hook delivers a...
CVE-2026-10072 Interinfo｜DreamMaker - Arbitrary File Upload
DreamMaker developed by Interinfo has an Arbitrary File Upload vulnerability, allowing privileged remote attackers to upload and execute web shell backdoors, thereby enabling arbitrary code execution on the server...
CVE-2026-10071 Interinfo｜DreamMaker - Arbitrary File Upload
DreamMaker developed by Interinfo has an Arbitrary File Upload vulnerability, allowing unauthenticated remote attackers to upload and execute web shell backdoors, thereby enabling arbitrary code execution on the server...
From Prompt Injection to Persistent Control: Defending Agentic Harness against Trojan Backdoors
LLM agents are evolving from conversational chatbots to operational tools in real-world workspaces. In local agentic harnesses, an LLM can read and write files, call tools, and reuse workspace state across sessions. While such capabilities enhance utility, they also expose a new attack surface fo...
Dokploy 安全漏洞
Dokploy is an open-source software developed by Dokploy itself. Versions of Dokploy prior to 0.26.5 contain security vulnerabilities. These vulnerabilities stem from path traversal during the application deployment process, which allows authenticated users to write arbitrary files on the file...
CVE-2026-3623
IBM Netezza Performance Server Replication Services 3.0.2.0 through 3.0.5.0 allows an attacker with low‑privileged access to escalate their privileges to root. By exploiting this flaw, the attacker can execute root‑level commands, obtain a root shell, and change the root user’s password. Successf...
CVE-2026-3623 Vulnerabilities exists in IBM Netezza Performance Server Replication Services
IBM Netezza Performance Server Replication Services 3.0.2.0 through 3.0.5.0 allows an attacker with low‑privileged access to escalate their privileges to root. By exploiting this flaw, the attacker can execute root‑level commands, obtain a root shell, and change the root user’s password. Successf...
PT-2026-43704
Name of the Vulnerable Software and Affected Versions IBM Netezza Performance Server Replication Services versions 3.0.2.0 through 3.0.5.0 Description An attacker with low-privileged access can escalate their privileges to root. This allows the execution of root-level commands, obtaining a root...
EternalBlue
EternalBlue MS17-010 Exploitation Lab A professional, end-t...
Awakening the Hydra: Stabilizing Multi-Concept Backdoor Injection in Text-To-Image Diffusion Models
Text-to-image diffusion models are increasingly developed through open-source reuse and repeated downstream fine-tuning, where reused checkpoints are difficult to verify and thus more susceptible to hidden backdoor behaviors. In such ecosystems, a single pretrained model may be sequentially adapt...
Critical ‘Claw Chain’ Vulnerabilities Put Thousands of OpenClaw AI Servers at Risk
Critical Claw Chain vulnerabilities in OpenClaw expose thousands of AI servers to data theft, backdoors, and admin-level attacks globally this week...
Path Traversal
github.com/ctfer-io/monitoring is vulnerable to a Path Traversal. The vulnerability is due to a missing trailing path separator in the strings.HasPrefix check within the sanitizeArchivePath function, which allows an attacker to perform arbitrary file writes via a crafted archive, potentially...
Four OpenClaw Flaws Enable Data Theft, Privilege Escalation, and Persistence
Cybersecurity researchers have disclosed a set of four security flaws in OpenClaw that could be chained to achieve data theft, privilege escalation, and persistence. The vulnerabilities, collectively dubbed Claw Chain by Cyera, can permit an attacker to establish a foothold, expose sensitive data...
PT-2026-40781
007% · CVE-2026-22841 · v200.0.1 → 0.007 The Zero-Day Supply Chain: How Lyrie Caught 7 Backdoors Before Public Disclosure...
Backdoor Channels Hidden in Latent Space: Cryptographic Undetectability in Modern Neural Networks
Recent cryptographic results establish that neural networks can be backdoored such that no efficient algorithm can distinguish them from a clean model. These guarantees, however, have been confined to stylised architectures of limited practical relevance, leaving open whether comparable...
CVE-2026-27662
Affected devices do not properly restrict access to the web browser via the Control Panel when no corresponding security mechanisms are in place. This could allow an unauthenticated attacker to gain unauthorized access to the web browser, potentially enabling the discovery of backdoors, performin...
