Arbitrary Code Injection

Overview Affected versions of this package are vulnerable to Arbitrary Code Injection through the escapeNodeAttributeValues process. An attacker can execute arbitrary operating system commands by crafting a malicious .sy.zip file containing specially formatted block attribute values, which, when...

TorchGeo Remote Code Execution Vulnerability

Impact TorchGeo 0.4–0.6.0 used an eval statement in its model weight API that could allow an unauthenticated, remote attacker to execute arbitrary commands. All platforms that expose torchgeo.models.getweight or torchgeo.trainers as an external API could be affected. Patches The eval statement wa...

Quantum-Safe Code Auditing: LLM-Assisted Static Analysis and Quantum-Aware Risk Scoring for Post-Quantum Cryptography Migration

The impending arrival of cryptographically relevant quantum computers CRQCs threatens the security foundations of modern software: Shor's algorithm breaks RSA, ECDSA, ECDH, and Diffie-Hellman, while Grover's algorithm reduces the effective security of symmetric and hash-based schemes. Despite NIS...

PT-2026-29577

Name of the Vulnerable Software and Affected Versions ONNX versions prior to 1.21.0 Description The ExternalDataInfo class in ONNX used Python’s setattr function to load metadata from ONNX model files without validating the keys. This allowed an attacker to craft a malicious model that could...

Automated Framework to Evaluate and Harden LLM System Instructions against Encoding Attacks

System Instructions in Large Language Models LLMs are commonly used to enforce safety policies, define agent behavior, and protect sensitive operational context in agentic AI applications. These instructions may contain sensitive information such as API credentials, internal policies, and...

When Safe Models Merge into Danger: Exploiting Latent Vulnerabilities in LLM Fusion

Model merging has emerged as a powerful technique for combining specialized capabilities from multiple fine-tuned LLMs without additional training costs. However, the security implications of this widely-adopted practice remain critically underexplored. In this work, we reveal that model merging...

llama.cpp 缓冲区错误漏洞

Llama.cpp is a multimodal model developed by Georgi Gerganov. Prior versions of llama.cpp b8492 contained a buffer error vulnerability. This vulnerability stemmed from the deserializetensor function in the RPC backend, which skipped all boundary verifications when the buffer field of the tensor w...

Open Neural Network Exchange 安全漏洞

Open Neural Network Exchange is an open-source ecosystem developed by Open Neural Network Exchange. It enables AI developers to choose the appropriate tools as the project evolves. Versions of Open Neural Network Exchange prior to 1.21.0 contained a security vulnerability. This vulnerability...

PT-2026-29672

Name of the Vulnerable Software and Affected Versions Go MCP SDK versions prior to 1.4.0 Description The Go MCP SDK, utilizing Go's standard encoding/json, did not enable DNS rebinding protection by default for HTTP-based servers prior to version 1.4.0. When an HTTP-based MCP server was run on...

Docker Model Runner 代码问题漏洞

Docker Model Runner is an open-source Docker model runner developed by Docker. Versions of Docker Model Runner prior to 1.1.25 contained code vulnerabilities. These vulnerabilities stemmed from a server-side request forgeing attack during the OCI registry token exchange process. When pulling...

PT-2026-29500

Improper Neutralization of Input During Web Page Generation 'Cross-site Scripting' vulnerability in Brainstorm Force Ultimate Addons for WPBakery Page Builder allows DOM-Based XSS.This issue affects Ultimate Addons for WPBakery Page Builder: from n/a before 3.21.4...

PT-2026-29827

Name of the Vulnerable Software and Affected Versions KubeAI versions prior to 0.23.2 Description The ollamaStartupProbeScript function in KubeAI constructs a shell command string using fmt.Sprintf with unsanitized model URL components ref, modelParam. This command is executed via bash -c as a...

CVE-2026-5153

A flaw has been found in Tenda CH22 1.0.0.1. The affected element is the function FormWriteFacMac of the file /goform/WriteFacMac. Executing a manipulation of the argument mac can lead to command injection. The attack may be launched remotely. The exploit has been published and may be used...

CVE-2026-33032

Nginx UI is a web user interface for the Nginx web server. In versions 2.3.5 and prior, the nginx-ui MCP Model Context Protocol integration exposes two HTTP endpoints: /mcp and /mcpmessage. While /mcp requires both IP whitelisting and authentication AuthRequired middleware, the /mcpmessage endpoi...

GHSA-3R9X-F23J-GC73 onnx Vulnerable to Path Traversal via Symlink

Summary A path traversal vulnerability via symlink allows to read arbitrary files outside model or user-provided directory. Details The following check for symlink is ineffective and it is possible to point a symlink to an arbitrary location on the file system:...

CVE-2025-15379

A flaw was found in MLflow. When deploying a model with envmanager=LOCAL, MLflow's model serving container initialization code, specifically the installmodeldependenciestoenv function, reads dependency specifications from the model artifact's pythonenv.yaml file. An attacker can supply a maliciou...

CVE-2026-34237

MCP Java SDK is the official Java SDK for Model Context Protocol servers and clients. Prior to versions 0.83.0, 1.0.1, and 1.1.1, there is a hardcoded wildcard CORS vulnerability. This issue has been patched in versions 0.83.0, 1.0.1, and 1.1.1...

CVE-2026-34237

CVE-2026-34237 affects MCP Java SDK. A hardcoded wildcard CORS configuration (Access-Control-Allow-Origin: *) existed in versions before 0.83.0, 1.0.1, and 1.1.1, allowing cross-origin requests to server endpoints (including SSE paths). The issue has been patched in those versions (0.83.0, 1.0.1,...

CVE-2026-34237 MCP Java SDK has a Hardcoded Wildcard CORS (Access-Control-Allow-Origin: *)

MCP Java SDK is the official Java SDK for Model Context Protocol servers and clients. Prior to versions 0.83.0, 1.0.1, and 1.1.1, there is a hardcoded wildcard CORS vulnerability. This issue has been patched in versions 0.83.0, 1.0.1, and 1.1.1...

CVE-2026-34237 MCP Java SDK has a Hardcoded Wildcard CORS (Access-Control-Allow-Origin: *)

MCP Java SDK is the official Java SDK for Model Context Protocol servers and clients. Prior to versions 0.83.0, 1.0.1, and 1.1.1, there is a hardcoded wildcard CORS vulnerability. This issue has been patched in versions 0.83.0, 1.0.1, and 1.1.1...