PT-2026-42943

Name of the Vulnerable Software and Affected Versions HuggingFace transformers versions prior to 5.3.0 Description A critical remote code execution issue exists where an attacker can craft a malicious config.json file. By setting the attn implementation internal field to a repository ID controlle...

Edimax BR-6675nD security vulnerabilities

The Edimax BR-6675nD is a dual-band broadband wireless router produced by Edimax Corporation. Version 1.12 of the Edimax BR-6675nD contains a security vulnerability. This vulnerability stems from the formsetPPPoE function in the POST Request Handler component, where improper handling of the...

Improper Integrity Verification

Amazon SageMaker Python SDK is vulnerable to improper integrity verification. The vulnerability is due to missing integrity verification in the Triton inference handler, which allows an authenticated attacker with S3 write access to replace model artifacts with a specially crafted pickle payload...

New API SQL injection vulnerability

The New API is an interface software developed by QuantumNous. Versions of the New API prior to 0.12.1 have a SQL injection vulnerability. This vulnerability originates from the SearchUserTopUps/SearchAllTopUps function in the model/topup.go file of the self Endpoint component, which may lead to...

Malicious code in model-switch-router (npm)

Ten packages published by npm user asdxzxc at version 1.0.10 target developers working on AI and LLM tooling. Each package masquerades as a developer utility while executing a two-stage payload triggered via postinstall: package.json → lib/setup.js → lib/worker.js. Credential harvesting:...

MAL-2026-4279 Malicious code in model-switch-router (npm)

Ten packages published by npm user asdxzxc at version 1.0.10 target developers working on AI and LLM tooling. Each package masquerades as a developer utility while executing a two-stage payload triggered via postinstall: package.json → lib/setup.js → lib/worker.js. Credential harvesting:...

CVE-2026-9305: Improper Neutralization of Special Elements used in an SQL Command

A weakness has been identified in QuantumNous new-api up to 0.12.1. The impacted element is the function SearchUserTopUps/SearchAllTopUps of the file model/topup.go of the component self Endpoint. This manipulation causes sql injection. The attack can be initiated remotely. The exploit has been...

Reasoning As an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs

Large Reasoning Models LRMs have demonstrated remarkable capabilities in reasoning and generation tasks and are increasingly deployed in real-world applications. However, their explicit chain-of-thought CoT mechanism introduces new security risks, making them particularly vulnerable to jailbreak...

CVE-2026-8962

A flaw was found in Firefox. The Mozilla Foundation's Security Advisory describes the following issue: Mitigation bypass in the DOM: Security component...

CVE-2026-5843

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

CVE-2026-5843 Docker Model Runner container-to-host code execution via MLX-LM model_file importlib loading

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

CVE-2026-5843

The CVE describes a vulnerability in Docker Model Runner (macOS) where the MLX-LM backend unconditionally imports and executes arbitrary Python files specified by model_file in a model's config.json via importlib, without a trust_remote_code gate or sandboxing. This enables container-to-host arbi...

CVE-2026-5843 Docker Model Runner container-to-host code execution via MLX-LM model_file importlib loading

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

EUVD-2026-31491

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

CVE-2026-5843

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

CVE-2026-5817 Docker Model Runner container-to-host code execution via unsandboxed trust_remote_code in Python inference backends

The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trustremotecode=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.frompretrained to import and execute arbitrary Python files included in any model pulled fr...

CVE-2026-5817

The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trustremotecode=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.frompretrained to import and execute arbitrary Python files included in any model pulled fr...

EUVD-2026-31493

The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trustremotecode=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.frompretrained to import and execute arbitrary Python files included in any model pulled fr...

CVE-2026-5817 Docker Model Runner container-to-host code execution via unsandboxed trust_remote_code in Python inference backends

The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trustremotecode=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.frompretrained to import and execute arbitrary Python files included in any model pulled fr...

CVE-2026-5817

CVE-2026-5817 affects the vllm-metal backend used by Docker Model Runner on macOS. The backend loads model tokenizers with trust_remote_code=True, causing transformers.AutoTokenizer.from_pretrained() to import and execute arbitrary Python files from models pulled from an OCI registry. This can en...