CVE-2026-46517

LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, hardcoded "trustremotecode=True" enables HF supply-chain RCE without user opt-in. Version 0.13.0 patches the issue...

CVE-2026-46432

LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, LMDeploy is vulnerable to arbitrary code execution through hardcoded "trustremotecode=True" in multiple HuggingFace model-loading call sites. At time of publication, there are no...

CVE-2026-41732

JsonPulsarHeaderMapper matched type headers against trusted packages using a prefix check, meaning that trusting any package implicitly trusted all of its subpackages. Additionally, an empty trusted-packages configuration fell back to trusting all packages rather than applying a safe default...

VMware Spring AMQP 信任管理问题漏洞

VMware Spring AMQP is a message queue integration framework developed by VMware, Inc. There is a vulnerability related to trust management in VMware Spring AMQP. This vulnerability arises when configuring a proxy connection using RabbitConnectionFactoryBean.setUriamqps://…, without calling...

S2OPC OPC UA Toolkit 信任管理问题漏洞

S2OPC OPC UA Toolkit is an open-source development toolkit for OPC UA communication, developed by Systerel. The S2OPC OPC UA Toolkit has a trust management vulnerability, which stems from improper comparison with a list of trusted certificates. This vulnerability may cause well-formatted, untrust...

PT-2026-48383

A vulnerability has been found in some Dahua products. An attacker may obtain the device’s CA root certificate. If that CA is installed and trusted on client systems, the attacker could issue fraudulent certificates trusted by those clients and undermine the certificate trust chain...

Dahua IPC 安全漏洞

Dahua IPC is a series of industrial control computers produced by Dahua Corporation in China. There is a security vulnerability in Dahua IPC. This vulnerability stems from the possibility of obtaining the CA root certificate. If this CA is installed and trusted on the client system, an attacker c...

VMware Spring for Apache Pulsar 代码问题漏洞

VMware Spring for Apache Pulsar is a Pulsar messaging integration framework developed by the company VMware. Versions of VMware Spring for Apache Pulsar such as 2.0.0, 1.2.0, and 1.1.0 have code vulnerabilities. These vulnerabilities stem from the use of JsonPulsarHeaderMapper to check header typ...

lmdeploy 代码注入漏洞

lmdeploy is a toolkit developed by InternLM for compressing, deploying, and serving LLMs. Versions of lmdeploy prior to 0.12.3 have a code injection vulnerability, which stems from the hard-coded trustremotecode=True setting. This vulnerability could lead to remote code execution within the...

lmdeploy 代码注入漏洞

lmdeploy is a toolkit developed by InternLM for compressing, deploying, and serving LLMs. Versions of lmdeploy prior to 0.12.3 have a code injection vulnerability. This vulnerability stems from the hardcoding of trustremotecode=True at multiple HuggingFace model loading points, which may allow...

PT-2026-48474

Name of the Vulnerable Software and Affected Versions @hulumi/policies versions prior to 1.4.0 Description IAM-role policy checks can be bypassed when a role trusts multiple OIDC OpenID Connect providers. The issue occurs when the Principal.Federated field contains a JSON array of multiple...

CVE-2026-29114: Insertion of Sensitive Information into Externally-Accessible File or Directory

A vulnerability has been found in some Dahua products. An attacker may obtain the device’s CA root certificate. If that CA is installed and trusted on client systems, the attacker could issue fraudulent certificates trusted by those clients and undermine the certificate trust chain...

CVE-2026-41732 In Spring for Apache Pulsar, overly broad trusted-package matching in header mapper exposes JDK classes to deserialization

JsonPulsarHeaderMapper matched type headers against trusted packages using a prefix check, meaning that trusting any package implicitly trusted all of its subpackages. Additionally, an empty trusted-packages configuration fell back to trusting all packages rather than applying a safe default...

CVE-2026-41714 In Spring AMQP the RabbitConnectionFactoryBean.setUri("amqps://...") bypasses secure SSL setup, uses TrustEverythingTrustManager

Applications that configure their broker connection via RabbitConnectionFactoryBean.setUri"amqps://..." without also calling setUseSSLtrue get TLS encryption with no certificate validation and no hostname verification. Affected versions: Spring AMQP 4.0.0 through 4.0.3; 3.2.0 through 3.2.10; 3.1....

CVE-2026-41714

Spring AMQP 2.4.x/3.1.x/3.2.x/4.0.x (versions 2.4.0–2.4.17, 3.1.0–3.1.15, 3.2.0–3.2.10, 4.0.0–4.0.3) are affected by CVE-2026-41714. The issue occurs when a broker connection is configured via RabbitConnectionFactoryBean.setUri("amqps://...") without calling setUseSSL(true). This leads to TLS enc...

CVE-2026-41714 In Spring AMQP the RabbitConnectionFactoryBean.setUri("amqps://...") bypasses secure SSL setup, uses TrustEverythingTrustManager

Applications that configure their broker connection via RabbitConnectionFactoryBean.setUri"amqps://..." without also calling setUseSSLtrue get TLS encryption with no certificate validation and no hostname verification. Affected versions: Spring AMQP 4.0.0 through 4.0.3; 3.2.0 through 3.2.10; 3.1....

CVE-2026-46517 LMDeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-out

LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, hardcoded "trustremotecode=True" enables HF supply-chain RCE without user opt-in. Version 0.13.0 patches the issue...

CVE-2026-46517 LMDeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-out

LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, hardcoded "trustremotecode=True" enables HF supply-chain RCE without user opt-in. Version 0.13.0 patches the issue...

CVE-2026-46517

LMDeploy has a hardcoded trust_remote_code=True path in multiple code locations (e.g., get_model_arch and related calls) that is invoked for every model load. This creates an implicit unsafe remote-code load path when loading HuggingFace models from a repository, with no user opt-out or CLI flag ...

CVE-2026-46432 LMDeploy: Arbitrary code execution via hardcoded trust_remote_code=True in lmdeploy model initialization

LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, LMDeploy is vulnerable to arbitrary code execution through hardcoded "trustremotecode=True" in multiple HuggingFace model-loading call sites. At time of publication, there are no...