CVE-2026-39981

AGiXT is a dynamic AI Agent Automation Platform. Prior to 1.9.2, the safejoin function in the essentialabilities extension fails to validate that resolved file paths remain within the designated agent workspace. An authenticated attacker can use directory traversal sequences to read, write, or...

Trezor多款产品 安全漏洞

Trezor One, among others, is a product of the Czech Republic-based Trezor company. Trezor One is a digital currency wallet device. Trezor T is a hardware cryptocurrency wallet device. Trezor Safe is also a hardware cryptocurrency wallet device. Several Trezor products have security vulnerabilitie...

HP System Optimizer - Escalation of Privilege

HP System Optimizer might potentially be vulnerable to escalation of privilege. HP is releasing an update to mitigate this potential vulnerability. HP has identified affected versions and the minimum version 1101.2603 that mitigates the potential vulnerability...

PT-2026-32627

A side-channel vulnerability exists in the implementation of BIP-39 mnemonic processing, as observed in Trezor One v1.13.0 to v1.14.0, Trezor T v1.13.0 to v1.14.0, and Trezor Safe v1.13.0 to v1.14.0 hardware wallets. This originates from the BIP-39 standard guidelines, which induce non-constant...

Quantum-Safe IPsec in the Banking Industry

The emergence of Cryptographically Relevant Quantum Computers CRQCs presents a critical threat to classical cryptographic systems, particularly widely adopted protocols such as RSA, Diffie-Hellman DH, and Elliptic Curve Cryptography ECC. Given their extensive use in the financial sector, the adve...

PT-2026-32697

Name of the Vulnerable Software and Affected Versions October versions prior to 3.7.13 October versions 4.0.0 through 4.1.4 Description A sandbox bypass exists in the optional Twig safe mode feature CMS SAFE MODE. Certain methods on the collect helper were not properly restricted, allowing...

PT-2026-32977

Name of the Vulnerable Software and Affected Versions @adonisjs/http-server versions prior to 7.8.1 @adonisjs/http-server versions 8.0.0-next.0 through 8.1.3 @adonisjs/core versions prior to 7.4.0 Description The response.redirect.back method reads the Referer header from the incoming HTTP reques...

PT-2026-32911

A server-side information disclosure vulnerability was identified in the INI settings parser. PHP's parse ini string function supports $ syntax for environment variable interpolation. Attackers with Editor access could inject $APP KEY, $DB PASSWORD, or similar patterns into CMS page settings...

CVE-2026-6224

A security flaw has been discovered in nocobase plugin-workflow-javascript up to 2.0.23. This issue affects the function createSafeConsole of the file packages/plugins/@nocobase/plugin-workflow-javascript/src/server/Vm.js. Performing a manipulation results in sandbox issue. The attack can be...

CVE-2026-6224 nocobase plugin-workflow-javascript Vm.js createSafeConsole sandbox

A security flaw has been discovered in nocobase plugin-workflow-javascript up to 2.0.23. This issue affects the function createSafeConsole of the file packages/plugins/@nocobase/plugin-workflow-javascript/src/server/Vm.js. Performing a manipulation results in sandbox issue. The attack can be...

CVE-2026-1462

A flaw was found in the keras package. This vulnerability allows an attacker to execute unauthorized code on a victim's system. It occurs when a victim loads a specially crafted .keras model, even if the safemode security feature is active. The issue arises because the keras package can...

EUVD-2026-21970

A vulnerability in the TFSMLayer class of the keras package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of .keras models, even when safemode=True. This bypasses the security guarantees of safemode and enables arbitrary attacker-controlled...

GHSA-4F3F-G24H-FR8M Keras has an untrusted deserialization vulnerability

A vulnerability in the TFSMLayer class of the keras package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of .keras models, even when safemode=True. This bypasses the security guarantees of safemode and enables arbitrary attacker-controlled...

DEBIAN-CVE-2026-1462

A vulnerability in the TFSMLayer class of the keras package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of .keras models, even when safemode=True. This bypasses the security guarantees of safemode and enables arbitrary attacker-controlled...

CVE-2026-1462

A vulnerability in the TFSMLayer class of the keras package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of .keras models, even when safemode=True. This bypasses the security guarantees of safemode and enables arbitrary attacker-controlled...

UBUNTU-CVE-2026-1462

A vulnerability in the TFSMLayer class of the keras package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of .keras models, even when safemode=True. This bypasses the security guarantees of safemode and enables arbitrary attacker-controlled...

CVE-2026-1462

A vulnerability in the TFSMLayer class of the keras package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of .keras models, even when safemode=True. This bypasses the security guarantees of safemode and enables arbitrary attacker-controlled...

CVE-2026-1462 Safe Mode Bypass in keras-team/keras

A vulnerability in the TFSMLayer class of the keras package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of .keras models, even when safemode=True. This bypasses the security guarantees of safemode and enables arbitrary attacker-controlled...

CVE-2026-1462

A vulnerability in the TFSMLayer class of the keras package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of .keras models, even when safemode=True. This bypasses the security guarantees of safemode and enables arbitrary attacker-controlled...

CVE-2026-1462

The CVE-2026-1462 issue affects the keras package (v3.13.0) via the TFSMLayer deserialization path. The vulnerability allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of .keras models, even when safe_mode is enabled, due to unconditional loading of external Sa...