CVE-2026-34450 Claude SDK for Python: Insecure Default File Permissions in Local Filesystem Memory Tool

The Claude SDK for Python provides access to the Claude API from Python applications. From version 0.86.0 to before version 0.87.0, the local filesystem memory tool in the Anthropic Python SDK created memory files with mode 0o666, leaving them world-readable on systems with a standard umask and...

CVE-2026-34450

The Claude SDK for Python provides access to the Claude API from Python applications. From version 0.86.0 to before version 0.87.0, the local filesystem memory tool in the Anthropic Python SDK created memory files with mode 0o666, leaving them world-readable on systems with a standard umask and...

PT-2026-29378

The Claude SDK for Python provides access to the Claude API from Python applications. From version 0.86.0 to before version 0.87.0, the local filesystem memory tool in the Anthropic Python SDK created memory files with mode 0o666, leaving them world-readable on systems with a standard umask and...

Debt behind the AI Boom: A Large-Scale Empirical Study of AI-Generated Code in the Wild

AI coding assistants are now widely used in software development. Software developers increasingly integrate AI-generated code into their codebases to improve productivity. Prior studies have shown that AI-generated code may contain code quality issues under controlled settings. However, we still...

CVE-2026-4982

A user with permission "update world" in any Venueless world is able to exfiltrate chat messages from direct messages or channels in other worlds on the same server due to a bug in the reporting feature. The exploitability is limited by the fact that the attacker needs to know the internal channe...

Finding Memory Leaks in C/C++ Programs Via Neuro-Symbolic Augmented Static Analysis

Memory leaks remain prevalent in real-world C/C++ software. Static analyzers such as CodeQL provide scalable program analysis but frequently miss such bugs because they cannot recognize project-specific custom memory-management functions and lack path-sensitive control-flow modeling. We present...

How Microsoft Defender protects high-value assets in real-world attack scenarios

In this article 1. Using asset context to strengthen detection 2. How high-value asset protection works 3. Real-world high-value asset protection scenarios 4. Protecting your HVAs 5. Learn more High-value assets including domain controllers, web servers, and identity infrastructure are frequent...

CVE-2026-4982

A user with permission "update world" in any Venueless world is able to exfiltrate chat messages from direct messages or channels in other worlds on the same server due to a bug in the reporting feature. The exploitability is limited by the fact that the attacker needs to know the internal channe...

EUVD-2026-16593

A user with permission "update world" in any Venueless world is able to exfiltrate chat messages from direct messages or channels in other worlds on the same server due to a bug in the reporting feature. The exploitability is limited by the fact that the attacker needs to know the internal channe...

CVE-2026-4982 Unauthorized access to chat contents

A user with permission "update world" in any Venueless world is able to exfiltrate chat messages from direct messages or channels in other worlds on the same server due to a bug in the reporting feature. The exploitability is limited by the fact that the attacker needs to know the internal channe...

CVE-2026-4982

CVE-2026-4982 affects Venueless where a user with the privilege “update world” can exfiltrate chat messages from direct messages or channels in other worlds on the same server due to a bug in the reporting feature. The vulnerability arises from the reporting component allowing cross-world access ...

CVE-2026-4982 Unauthorized access to chat contents

A user with permission "update world" in any Venueless world is able to exfiltrate chat messages from direct messages or channels in other worlds on the same server due to a bug in the reporting feature. The exploitability is limited by the fact that the attacker needs to know the internal channe...

CVE-2026-4982

A user with permission "update world" in any Venueless world is able to exfiltrate chat messages from direct messages or channels in other worlds on the same server due to a bug in the reporting feature. The exploitability is limited by the fact that the attacker needs to know the internal channe...

venueless 安全漏洞

Venueless is an open-source online activity platform developed by Venueless. There are security vulnerabilities in Venueless; these vulnerabilities stem from defects in the reporting feature, which may allow users with update world permissions to disclose chat messages from other worlds on the sa...

PT-2026-28703

Name of the Vulnerable Software and Affected Versions Venueless affected versions not specified Description A user possessing the "update world" permission within any Venueless world can potentially extract chat messages from direct messages or channels in other worlds on the same server. This is...

AI Runtime Threat Detection: From Input to Real-World Impact

Understanding and detecting AI-driven behavior across model, workload, and cloud...

Capgo CLI: symlink-following local secret writes enable arbitrary file overwrite + world-readable credentials (0600 missing)

Summary The Capgo CLI writes sensitive local files .capgo API key file and build credentials JSON using unsafe file operations that follow symlinks and do not enforce safe permissions. This allows an attacker-controlled repository to cause arbitrary file overwrite on the developer’s machine when...

CVE-2026-24063 World-writable uninstall script executed as root in Arturia Software Center

When a plugin is installed using the Arturia Software Center MacOS, it also installs an uninstall.sh bash script in a root owned path. This script is written to disk with the file permissions 777, meaning it is writable by any user. When uninstalling a plugin via the Arturia Software Center the...

CVE-2026-24063

The CVE concerns Arturia Software Center on macOS. A plugin install creates an uninstall.sh script in a root-owned path with 777 permissions, writable by any user. During plugin uninstall, the Privileged Helper is instructed to execute this script. If an attacker manipulates the script, this can ...

CVE-2026-24063

When a plugin is installed using the Arturia Software Center MacOS, it also installs an uninstall.sh bash script in a root owned path. This script is written to disk with the file permissions 777, meaning it is writable by any user. When uninstalling a plugin via the Arturia Software Center the...