PurpleLlama

🤗 Models on Hugging Face | Blog | Website | CyberSec Eval Paper | Llama Guard Paper Purple Llama Purple Llama is an umbrella project that over time will bring together tools and evals to help the community build responsibly with open generative AI models. The initial release will include tools an...

pcc

PHP Secure Configuration Checker Check current PHP configuration for potential security flaws. Simply access this file from your webserver or run on CLI. Author This software was written by Ben Fuhrmannek, SektionEins GmbH, in an effort to automate php.ini checks and spend more time on cheerful...

CVE-2026-76341 Risky Commands Safeguards Bypass through Table Editor Dataset Initial Data in Splunk Enterprise

In Splunk Enterprise versions below 10.4.2, 10.2.6, 10.0.9, and 9.4.14, a user who holds the "power" Splunk role could store attacker-controlled Search Processing Language SPL in a Table Editor dataset and share the dataset. A user who holds the "admin" Splunk role triggers the SPL when that user...

CVE-2026-76323 SPL Risky Command Safeguards Bypass through the Job Details Dashboard in Splunk Enterprise

In Splunk Enterprise versions below 10.4.2, 10.2.6, 10.0.9, and 9.4.14, a user who does not hold the "admin" or "power" Splunk roles could bypass Search Processing Language SPL safeguards for risky commands through the Job Details dashboard. The injected SPL could run using the permissions of an...

CVE-2026-76323

CVE-2026-76323 is an SPL risky command safeguards bypass in the Job Details dashboard of Splunk Enterprise versions below 10.4.2, 10.2.6, 10.0.9, and 9.4.14 . The root cause is that the dashboard does not correctly neutralize a caller-supplied search identifier before embedding it into SPL search...

CVE-2026-76254

Splunk Enterprise versions below 10.4.2, 10.2.6, 10.0.9, 9.4.14, and 9.3.14 are affected. Dataset Explorer fails to validate or escape dataset names when building SPL searches and skips SPL safeguards for risky commands, enabling a phishing attack to execute arbitrary SPL under the victim's privi...

CVE-2026-76254 SPL Command Safeguards Bypass through Splunk Web in Splunk Enterprise

In Splunk Enterprise versions below 10.4.2, 10.2.6, 10.0.9, 9.4.14, and 9.3.14, an unauthenticated user could cause another user to dispatch arbitrary Search Processing Language SPL pipelines from Dataset Explorer with the same privileges as that user, which can allow for access to all relevant...

CVE-2026-76255 Risky Command Safeguards Bypass through Splunk Web in Splunk Enterprise

In Splunk Enterprise versions below 10.4.1, 10.2.6, 10.0.8, and 9.4.13, a user who does not hold the "admin" or "power" Splunk roles could trick another user into running arbitrary Search Processing Language SPL commands through the Data Model Editor using the permissions of the affected user. Th...

XRP Volatility Surges as Cybersecurity Threats and Market Changes Raise New Concerns

XRP volatility drives faster crypto trading as AI tools gain traction, while phishing, exchange attacks and automation risks test digital asset safeguards...

CVE-2026-12251

CVE-2026-12251 affects the Ultimate Member WordPress plugin prior to 2.12.1. The vulnerability arises because administrator-level capabilities are not filtered from the roles shown on registration forms, and the post-registration safeguard against elevated accounts is disabled by default. Consequ...

EUVD-2026-51430

The Ultimate Member WordPress plugin before 2.12.1 does not filter administrator-level capabilities from the roles it makes selectable on its registration forms, and its post-registration safeguard against elevated accounts is disabled by default, allowing unauthenticated users to register with a...

EUVD-2026-50555

Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. In versions 1.88.0 up to but not including 1.107.1 and 2.0.0b1 up to but not including 2.5.0, the UI adapters AG-UI via Agent.toagui/AGUIAdapter, and Vercel AI via VercelAIAdapter use...

⚡ Weekly Recap: Rogue AI Agents, Check Point Exploit, Slopsquatting, ClickFix Lures and More

Monday starts with the usual promise that everything is under control. Then the logs wake up. This week, trusted tools crossed lines, old flaws found new work, exposed systems stayed exposed, and attackers kept hiding inside normal-looking services. Nothing looked strange at first. That helped...

CVE-2026-14890

SGLang uses an expert-parallel backup subsystem that exposes a ZeroMQ PULL socket on a routable network interface that does not contain authentication or deserialization safeguards, allowing an attacker to provide a malicious pickle file that results in unauthenticated remote code execution when...

CVE-2026-14890

SGLang is vulnerable to unauthenticated remote code execution via its expert-parallel backup subsystem. The issue centers on a ZeroMQ PULL socket exposed on a routable network interface with no authentication or deserialization safeguards, allowing an attacker to send a malicious pickle payload t...

PYSEC-2026-2546 Keras has a Local File Disclosure via HDF5 External Storage During Keras Weight Loading

Summary TensorFlow / Keras continues to honor HDF5 “external storage” and ExternalLink features when loading weights. A malicious .weights.h5 or a .keras archive embedding such weights can direct loadweights to read from an arbitrary readable filesystem path. The bytes pulled from that path...

CVE-2026-58203

A flaw was found in the pydantic-settings component. When configured to handle nested secrets, this vulnerability allows an attacker to read sensitive files from outside the designated secrets directory. By placing a specially crafted symbolic link within the secrets directory, an attacker can...

Security update for transmission (moderate)

openSUSE Security Update: Security update for transmission Announcement ID: openSUSE-SU-2026:0237-1 Rating: moderate References: 1267404 Cross-References: CVE-2026-38978 Affected Products: openSUSE Backports SLE-15-SP7 An update that fixes one vulnerability is now available. Description: This...

CVE-2026-10706

In Adalo’s no-code app builder, Versions 1 and 2 the attackers may extract full user records and correlate user behavior across multiple applications via dbId enumeration. The platform does not implement data minimization, privacy by design, or implement appropriate technical safeguards, allowing...

EUVD-2026-42277

In Adalo’s no-code app builder, Versions 1 and 2 the attackers may extract full user records and correlate user behavior across multiple applications via dbId enumeration. The platform does not implement data minimization, privacy by design, or implement appropriate technical safeguards, allowing...