LLM Causality Analysis Framework

A comprehensive framework for multi-level causality analysis in Large Language Models LLMs, enabling systematic investigation of safety mechanisms and misbehavior detection across token, neuron, layer, and representation levels. Includes the whitepaper 2512.04841.pdf titled SoK: A Comprehensive...

The Road of Adaptive AI for Precision in Cybersecurity

Cybersecurity's evolving complexity presents unique challenges and opportunities for AI research and practice. This paper shares key lessons and insights from designing, building, and operating production-grade GenAI pipelines in cybersecurity, with a focus on the continual adaptation required to...

MINI-54HV-MRVM-W2QV

Bulletin has no description...

CVE-2025-66563

Monkeytype is a minimalistic and customizable typing test. In 25.49.0 and earlier, there is improper handling of user input which allows an attacker to execute malicious javascript on anyone viewing a malicious quote submission. quote.text and quote.source are user input, and they're inserted...

expat: Integer overflow in build_model in xmlparse.c

expat libexpat is susceptible to a software flaw that causes process interruption. When processing a large number of prefixed XML attributes on a single tag can libexpat can terminate unexpectedly due to integer overflow. The highest threat from this vulnerability is to availability,...

CVE-2025-66448

A remote code execution vulnerability has been identified in vLLM. An attacker can exploit a weakness in the model loading process to silently fetch and run unauthorized, malicious Python code on the host system. This happens because the engine mistakenly executes code from a remote repository...

MINI-JP88-J82C-2QQW

Bulletin has no description...

CVE-2025-66222

DeepChat is a smart assistant uses artificial intelligence. In 0.5.0 and earlier, there is a Stored Cross-Site Scripting XSS vulnerability in the Mermaid diagram renderer allows an attacker to execute arbitrary JavaScript within the application context. By leveraging the exposed Electron IPC...

CVE-2025-29269

Summary: CVE-2025-29269 affects ALLNET ALL-RUT22GW v3.3.8 and is an OS command injection via the parameter named “command” in the popen.cgi endpoint. The vulnerability’s root cause is improper handling of the command parameter, enabling arbitrary command execution. Several sources corroborate the...

CVE-2025-66222

DeepChat is a smart assistant uses artificial intelligence. In 0.5.0 and earlier, there is a Stored Cross-Site Scripting XSS vulnerability in the Mermaid diagram renderer allows an attacker to execute arbitrary JavaScript within the application context. By leveraging the exposed Electron IPC...

DRUPAL-CONTRIB-2025-119

This modules provides the ability to chat with an AI Agent using a large-language model LLM provider for different purposes. The module doesn’t sufficiently filter LLM responses. This leads to a cross-site scripting XSS vulnerability where an attacker can use prompt injections on user-generated...

CVE-2025-66222 DeepChat Cross-Site Scripting(XSS) escalate to Remote Code Execution(RCE)

DeepChat is a smart assistant uses artificial intelligence. In 0.5.0 and earlier, there is a Stored Cross-Site Scripting XSS vulnerability in the Mermaid diagram renderer allows an attacker to execute arbitrary JavaScript within the application context. By leveraging the exposed Electron IPC...

CVE-2025-66222 DeepChat Cross-Site Scripting(XSS) escalate to Remote Code Execution(RCE)

DeepChat is a smart assistant uses artificial intelligence. In 0.5.0 and earlier, there is a Stored Cross-Site Scripting XSS vulnerability in the Mermaid diagram renderer allows an attacker to execute arbitrary JavaScript within the application context. By leveraging the exposed Electron IPC...

CVE-2025-66222

DeepChat (0.5.0 and earlier) is affected by a Stored XSS in the Mermaid diagram renderer, exploitable via the Electron IPC bridge to escalate to RCE by starting a malicious MCP server. Affected product/version: DeepChat prior to 0.5.0. Root cause: XSS within Mermaid rendering allows arbitrary Jav...

EUVD-2025-201091

DeepChat is a smart assistant uses artificial intelligence. In 0.5.0 and earlier, there is a Stored Cross-Site Scripting XSS vulnerability in the Mermaid diagram renderer allows an attacker to execute arbitrary JavaScript within the application context. By leveraging the exposed Electron IPC...

CVE-2025-66222 DeepChat Cross-Site Scripting(XSS) escalate to Remote Code Execution(RCE)

DeepChat is a smart assistant uses artificial intelligence. In 0.5.0 and earlier, there is a Stored Cross-Site Scripting XSS vulnerability in the Mermaid diagram renderer allows an attacker to execute arbitrary JavaScript within the application context. By leveraging the exposed Electron IPC...

CVE-2025-20381

In Splunk MCP Server app versions below 0.2.4, a user with access to the "runsplunkquery" Model Context Protocol MCP tool could bypass the SPL command allowlist controls in MCP by embedding SPL commands as sub-searches, leading to unauthorized actions beyond the intended MCP restrictions...

CVE-2025-20381 SPL commands allowlist controls bypass in Splunk MCP Server app through "run_splunk_query" MCP tool

In Splunk MCP Server app versions below 0.2.4, a user with access to the "runsplunkquery" Model Context Protocol MCP tool could bypass the SPL command allowlist controls in MCP by embedding SPL commands as sub-searches, leading to unauthorized actions beyond the intended MCP restrictions...

CVE-2025-20381

In Splunk MCP Server App

CVE-2025-20381 SPL commands allowlist controls bypass in Splunk MCP Server app through "run_splunk_query" MCP tool

In Splunk MCP Server app versions below 0.2.4, a user with access to the "runsplunkquery" Model Context Protocol MCP tool could bypass the SPL command allowlist controls in MCP by embedding SPL commands as sub-searches, leading to unauthorized actions beyond the intended MCP restrictions...