CVE-2025-63681

Open-WebUI CVE-2025-63681 affects v0.6.33. The API endpoint /api/tasks/stop/ allows direct cancellation of tasks without verifying ownership, enabling a normal user to stop arbitrary LLM response tasks (Incorrect Access Control). Base score 4.3 (Medium); attack vector NETWORK, privileges required...

CVE-2025-63317

Todoist version v8896 is vulnerable to a Cross-Site Scripting (XSS) flaw within the /api/v1/uploads endpoint. The root cause is a lack of sanitization for uploaded SVG files , which allows for the execution of embedded JavaScript when a user opens an attachment from a task or comment. The vulnera...

COGNITION: From Evaluation to Defense against Multimodal LLM CAPTCHA Solvers

This paper studies how multimodal large language models MLLMs undermine the security guarantees of visual CAPTCHA. We identify the attack surface where an adversary can cheaply automate CAPTCHA solving using off-the-shelf models. We evaluate 7 leading commercial and open-source MLLMs across 18...

Red Teaming Large Reasoning Models

Large Reasoning Models LRMs have emerged as a powerful advancement in multi-step reasoning tasks, offering enhanced transparency and logical consistency through explicit chains of thought CoT. However, these models introduce novel safety and reliability risks, such as CoT-hijacking and...

MAL-2025-191205 Malicious code in @clausehq/flows-step-taskscreateurl (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector aea06d73112475520b51b21f45ccc22e8ddcf55245e7045f7efcdeefda036595 The package @clausehq/flows-step-taskscreateurl was found to contain malicious code. Source: ghsa-malware...

LLMs As Firmware Experts: A Runtime-Grown Tree-Of-Agents Framework

Large Language Models LLMs and their agent systems have recently demonstrated strong potential in automating code reasoning and vulnerability detection. However, when applied to large-scale firmware, their performance degrades due to the binary nature of firmware, complex dependency structures, a...

Beyond Jailbreak: Unveiling Risks in LLM Applications Arising from Blurred Capability Boundaries

LLM applications i.e., LLM apps leverage the powerful capabilities of LLMs to provide users with customized services, revolutionizing traditional application development. While the increasing prevalence of LLM-powered applications provides users with unprecedented convenience, it also brings fort...

CVE-2025-64708: Insufficient Session Expiration

authentik is an open-source Identity Provider. Prior to versions 2025.8.5 and 2025.10.2, in previous authentik versions, invitations were considered valid regardless if they are expired or not, thus relying on background tasks to clean up expired ones. In a normal scenario this can take up to 5...

TaskHound - Windows Privileged Scheduled Task Discovery Tool

TaskHound is a tool that hunts for Windows scheduled tasks that run with privileged accounts and stored credentials. It enumerates tasks over SMB, parses XMLs, and identifies high-value attack opportunities through BloodHound support...

The Role of Humans in an AI-Powered World

As AI capabilities grow, we must delineate the roles that should remain exclusively human. The line seems to be between fact-based decisions and judgment-based decisions. For example, in a medical context, if an AI was demonstrably better at reading a test result and diagnosing cancer than a huma...

Devolutions Server <= 2025.2.15.0 Improper Input Validation (DEVO-2025-0015) (CVE-2025-11958)

The version of Devolutions Server installed on the remote host is prior or equal to 2025.2.15.0 and is, therefore, affected by an improper authorization vulnerability: - An improper input validation in the Security Dashboard ignored-tasks API of Devolutions Server 2025.2.15.0 and earlier allows a...

CVE-2025-60710

Improper link resolution before file access 'link following' in Host Process for Windows Tasks allows an authorized attacker to elevate privileges locally...

CVE-2025-60710

Improper link resolution before file access 'link following' in Host Process for Windows Tasks allows an authorized attacker to elevate privileges locally...

CVE-2025-60710

Improper link resolution before file access 'link following' in Host Process for Windows Tasks allows an authorized attacker to elevate privileges locally...

CVE-2025-60710 Host Process for Windows Tasks Elevation of Privilege Vulnerability

...

EUVD-2025-93436

Improper link resolution before file access 'link following' in Host Process for Windows Tasks allows an authorized attacker to elevate privileges locally...

CVE-2025-60710 Host Process for Windows Tasks Elevation of Privilege Vulnerability

...

CVE-2025-60710

CVE-2025-60710 is a Windows privilege-escalation vulnerability in the Host Process for Windows Tasks caused by improper link resolution before file access (link following). The CVE entry notes local abuse by an authorized attacker and a high impact (C:H, I:H, A:H) with local attack vector and no ...

Host Process for Windows Tasks Elevation of Privilege Vulnerability

Improper link resolution before file access 'link following' in Host Process for Windows Tasks allows an authorized attacker to elevate privileges locally...

kernel: bpf: Fix UAF via mismatching bpf_prog/attachment RCU flavors

In the Linux kernel, the following vulnerability has been resolved: bpf: Fix UAF via mismatching bpfprog/attachment RCU flavors Uprobes always use bpfprogrunarrayuprobe under tasks-trace-RCU protection. But it is possible to attach a non-sleepable BPF program to a uprobe, and non-sleepable BPF...