Trust in LLM-Controlled Robotics: A Survey of Security Threats, Defenses and Challenges

The integration of Large Language Models LLMs into robotics has revolutionized their ability to interpret complex human commands and execute sophisticated tasks. However, such paradigm shift introduces critical security vulnerabilities stemming from the ''embodiment gap'', a discord between the...

Weak Enforcement and Low Compliance in PCI~DSS: A Comparative Security Study

Although credit and debit card data continue to be a prime target for attackers, organizational adherence to the Payment Card Industry Data Security Standard PCI DSS remains surprisingly low. Despite prior work showing that PCI DSS can reduce card fraud, only 32.4% of organizations were fully...

CeLLMate: Sandboxing Browser AI Agents

Browser-using agents BUAs are an emerging class of autonomous agents that interact with web browsers in human-like ways, including clicking, scrolling, filling forms, and navigating across pages. While these agents help automate repetitive online tasks, they are vulnerable to prompt injection...

CVE-2025-66452

LibreChat is a ChatGPT clone with additional features. In versions 0.8.0 and below, there is no handler for JSON parsing errors; SyntaxError from express.json includes user input in the error message, which gets reflected in responses. User input including HTML/JavaScript can be exposed in error...

CVE-2025-66452

LibreChat (versions ≤ 0.8.0) is affected by a lack of handling for JSON parsing errors in express.json(). A SyntaxError triggered by user input can be reflected in error responses, exposing input (including HTML/JavaScript) and creating an XSS risk if Content-Type isn’t strictly enforced. The iss...

CVE-2025-66452 LibreChat's lack of JSON parsing error handling can lead to XSS

LibreChat is a ChatGPT clone with additional features. In versions 0.8.0 and below, there is no handler for JSON parsing errors; SyntaxError from express.json includes user input in the error message, which gets reflected in responses. User input including HTML/JavaScript can be exposed in error...

EUVD-2025-202928

LibreChat is a ChatGPT clone with additional features. In versions 0.8.0 and below, there is no handler for JSON parsing errors; SyntaxError from express.json includes user input in the error message, which gets reflected in responses. User input including HTML/JavaScript can be exposed in error...

CVE-2025-66452 LibreChat's lack of JSON parsing error handling can lead to XSS

LibreChat is a ChatGPT clone with additional features. In versions 0.8.0 and below, there is no handler for JSON parsing errors; SyntaxError from express.json includes user input in the error message, which gets reflected in responses. User input including HTML/JavaScript can be exposed in error...

Remote Code Execution (RCE)

Apache Syncope is vulnerable to Remote Code Execution RCE. The vulnerability is due to unsafe handling of custom Groovy implementations, where a malicious administrator can inject Groovy code that is executed by the Syncope Core at runtime, enabling remote code execution until sandboxing is...

CVE-2025-36102

IBM Controller 11.1.0 through 11.1.1 and IBM Cognos Controller 11.0.0 through 11.0.1 FP6 could allow a privileged user to bypass validation, passing user input into the application as trusted data, due to client-side enforcement of server-side security...

Police Dismantle EUR 700 Million Crypto Scam That Used Deepfakes

Europol and Eurojust led a massive international police operation that successfully dismantled a crypto fraud network that laundered over €700M using deepfake ads...

Neuron MySQLWriteTool allows arbitrary/destructive SQL when exposed to untrusted prompts (agent “footgun”)

Impact MySQLWriteTool executes arbitrary SQL provided by the caller using PDO::prepare + execute without semantic restrictions. This is consistent with the name “write tool”, but in an LLM/agent context it becomes a high-risk capability: prompt injection or indirect prompt manipulation can cause...

CVE-2025-42616

Some endpoints in vulnerability-lookup that modified application state e.g. changing database entries, user data, configurations, or other privileged actions may have been accessible via HTTP GET requests without requiring a CSRF token. This flaw leaves the application vulnerable to Cross-Site...

SUSE CVE-2022-50621

In the Linux kernel, the following vulnerability has been resolved: dm: verity-loadpin: Only trust verity targets with enforcement Verity targets can be configured to ignore corrupted data blocks. LoadPin must only trust verity targets that are configured to perform some kind of enforcement when...

EUVD-2025-201817

IBM Controller 11.1.0 through 11.1.1 and IBM Cognos Controller 11.0.0 through 11.0.1 FP6 could allow a privileged user to bypass validation, passing user input into the application as trusted data, due to client-side enforcement of server-side security...

PT-2025-49837

A vulnerability has been identified in SINEMA Remote Connect Server All versions V3.2 SP4. Affected applications do not properly validate license restrictions against the database, allowing direct modification of the system ticketinfo table to bypass license limitations without proper enforcement...

CVE-2025-36102

IBM Controller 11.1.0 through 11.1.1 and IBM Cognos Controller 11.0.0 through 11.0.1 FP6 could allow a privileged user to bypass validation, passing user input into the application as trusted data, due to client-side enforcement of server-side security...

CVE-2025-36102 IBM Controller Validation Bypass

IBM Controller 11.1.0 through 11.1.1 and IBM Cognos Controller 11.0.0 through 11.0.1 FP6 could allow a privileged user to bypass validation, passing user input into the application as trusted data, due to client-side enforcement of server-side security...

CVE-2025-36102 IBM Controller Validation Bypass

IBM Controller 11.1.0 through 11.1.1 and IBM Cognos Controller 11.0.0 through 11.0.1 FP6 could allow a privileged user to bypass validation, passing user input into the application as trusted data, due to client-side enforcement of server-side security...

CVE-2025-36102

CVE-2025-36102 affects IBM Controller 11.1.0–11.1.1 and IBM Cognos Controller 11.0.0–11.0.1 FP6. The issue arises from client-side enforcement of server-side security, allowing a privileged user to bypass validation by passing user input into the application as trusted data. Impact described acro...