AI and LLM Bot Management Has Become a Business-Critical Issue: Do It Right

AI bots, agents, and LLM scrapers all want your content. Here’s how to manage them so that they help, not hinder, your business...

CVE-2025-53630

llama.cpp is an inference of several LLM models in C/C++. Integer Overflow in the ggufinitfromfileimpl function in ggml/src/gguf.cpp can lead to Heap Out-of-Bounds Read/Write. This vulnerability is fixed in commit 26a48ad699d50b6268900062661bd22f3e792579...

CAVGAN: Unifying Jailbreak and Defense of LLMs Via Generative Adversarial Attacks on Their Internal Representations

Security alignment enables the Large Language Model LLM to gain the protection against malicious queries, but various jailbreak attack methods reveal the vulnerability of this security mechanism. Previous studies have isolated LLM jailbreak attacks and defenses. We analyze the security protection...

dify 跨站脚本漏洞

dify is an open source LLM application development platform from LangGenius Open Source. A cross-site scripting vulnerability exists in versions prior to dify 1.1.3, which stems from uncleared input in the Firefox browser and could lead to cross-site scripting attacks...

Attention Slipping: a Mechanistic Understanding of Jailbreak Attacks and Defenses in LLMs

As large language models LLMs become more integral to society and technology, ensuring their safety becomes essential. Jailbreak attacks exploit vulnerabilities to bypass safety guardrails, posing a significant threat. However, the mechanisms enabling these attacks are not well understood. In thi...

MCP Inspector < 0.14.1 Remote Code Execution

According to the self-reported version number, the version of MCP Inspector hosted on the remote is affected by a Remote Code Execution vulnerability. Note that the scanner has not tested for these issues but has instead relied only on the application's self-reported version number. This detectio...

CVE-2025-45809

SQL Injection vulnerability in BerriAI LiteLLM before 1.81.0 allows attackers to execute arbitrary commands via the key parameter to the "/key/block" and "/key/unblock" API endpoints...

DocsGPT Detected

This is an informational plugin to inform the user that the scanner has detected a publicly accessible DocsGPT instance on the target application. DocsGPT is an open-source genAI tool that helps users get answers from knowledge source. This detection is included in the AI and LLM category. No...

CVE-2025-52566

llama.cpp is an inference of several LLM models in C/C++. Prior to version b5721, there is a signed vs. unsigned integer overflow in llama.cpp's tokenizer implementation llamavocab::tokenize src/llama-vocab.cpp:3036 resulting in unintended behavior in tokens copying size comparison. Allowing...

CVE-2025-6518 PySpur-Dev pyspur Jinja2 Template single_llm_call.py SingleLLMCallNode special elements used in a template engine

A vulnerability was found in PySpur-Dev pyspur up to 0.1.18. It has been classified as critical. Affected is the function SingleLLMCallNode of the file backend/pyspur/nodes/llm/singlellmcall.py of the component Jinja2 Template Handler. The manipulation of the argument usermessage leads to imprope...

CVE-2025-49149

Dify is an open-source LLM app development platform. In version 1.2.0, there is insufficient filtering of user input by web applications. Attackers can use website vulnerabilities to inject malicious script code into web pages. This may result in a cross-site scripting XSS attack when a user...

Watermarking LLM-Generated Datasets in Downstream Tasks

Large Language Models LLMs have experienced rapid advancements, with applications spanning a wide range of fields, including sentiment classification, review generation, and question answering. Due to their efficiency and versatility, researchers and companies increasingly employ LLM-generated da...

MCP Server SSE DNS Rebinding

ModelContextProtocol MCP servers using SSE Server-Sent Events transport mode are prone to DNS rebinding attacks when they do not enforce strict verification of both the 'Origin' and 'Host' headers. This vulnerability allows an attacker to bypass same-origin policies, potentially leading to...

AgentSmith Flaw in LangSmith’s Prompt Hub Exposed User API Keys, Data

A CVSS 8.8 AgentSmith flaw in LangSmith's Prompt Hub exposed AI agents to data theft and LLM manipulation. Learn how malicious AI agents could steal API keys and hijack LLM responses. Fix deployed...

CVE-2025-49149

Dify, an open-source LLM app platform, has a vulnerability in version 1.2.0 caused by insufficient filtering of user input in web applications, allowing injection of malicious script and potentially leading to cross-site scripting (XSS) when users browse affected pages. The CVE entries consistent...

AgentVigil: Generic Black-Box Red-Teaming for Indirect Prompt Injection against LLM Agents

The strong planning and reasoning capabilities of Large Language Models LLMs have fostered the development of agent-based systems capable of leveraging external tools and interacting with increasingly complex environments. However, these powerful features also introduce a critical security risk:...

Zero-Click AI Vulnerability Exposes Microsoft 365 Copilot Data Without User Interaction

A novel attack technique named EchoLeak has been characterized as a "zero-click" artificial intelligence AI vulnerability that allows bad actors to exfiltrate sensitive data from Microsoft 365 M365 Copilot's context sans any user interaction. The critical-rated vulnerability has been assigned the...

Llms.txt File Detected

This is an informational plugin to inform the user that the scanner has detected a publicly accessible 'llms.txt' file on the target application. The 'llms.txt' file is a proposal designed to provide LLM-friendly content written in markdown for LLMs usage. This detection is included in the AI and...

Enabling Secure AI Inference: Trend Cybertron Leverages NVIDIA Universal LLM NIM Microservices

Learn how Trend's Cybertron has been harnessing the power of NVIDIA Universal LLM NIM Microservices...

JavelinGuard: Low-Cost Transformer Architectures for LLM Security

We present JavelinGuard, a suite of low-cost, high-performance model architectures designed for detecting malicious intent in Large Language Model LLM interactions, optimized specifically for production deployment. Recent advances in transformer architectures, including compact BERTDevlin et al...