Detecting and analyzing prompt abuse in AI tools

This second post in our AI Application Security series is all about moving from planning to practice. AI Application Series 1: Security considerations when adopting AI tools established how AI adoption expands the attack surface and our threat-modelling guidance on the Microsoft security blog...

DRUPAL-CONTRIB-2026-028

The module and certain submodules AI Automators, AI Translate, AI API Explorer, AI Content Suggestions provide the ability to use an LLM to generate HTML or Markdown and preview it in a browser. Under certain circumstances, rendering of this HTML can lead to exposing secret communications in the...

AttriGuard: Defeating Indirect Prompt Injection in LLM Agents Via Causal Attribution of Tool Invocations

LLM agents are highly vulnerable to Indirect Prompt Injection IPI, where adversaries embed malicious directives in untrusted tool outputs to hijack execution. Most existing defenses treat IPI as an input-level semantic discrimination problem, which often fails to generalize to unseen payloads. We...

CLIOPATRA: Extracting Private Information from LLM Insights

As AI assistants become widely used, privacy-aware platforms like Anthropic's Clio have been introduced to generate insights from real-world AI use. Clio's privacy protections rely on layering multiple heuristic techniques together, including PII redaction, clustering, filtering, and LLM-based...

PT-2026-24113

vLLM is an inference and serving engine for large language models LLMs. The SSRF protection fix for CVE-2026-24779 add in 0.15.1 can be bypassed in the load from url async method due to inconsistent URL parsing behavior between the validation layer and the actual HTTP client. The SSRF fix uses...

OpenAnt LLM-Based Vulnerability Discovery

OpenAnt from Knostic is an open source LLM-based vulnerability discovery product that helps defenders proactively find verified security flaws while minimizing both false positives and false negatives. Stage 1 detects. Stage 2 attacks. What survives is real...

Targeted Bit-Flip Attacks on LLM-Based Agents

Targeted bit-flip attacks BFAs exploit hardware faults to manipulate model parameters, posing a significant security threat. While prior work targets single-step inference models e.g., image classifiers, LLM-based agents with multi-stage pipelines and external tools present new attack surfaces,...

SUSE CVE-2026-25802

New API is a large language mode LLM gateway and artificial intelligence AI asset management system. Prior to version 0.10.8-alpha.9, a potential unsafe operation occurs in component MarkdownRenderer.jsx, allowing for Cross-Site ScriptingXSS when the model outputs items containing tag. Version...

Jailbreak Foundry: From Papers to Runnable Attacks for Reproducible Benchmarking

Jailbreak techniques for large language models LLMs evolve faster than benchmarks, making robustness estimates stale and difficult to compare across papers due to drift in datasets, harnesses, and judging protocols. We introduce JAILBREAK FOUNDRY JBF, a system that addresses this gap via a...

CVE-2026-25802

New API is a large language mode LLM gateway and artificial intelligence AI asset management system. Prior to version 0.10.8-alpha.9, a potential unsafe operation occurs in component MarkdownRenderer.jsx, allowing for Cross-Site ScriptingXSS when the model outputs items containing...

ai-security-toolkit

...

CVE-2026-25802 New API has Potential XSS in its MarkdownRenderer component

New API is a large language mode LLM gateway and artificial intelligence AI asset management system. Prior to version 0.10.8-alpha.9, a potential unsafe operation occurs in component MarkdownRenderer.jsx, allowing for Cross-Site ScriptingXSS when the model outputs items containing tag. Version...

ICSSPulse: A Modular LLM-Assisted Platform for Industrial Control System Penetration Testing

It is well established that industrial control systems comprise the operational backbone of modern critical infrastructures, yet their increasing connectivity exposes them to cyber threats that are difficult to study and remedy safely under real-time operational conditions. In this paper, we...

LLM Scalability Risk for Agentic-AI and Model Supply Chain Security

Large Language Models LLMs & Generative AI are transforming cybersecurity, enabling both advanced defenses and new attacks. Organizations now use LLMs for threat detection, code review, and DevSecOps automation, while adversaries leverage them to produce malwares and run targeted social-engineeri...

Kestrel

Kestrel LLM-Assisted Bug Bounty Hunting Platform for Kali L...

OpenClaw: Unsanitized CWD path injection into LLM prompts

Overview OpenClaw embedded the current working directory workspace path into the agent system prompt without sanitization. If an attacker can cause OpenClaw to run inside a directory whose name contains control/format characters for example newlines or Unicode bidi/zero-width markers, those...

Can Adversarial Code Comments Fool AI Security Reviewers -- Large-Scale Empirical Study of Comment-Based Attacks and Defenses against LLM Code Analysis

AI-assisted code review is widely used to detect vulnerabilities before production release. Prior work shows that adversarial prompt manipulation can degrade large language model LLM performance in code generation. We test whether similar comment-based manipulation misleads LLMs during...

GHSA-W5CR-2QHR-JQC5 Cloudflare Agents has a Reflected Cross-Site Scripting (XSS) vulnerability in AI Playground site

Summary A Reflected Cross-Site Scripting XSS vulnerability was discovered in the AI Playground's OAuth callback handler. The errordescription query parameter was directly interpolated into an HTML script tag without proper escaping, allowing attackers to execute arbitrary JavaScript in the contex...

VulReaD: Knowledge-Graph-Guided Software Vulnerability Reasoning and Detection

Software vulnerability detection SVD is a critical challenge in modern systems. Large language models LLMs offer natural-language explanations alongside predictions, but most work focuses on binary evaluation, and explanations often lack semantic consistency with Common Weakness Enumeration CWE...

LLM-FS: Zero-Shot Feature Selection for Effective and Interpretable Malware Detection

Feature selection FS remains essential for building accurate and interpretable detection models, particularly in high-dimensional malware datasets. Conventional FS methods such as Extra Trees, Variance Threshold, Tree-based models, Chi-Squared tests, ANOVA, Random Selection, and Sequential...