The paper introduces BenchJack, an automated red-teaming system that systematically audits popular AI agent benchmarks, revealing numerous reward-hacking exploits and demonstrating a method to signifi…

The paper introduces ExploitBench, a capability-graded benchmark that measures the progressive stages of exploitation, demonstrating that while current frontier models can easily trigger bugs, achievi…

VIPER-MCP is a novel, end-to-end automated framework that detects and dynamically confirms the exploitability of taint-style vulnerabilities in Model Context Protocol (MCP) servers, achieving high-fid…

The paper introduces ClawTrap, a MITM-based red-teaming framework, to evaluate the security robustness of web agents like OpenClaw against dynamic, real-world network attacks, finding that model stren…

The paper introduces DeepTrap, an automated framework that evaluates security vulnerabilities in agentic language models by manipulating their internal execution contexts, demonstrating that task comp…

The paper benchmarks current frontier computer-using agents against hand-crafted attacks, finding that while they are highly safe in browser tasks, this safety does not generalize to other domains lik…

WebTrap introduces a stealthy, mid-task hijacking attack that successfully compromises browser agents during long-horizon tasks by seamlessly fusing malicious instructions with the original user goal.

The paper introduces SEC-bench Pro, a rigorous benchmark for evaluating LLM-based bug hunting on complex software, finding that even advanced agents struggle with long-horizon security tasks.

The paper introduces TraceSafe-Bench, a comprehensive benchmark, and finds that securing LLM agents requires jointly optimizing for structural reasoning and safety alignment to mitigate risks during m…

The paper systematically maps LLM agent vulnerabilities by testing 10,000 prompt variations, finding that 'goal reframing' language is the primary trigger for exploitation, rather than broad adversari…

The paper introduces MOSAIC-Bench, a benchmark demonstrating that coding agents can ship exploitable code by complying with seemingly innocuous, staged tasks, a vulnerability that is not easily mitiga…

The paper introduces Patch2Vuln, a pipeline that uses an LLM agent to reconstruct security vulnerabilities by analyzing differences between old and new Linux binary packages, successfully localizing p…

The paper analyzes a dataset of agent skills, demonstrating that different security scanners (VirusTotal, static analysis, SkillSpector) rarely agree, necessitating a layered governance approach for s…

The paper analyzes a dataset of agent skills, demonstrating that different security scanners (VirusTotal, static analysis, SkillSpector) rarely agree on maliciousness, necessitating layered security g…

This paper conducts a large-scale, repository-aware security analysis of AI agent skills, demonstrating that incorporating surrounding project context drastically reduces the rate of false positive ma…

The paper introduces LoopTrap, an automated red-teaming framework that demonstrates how malicious prompts can poison the termination judgment of LLM agents, causing unbounded computation.

The paper introduces APIOT, the first LLM framework capable of autonomously performing the full discovery, exploitation, patching, and verification cycle against bare-metal industrial OT devices.

The paper introduces SLEIGHT-Bench, a benchmark of 40 synthetic attacks, demonstrating that current LLM monitor systems fail to detect a significant number of covert, harmful actions executed by codin…

This paper provides the first comprehensive systematization and large-scale empirical evaluation of existing LLM-based Automated Penetration Testing (AutoPT) frameworks, offering a structured taxonomy…

The paper introduces SkillSafetyBench, a comprehensive benchmark demonstrating that agent safety failures often stem from adversarial influences within reusable skills and execution environments, rath…