The paper proposes Dynamic Cyber Ranges, an advanced cyber range environment using LLM-driven Defender agents to counter the saturation of traditional security benchmarks, demonstrating that these dyn…

MLEvolve is a novel self-evolving multi-agent framework that enables LLM agents to discover and optimize machine learning algorithms for complex, long-horizon tasks.

The paper introduces CyberGym-E2E, a large-scale, end-to-end benchmark designed to comprehensively evaluate AI agents' capabilities across the entire lifecycle of real-world software vulnerability dis…

The paper re-evaluates LLM agents on CTFs, finding that while general-purpose agents like claude-code are strong baselines, specialized, modular architectures significantly improve performance and con…

Evo-Attacker introduces a memory-augmented reinforcement learning framework to perform generalized, long-horizon tool attacks on LLM-MAS, significantly outperforming existing methods.

The EvoSafety framework enhances LLM safety by externalizing attack and defense mechanisms, enabling persistent, transferable, and model-agnostic robustness against adversarial prompts.

The paper proposes Meta-Team, an experience-driven framework that enables multi-agent systems (MAS) to collaboratively self-evolve by transforming complex execution experiences into reusable improveme…

The paper empirically evaluates domain-adapted and general-purpose LLMs for structured threat modelling (STRIDE on 5G security), finding that domain adaptation and model size do not guarantee reliable…

The paper introduces a meta-scaffold, CSI, which demonstrates that combining multiple, structurally heterogeneous LLM-driven cybersecurity agents via a shared blackboard architecture significantly out…

The paper proposes an autonomous red teaming framework combining LLMs and RL to generate sophisticated, multi-stage cyber attack campaigns, demonstrating its necessity for evaluating robust AI-enabled…

HarnessForge introduces a meta-adaptive framework that jointly evolves the execution structure (harness) and the reasoning policy of LLM agents, significantly improving overall system performance acro…

PoC-Adapt is an end-to-end framework that significantly improves the reliability and efficiency of automated vulnerability exploitation by integrating semantic state validation and reinforcement learn…

Red-MIRROR is a novel multi-agent LLM system that automates complex web penetration testing by integrating a memory-reflection backbone, achieving superior performance on industry benchmarks.

The paper proposes the Layered Attack Surface Model (LASM), a structural taxonomy that maps security threats and defenses across the complex, multi-layered architecture of AI agents, revealing signifi…

The paper distinguishes between a model's ability to generate useful updates for external agent components (harness-updating) and its ability to benefit from those updates (harness-benefit), finding t…

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.

CoopGuard is a novel stateful, multi-round defense framework using cooperative agents to significantly reduce the success rate of evolving adversarial attacks against Large Language Models.

The paper evaluates Language Model Agents (LMAs) for red-teaming by benchmarking their ability to perform lateral movement, finding that expert-defined action plans are most effective, though all moda…

Adaptive Auto-Harness introduces a framework that enables LLM agents to sustain self-improvement and maintain high performance over open-ended, shifting task streams, outperforming existing fixed-benc…

This paper provides the first comprehensive security analysis of the Agent Skills framework, identifying severe structural vulnerabilities that require fundamental architectural changes rather than si…