CyberLLM is presented, a multi-agent, LLM-orchestrated framework that autonomously detects vulnerabilities and executes remediations under a formal, runtime safety guard, and indicates that LLM agents can perform useful autonomous cyber-defense when wrapped in a deterministic, auditable safety envelope.
Abstract
Software-Defined Vehicles (SDVs) expand the automotive attack surface across source code, runtime logs, and deployment topologies, while safety constraints forbid autonomous agents from acting without oversight. This paper presents CyberLLM, a multi-agent, LLM-orchestrated framework that autonomously detects vulnerabilities and executes remediations under a formal, runtime safety guard. Detection combines a deterministic layer (regex rules, AST analyzers, and topology graph checks) with an LLM refinement pass, so a high-recall floor is complemented by high-precision reasoning. A decision agent aggregates findings, tags them with a human-centric asset taxonomy, and selects a tiered response, ratcheting its confidence with signed cross-session memory and re-planning feedback. Every action is validated against four contextual security properties and an independent action-alignment oracle before it is allowed to run, and refused actions trigger escalation and re-planning. A symmetric attack pipeline generates and replays exploits so both sides can be exercised on the same scenarios. On an independent, ground-truthed benchmark of nine original automotive ECU modules in C, C++, and Rust seeding 47 layered vulnerabilities plus clean controls, the always-on deterministic layer covers 34\% of the labeled vulnerabilities at perfect precision, and adding the grounded LLM refinement and completeness passes roughly doubles coverage to about 70\% (F1 $0.83$) while producing zero false positives on the clean controls. The results indicate that LLM agents can perform useful autonomous cyber-defense when wrapped in a deterministic, auditable safety envelope.
Agentic artificial intelligence (AI) is increasingly being explored for automating offensive security and red teaming tasks, enabling systems that can coordinate multi-step cyber operations through structured decision-making. While prior research has investigated reinforcement learning (RL) agents and large language models (LLMs) for penetration testing, most studies are evaluated in simulated or abstract environments, with limited empirical validation in real cyber-range settings. This paper presents a controlled experimental evaluation of an LLM-guided offensive security pipeline against deterministic scripted baselines in a cyber-range environment. Using a vulnerable Kioptrix virtual machine and a Kali Linux attacker, we implement three deterministic pipelines, fixed-path, sequential and rule-based, alongside two configurations of an LLM-guided agent: an initial version (LLM V1) and a refined constrained controller (LLM V2). All approaches operate within a restricted and auditable action space executed through predefined tools. Across repeated trials, the initial LLM configuration exhibits reduced reliability and increased execution cost due to exploratory behaviour. In contrast, the refined controller achieves a 100% success rate, reduces execution steps, eliminates wasted actions and matches the efficiency of rule-based automation. These results show that, within a controlled cyber-range environment, LLM-guided agents can approximate deterministic performance when appropriate constraints are applied. This suggests that agent-based approaches to offensive security can support semi-autonomous red teaming workflows, provided that decision-making is governed by structured control policies.
Atif Chowhan, Yasir Hamid· Journal of Information &...· 0 citations
Public reports and open-source resources expose many EDR evasion techniques, but it remains unclear whether commercial Endpoint Detection and Response (EDR) systems can withstand these documented attacks. Evaluating them requires turning fragmented security knowledge into working payloads and refining those payloads from opaque alerts, tasks that existing automation does not address. We present AutoBypass, a knowledge-grounded, closed-loop multi-agent framework for automated EDR resilience assessment. A Detection-Aware Knowledge Base structures threat intelligence, expert analyses, and open-source proofs of concept into evasion techniques and operational constraints. Agents use this knowledge to plan attacks, generate polymorphic code, and compile binaries, while a telemetry-driven reasoning engine diagnoses failures and feeds corrective evidence back into the strategy. Across seven commercial endpoint security platforms, AutoBypass bypassed every target, reaching 90% evasion against Windows Defender and 86.7% against Trend Micro AV. Ablations show that the knowledge base raises the success rates of 8B open-weight models from 27--53% to 43--83%, bringing them close to large proprietary models. These results demonstrate a systematic way to operationalize public security knowledge for continuous, automated assessment of EDR resilience.
Wei Yuan, Wenbo Guo, Qingyun Du et al.· 0 citations
As large language model (LLM) agents move from conversation to executing code, reading local files, and orchestrating external tools, a single agent hijacked by a malicious third-party skill can cause data exfiltration, privilege escalation, or cascading compromise. We argue that agentic risk is progressive: it can enter at four loci of the agent control loop--skill admission, invocation-time intent, execution-time effect, and post-action consequence--while a denied dangerous objective can reappear across surface forms, tools, or turns; existing safeguards are typically local to one lifecycle boundary or one call. Guided by this threat model, we present ClawSentry, an open-source, framework-agnostic security supervision gateway for agent runtimes. Before a skill package is ever executed, First-use Skill Package Review (FSPR) audits it under a deterministic evidence floor, escalating unresolved cases to bounded read-only agentic review (locus A). At runtime, a three-tier progressive decision engine--a deterministic L1 layer, a rule-anchored L2 semantic reviewer, and a read-only L3 evidence-seeking agent--spends contextual review only on the residual ambiguity, while a session-level anti-bypass mechanism recognizes tool-switching and rephrased retries (loci B--C); a post-action path feeds high-severity evidence non-retroactively into later review (locus D). An Agent Harness Protocol (AHP) abstraction applies one policy across Codex, Claude Code, Kimi CLI, and Gemini CLI without modifying agent internals. On SkillInject with Codex/GPT-5.4, contextual ASR falls from 39.55% to 2.61% while contextual TSR moves only from 83.78% to 83.05%. Across five Work Agents on the full SkillsSafety benchmark, ClawSentry confines ASR to 9.09--15.03% from 33.5--49.7% unprotected, and aggregate TSR on clean skills remains 98.7%.
Kai Wang, Zeming Wei, Biaojie Zeng et al.· 0 citations
Large language model (LLM) agents combine language-based reasoning with external tools to perform complex tasks. Adversarial inputs can exploit interactions between the agent and its environment, causing the agent to violate safety policies during execution. Yet existing evaluations often reduce agent safety to a single attack success rate (ASR), collapsing exposure, execution, observation, and adjudication and potentially conflating actual violations with evidence visibility. We introduce REDAgentBench, an executable framework for autonomous red-teaming and faithful measurement. It derives attacks from explicit safety constraints and associated agent-system vulnerabilities, runs them in isolated service sandboxes, and verifies harmful effects from service receipts and final-state changes. The benchmark contains 1,661 cases across five service surfaces. Across six models and three agent harnesses, macro-average ASR is 65.69%; reported ASR varies with harness and evidence view, while evaluation-context disclosure changes execution behavior. In a state-grounded diagnostic cohort, almost one in five confirmed violations with resolved action anchors occurs after the agent states the relevant constraint or risk, revealing a Recognition--Execution Gap. Finally, a training-free policy reminder reduces confirmed violations by more than 70 percentage points in matched replay. These findings show that executable evaluation can improve safety measurement and identify actionable intervention points.
Zixing Chen, Xingyuan Liu, Jie Zhu et al.· 0 citations
Autonomous LLM agents with tool execution capabilities introduce severe security risks through prompt injection, goal hijacking, and unauthorized action invocation. Existing guardrails rely on unverified, host local software filters system prompts, semantic classifiers, policy engines that share the execution environment of the untrusted agent, offering no guarantee to an external observer that a safety policy was correctly evaluated. A compromised host produces no evidence of its own failure. This paper presents NiyamAI, an intent bound runtime guardrail architecture providing cryptographically verifiable execution integrity for autonomous agents. At session initialization, permitted tools and operational constraints are sealed into an immutable Intent Contract under a SHA256 commitment. Every tool invocation is intercepted by a deterministic authority gate and classified by a dedicated neural Judge (11->8->2 feedforward network). For each authorized action, NiyamAI generates a succinct zkSNARK proof certifying correct policy evaluation under the committed contract; execution proceeds only after that proof verifies. Across 2,000 AgentSafetyBench scenarios under 5fold stratified crossvalidation with out of fold scoring, NiyamAI achieves 88.8% F1 at a 1.0% false positive rate (bootstrap 95% CI [85.5%, 92.1%]), against 66.8% for Llama Prompt Guard 2, 46.2% for GPTOSSSafeguard, and 40.4% for NeMo Guardrails; McNemar's exact test confirms each margin at p<0.0001. Proof generation adds 1.7 s per approved action, verification 51 ms, with an 18.6 KB proof verifiable by any third party without access to model parameters. We further subject NiyamAI's own enforcement mechanism to 18 adversarial vectors across six classes, disclosing two implementation vulnerabilities identified and remediated during development.
Aditya Katkar, Om Karkele, Kartik Mandhane et al.· 0 citations
Cyber-capable AI agents combine language models with tools, memory, and execution environments to perform multi-step offensive-security tasks. Existing work separately measures cyber capability and catalogs attacks against agent components, but provides less guidance on containing a capable agent within the environments used to evaluate it. This review synthesizes five vulnerability classes at that boundary: multi-step offensive chains, objectives that conflict with sandbox boundaries, supply-chain and credential exposure, persistent command-and-control, and the speed of automated action. We use two separate preliminary incident records: the reported July 2026 Hugging Face/OpenAI evaluation breach and Anthropic's subsequent three-incident evaluation review. A comparative evidence protocol distinguishes record-specific factual claims from the shared systems lesson: the evaluation environment is itself part of the security boundary. Across the taxonomy and records, we examine controls for containment, privilege separation, provenance, and responder access, including the dual-use problem that defensive artifacts may also enable misuse. The review identifies practical priorities for evaluating cyber capability together with the security of the environment in which that capability is exercised.