CyberLLM: A Multi-Agent LLM Framework for Autonomous Detection and Guarded Response in Automotive Cybersecurity

2026-08-06

AI Summary

Software-defined vehicles significantly expand the attack surface, yet autonomous defense mechanisms must operate under stringent safety constraints. This work proposes a multi-agent large language model (LLM) framework augmented with runtime safety guards that integrates rule-based detection and LLM-driven reasoning within a deterministic safety envelope to enable automated threat detection and tiered response. The approach innovatively incorporates asset classification, cross-session memory, four-dimensional contextual safety attribute verification, and a symmetric attack replay mechanism to ensure action compliance and auditability. Evaluated on an ECU benchmark comprising 47 hierarchical vulnerabilities, the system achieves 70% vulnerability coverage (F1 = 0.83), with the deterministic layer attaining 34% perfect precision and zero false positives.

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.

Problem

• Automotive Cybersecurity
• Software-Defined Vehicles
• Autonomous Detection
• Safety Constraints
• Vulnerability Remediation

Innovation

• Multi-Agent LLM
• Automotive Cybersecurity
• Runtime Safety Guard
• Deterministic-LLM Hybrid Detection
• Action Alignment Oracle