SafeHarness: Lifecycle-Integrated Security Architecture for LLM-based Agent Deployment

TL;DR

SafeHarness is a comprehensive security architecture for LLM-based agents that integrates multiple defense layers into the agent lifecycle to mitigate attacks and reduce unsafe behaviors.

Contribution

It introduces a novel multi-layered security framework that addresses structural security gaps in LLM agent deployment, with integrated verification, privilege control, and rollback mechanisms.

Findings

Achieves 38% reduction in unsafe behavior rate.

Achieves 42% reduction in attack success rate.

Maintains core task utility while enhancing security.

Abstract

The performance of large language model (LLM) agents depends critically on the execution harness, the system layer that orchestrates tool use, context management, and state persistence. Yet this same architectural centrality makes the harness a high-value attack surface: a single compromise at the harness level can cascade through the entire execution pipeline. We observe that existing security approaches suffer from structural mismatch, leaving them blind to harness-internal state and unable to coordinate across the different phases of agent operation. In this paper, we introduce \safeharness{}, a security architecture in which four proposed defense layers are woven directly into the agent lifecycle to address above significant limitations: adversarial context filtering at input processing, tiered causal verification at decision making, privilege-separated tool control at action execution, and safe rollback with adaptive degradation at state update. The proposed cross-layer mechanisms tie these layers together, escalating verification rigor, triggering rollbacks, and tightening tool privileges whenever sustained anomalies are detected. We evaluate \safeharness{} on benchmark datasets across diverse harness configurations, comparing against four security baselines under five attack scenarios spanning six threat categories. Compared to the unprotected baseline, \safeharness{} achieves an average reduction of approximately 38\% in UBR and 42\% in ASR, substantially lowering both the unsafe behavior rate and the attack success rate while preserving core task utility.