Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents

arXiv:2605.08442v4 Announce Type: replace-cross Abstract: Persistent memory in LLM agents creates an attack surface that production safety classifiers do not observe: the payload enters via RAG retrieval and persists across sessions via tool-mediated memory. We evaluate six defenses across four architectural layers against delayed-trigger attacks on nine open-source models (5,040 runs, N=40 per condition). Five of six defenses fail: input-level filters never see the payload (it enters via RAG, not user input); retrieval-level classifiers observe it but cannot distinguish compliance-framed injection from legitimate policy; instruction-level hardening is overridden by the stored rule's compliance framing. Only tool-gating at the memory layer (Memory Sandbox) reduces ASR to 0% for eight of nine models, with zero utility cost. A reasoning model inverts this defense via goal-directed RAG fallback, a mechanism that replicates cross-family on Bedrock. A reasoning-mode ablation reveals a double dissociation: no single sandbox implementation is safe across both reasoning and non-reasoning model classes. We resolve this with a content-layer proof-of-concept (RATG), validated on non-reasoning models. A loaded-corpus frontier evaluation (21 models, 3 providers, N=40) overturns an initial empty-corpus screen showing 0/210 exfiltrations: that was a threat-model artifact, not model safety. Under realistic conditions, Gemini 3.1 Pro Preview exfiltrates at 95% ASR, GPT-5.1 regresses to 22.5% relative to GPT-5 (5%), and Anthropic blocks at the injection layer (0-17.5% storage, 0% ASR). Nearly all OpenAI and Gemini models store the rule at 100% regardless of execution resistance, creating supply-chain risk in shared-memory deployments. Defense effectiveness is determined by architectural layer and reasoning capability, not classifier quality.