The System Prompt Is the Attack Surface: How LLM Agent Configuration Shapes Security and Creates Exploitable Vulnerabilities

• URL: http://arxiv.org/abs/2603.25056v1
• Date: Thu, 26 Mar 2026 05:48:37 GMT
• Title: The System Prompt Is the Attack Surface: How LLM Agent Configuration Shapes Security and Creates Exploitable Vulnerabilities
• Abstract summary: PhishNChips is a study of 11 models under 10 prompt strategies.<n>A single model's phishing bypass rate ranges from under 1% to 97% depending on how it is configured.<n>We show that optimizing prompts around highly predictive signals can improve benchmark performance.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: System prompt configuration can make the difference between near-total phishing blindness and near-perfect detection in LLM email agents. We present PhishNChips, a study of 11 models under 10 prompt strategies, showing that prompt-model interaction is a first-order security variable: a single model's phishing bypass rate ranges from under 1% to 97% depending on how it is configured, while the false-positive cost of the same prompt varies sharply across models. We then show that optimizing prompts around highly predictive signals can improve benchmark performance, reaching up to 93.7% recall at 3.8% false positive rate, but also creates a brittle attack surface. In particular, domain-matching strategies perform well when legitimate emails mostly have matched sender and URL domains, yet degrade sharply when attackers invert that signal by registering matching infrastructure. Response-trace analysis shows that 98% of successful bypasses reason in ways consistent with the inverted signal: the models are following the instruction, but the instruction's core assumption has become false. A counter-intuitive corollary follows: making prompts more specific can degrade already-capable models by replacing broader multi-signal reasoning with exploitable single-signal dependence. We characterize the resulting tension between detection, usability, and adversarial robustness as a navigable tradeoff, introduce Safetility, a deployability-aware metric that penalizes false positives, and argue that closing the adversarial gap likely requires tool augmentation with external ground truth.

Related papers

• Behavioral Grammar: Detecting Adaptive Malware via Tiny Language Model Priors and Second-Order Temporal Analysis [1.560152792366037]
  This paper presents Behavioral Grammar, a detection architecture that treats host runtime behavior as a structured language.<n>Each system event is discretized into an 8-token representation spanning event type, process, argument skeleton, path category, parent process, user, destination, and inter-event timing.<n>We frame these findings within a coevolutionary economics model, arguing that behavior-grammar detection shifts the evasion cost from rule circumvention to distribution matching.
  arXiv  Detail & Related papers  (2026-08-01T16:28:38Z)
• GuardPhish: Securing Open-Source LLMs from Phishing Abuse [0.7009487789080343]
  Open-source Large Language Models (LLMs) in offline and enterprise environments have introduced a largely unexamined security risk.<n>We investigate this vulnerability through GuardPhish, a large scale multi-vector phishing prompt dataset.<n>Using a deterministic five model ensemble for labeling, we achieve near perfect inter model agreement.
  arXiv  Detail & Related papers  (2026-04-19T08:02:57Z)
• Repurposing Backdoors for Good: Ephemeral Intrinsic Proofs for Verifiable Aggregation in Cross-silo Federated Learning [30.440611659881494]
  We propose a lightweight architecture that shifts from cryptographic proofs to textitIntrinsic Proofs<n>We harness Catastrophic Forgetting to embed verification signals directly into model parameters.<n>Our approach achieves over $1000times$ speedup on ResNet-18 compared to cryptographic baselines, effectively scaling to large models.
  arXiv  Detail & Related papers  (2026-03-11T12:04:18Z)
• TraceGuard: Process-Guided Firewall against Reasoning Backdoors in Large Language Models [19.148124494194317]
  We propose TraceGuard, a process-guided security framework that transforms small-scale models into robust reasoning firewalls.<n>Our approach treats the reasoning trace as an untrusted payload and establishes a defense-in-depth strategy.<n>We demonstrate robustness against adaptive adversaries in a grey-box setting, establishing TraceGuard as a viable, low-latency security primitive.
  arXiv  Detail & Related papers  (2026-03-02T22:19:13Z)
• Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models [68.45272703833209]
  We show that state-of-the-art PRMs are systematically exploitable under adversarial optimization pressure.<n>We introduce a three-tiered diagnostic framework that applies increasing adversarial pressure to quantify these vulnerabilities.<n>We release PRM-BiasBench and a diagnostic toolkit to enable robustness evaluation before deployment.
  arXiv  Detail & Related papers  (2026-02-20T23:38:03Z)
• ReasAlign: Reasoning Enhanced Safety Alignment against Prompt Injection Attack [52.17935054046577]
  We present ReasAlign, a model-level solution to improve safety alignment against indirect prompt injection attacks.<n>ReasAlign incorporates structured reasoning steps to analyze user queries, detect conflicting instructions, and preserve the continuity of the user's intended tasks.
  arXiv  Detail & Related papers  (2026-01-15T08:23:38Z)
• DISTIL: Data-Free Inversion of Suspicious Trojan Inputs via Latent Diffusion [0.7351161122478707]
  Deep neural networks are vulnerable to Trojan (backdoor) attacks.<n> triggerAdaptive inversion reconstructs malicious "shortcut" patterns inserted by an adversary during training.<n>We propose a data-free, zero-shot trigger-inversion strategy that restricts the search space while avoiding strong assumptions on trigger appearance.
  arXiv  Detail & Related papers  (2025-07-30T16:31:13Z)
• MultiPhishGuard: An LLM-based Multi-Agent System for Phishing Email Detection [3.187381965457262]
  MultiPhishGuard is a dynamic multi-agent detection system that synergizes specialized expertise with adversarial-aware reinforcement learning.<n>Our framework employs five cooperative agents with automatically adjusted decision weights powered by a Proximal Policy Optimization reinforcement learning algorithm.<n>Experiments demonstrate that MultiPhishGuard achieves high accuracy (97.89%) with low false positive (2.73%) and false negative rates (0.20%)
  arXiv  Detail & Related papers  (2025-05-26T23:27:15Z)
• EXPLICATE: Enhancing Phishing Detection through Explainable AI and LLM-Powered Interpretability [44.2907457629342]
  EXPLICATE is a framework that enhances phishing detection through a three-component architecture.<n>It is on par with existing deep learning techniques but has better explainability.<n>It addresses the critical divide between automated AI and user trust in phishing detection systems.
  arXiv  Detail & Related papers  (2025-03-22T23:37:35Z)
• Trigger without Trace: Towards Stealthy Backdoor Attack on Text-to-Image Diffusion Models [70.03122709795122]
  Backdoor attacks targeting text-to-image diffusion models have advanced rapidly.<n>Current backdoor samples often exhibit two key abnormalities compared to benign samples.<n>We propose Trigger without Trace (TwT) by explicitly mitigating these consistencies.
  arXiv  Detail & Related papers  (2025-03-22T10:41:46Z)
• Neural Antidote: Class-Wise Prompt Tuning for Purifying Backdoors in CLIP [51.04452017089568]
  Class-wise Backdoor Prompt Tuning (CBPT) is an efficient and effective defense mechanism that operates on text prompts to indirectly purify CLIP.<n>CBPT significantly mitigates backdoor threats while preserving model utility.
  arXiv  Detail & Related papers  (2025-02-26T16:25:15Z)
• Lazy Layers to Make Fine-Tuned Diffusion Models More Traceable [70.77600345240867]
  A novel arbitrary-in-arbitrary-out (AIAO) strategy makes watermarks resilient to fine-tuning-based removal. Unlike the existing methods of designing a backdoor for the input/output space of diffusion models, in our method, we propose to embed the backdoor into the feature space of sampled subpaths. Our empirical studies on the MS-COCO, AFHQ, LSUN, CUB-200, and DreamBooth datasets confirm the robustness of AIAO.
  arXiv  Detail & Related papers  (2024-05-01T12:03:39Z)
• On Trace of PGD-Like Adversarial Attacks [77.75152218980605]
  Adversarial attacks pose safety and security concerns for deep learning applications. We construct Adrial Response Characteristics (ARC) features to reflect the model's gradient consistency. Our method is intuitive, light-weighted, non-intrusive, and data-undemanding.
  arXiv  Detail & Related papers  (2022-05-19T14:26:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.