Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents

• URL: http://arxiv.org/abs/2602.07652v1
• Date: Sat, 07 Feb 2026 18:27:47 GMT
• Title: Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents
• Abstract summary: We introduce **AgentFence**, an architecture-centric security evaluation that defines 14 trust-boundary attack classes spanning planning, memory, retrieval, tool use, and delegation.<n>Hold the base model fixed, we evaluate eight agent archetypes under persistent multi-turn interaction.<n>Highest-risk classes are operational: Denial-of-Wallet ($0.62 pm 0.08$), Authorization Confusion ($0.54 pm 0.10$), Retrieval Poisoning ($0.47 pm 0.09$), and Planning Manipulation ($0.44 pm 0.11$)
• Score: 0.9368753183086048
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models are increasingly deployed as *deep agents* that plan, maintain persistent state, and invoke external tools, shifting safety failures from unsafe text to unsafe *trajectories*. We introduce **AgentFence**, an architecture-centric security evaluation that defines 14 trust-boundary attack classes spanning planning, memory, retrieval, tool use, and delegation, and detects failures via *trace-auditable conversation breaks* (unauthorized or unsafe tool use, wrong-principal actions, state/objective integrity violations, and attack-linked deviations). Holding the base model fixed, we evaluate eight agent archetypes under persistent multi-turn interaction and observe substantial architectural variation in mean security break rate (MSBR), ranging from $0.29 \pm 0.04$ (LangGraph) to $0.51 \pm 0.07$ (AutoGPT). The highest-risk classes are operational: Denial-of-Wallet ($0.62 \pm 0.08$), Authorization Confusion ($0.54 \pm 0.10$), Retrieval Poisoning ($0.47 \pm 0.09$), and Planning Manipulation ($0.44 \pm 0.11$), while prompt-centric classes remain below $0.20$ under standard settings. Breaks are dominated by boundary violations (SIV 31%, WPA 27%, UTI+UTA 24%, ATD 18%), and authorization confusion correlates with objective and tool hijacking ($ρ\approx 0.63$ and $ρ\approx 0.58$). AgentFence reframes agent security around what matters operationally: whether an agent stays within its goal and authority envelope over time.

Related papers

• EvoSafeHarness: Evolving Model- and Domain-Specific Harnesses for Securing Agents [74.54929751174289]
  EvoSafeHarness is a safety-specific optimization framework that synthesizes a deployable harness for a frozen model in a target domain.<n>It reduces average attack success rate from 45.6% to 10.0% at a 3.3-point utility cost.<n>It also achieves the best score on Agent-SafetyBench for every victim.
  arXiv  Detail & Related papers  (2026-09-05T05:56:41Z)
• Privacy-Preserving Topology-Guided Safety for LLM-Based Multi-Agent Systems via Federated Graph Learning [67.04448659688579]
  FGLGuard cuts AgentDojo's ground-success rate by 43% at near-unguarded utility, zero API cost, and negligible capability loss.<n>Federation is not optional: off-the-shelf transfer collapses under distribution shift.<n>Live FGLGuard cuts AgentDojo's ground-success rate by 43% at near-unguarded utility, zero API cost, and negligible capability loss.
  arXiv  Detail & Related papers  (2026-09-02T07:57:12Z)
• AgentWorld: Personality-Aware Reliability Evaluation for Agentic Information Retrieval [1.1647715835517187]
  AgentWorld is a simulation framework combining personality-driven user populations with stateful tool-use environments.<n>Three experiments demonstrate the framework: a conversational analytics agent across 10 OCEAN personas (240 evaluator judgments); a customer-support agent across 5 tasks $times$ 4 persona variants; and adversarial stress-testing of 5 tasks revealing pre-existing trajectory brittleness.
  arXiv  Detail & Related papers  (2026-08-25T05:15:33Z)
• STAIN-FL: Stealthy Targeted Attack Injection with Contextual Triggers in Federated Learning [4.717908620608989]
  This paper introduces STAIN-FL, a stealthy targeted backdoor attack injection framework.<n>It uses naturally occurring surveillance conditions, including low-light scenes, indoor settings, and crowd density, as contextual triggers.<n>We evaluate STAIN-FL on textttUCF-Crime using 1024-dimensional I3D features in a non-IID four-client multi-agency setting.
  arXiv  Detail & Related papers  (2026-08-25T01:24:16Z)
• Test-time reasoning effort and unauthorized tool use in language-model agents: a prespecified equivalence study [1.6787220460106658]
  Language-model agents that execute through tool calls operate under access-control policies that restrict which operations each role may perform.<n>We vary reasoning effort (low, max) inside GPT-5.6 across 14 confirmatory scenarios of TRIO-20.<n>Across 840 trajectories and two model tiers, no unauthorized tool call occurred.
  arXiv  Detail & Related papers  (2026-08-04T06:01:14Z)
• Safety, or Just Capability? A Validity Audit of Agent-Safety Benchmarks [10.839956062772792]
  Three agent-safety benchmarks rank the same 18 models differently.<n>R-Judge specificity against AgentHarm safety correlates $-0.64$ at $n=7$ and $+0.02$ at $n=18$, and a quarter of random size-7 subsets reach $|| geq 0.5$ around that near-zero value.<n>AgentHarm shows the strongest held-out association, $=+0.72$ with three-template jailbreak safety after controlling capability.
  arXiv  Detail & Related papers  (2026-07-30T03:45:10Z)
• AgentSentry: Mitigating Indirect Prompt Injection in LLM Agents via Temporal Causal Diagnostics and Context Purification [25.817251923574286]
  We propose a novel inference-time detection and mitigation framework for large language model (LLM) agents.<n>AgentSentry is the first inference-time defense to model multi-turn IPI as a temporal causal takeover.<n>We evaluate AgentSentry on the textscAgentDojo benchmark across four task suites, three IPI attack families, and multiple black-box LLMs.
  arXiv  Detail & Related papers  (2026-02-26T07:59:10Z)
• StepShield: When, Not Whether to Intervene on Rogue Agents [1.472404880217315]
  Existing agent safety benchmarks report binary accuracy, conflating early intervention with post-mortem analysis.<n>We introduce StepShield, the first benchmark to evaluate when violations are detected, not just whether.<n>By shifting the focus of evaluation from whether to when, StepShield provides a new foundation for building safer and more economically viable AI agents.
  arXiv  Detail & Related papers  (2026-01-29T18:55:46Z)
• $α^3$-SecBench: A Large-Scale Evaluation Suite of Security, Resilience, and Trust for LLM-based UAV Agents over 6G Networks [3.099103925863002]
  We introduce $3$-SecBench, the first large-scale evaluation suite for assessing the security-aware autonomy of LLM-based UAV agents under realistic adversarial interference.<n>We evaluate 23 state-of-the-art LLMs from major industrial providers and leading AI labs using thousands of adversarially augmented UAV episodes sampled from a corpus of 113,475 missions spanning 175 threat types. Normalized overall scores range from 12.9% to 57.1%, highlighting a significant gap between anomaly detection and security-aware autonomous decision-making.
  arXiv  Detail & Related papers  (2026-01-26T18:25:07Z)
• BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents [58.83028403414688]
  Large language model (LLM) agents execute tasks through multi-step workflow that combine planning, memory, and tool use.<n>Backdoor triggers injected into specific stages of an agent workflow can persist through multiple intermediate states and adversely influence downstream outputs.<n>We propose textbfBackdoorAgent, a modular and stage-aware framework that provides a unified agent-centric view of backdoor threats in LLM agents.
  arXiv  Detail & Related papers  (2026-01-08T03:49:39Z)
• ReliabilityBench: Evaluating LLM Agent Reliability Under Production-Like Stress Conditions [0.32928123659012326]
  Existing benchmarks for tool-using LLM agents primarily report single-run success rates and miss reliability properties required in production.<n>We introduce textbfReliabilityBench, a benchmark for evaluating agent reliability across three dimensions.<n>We evaluate two models (Gemini 2.0 Flash, GPT-4o) and two agent architectures (ReAct, Reflexion) across four domains (scheduling, travel, customer support, e-commerce) over 1,280 episodes.
  arXiv  Detail & Related papers  (2026-01-03T13:41:33Z)
• Securing the Model Context Protocol: Defending LLMs Against Tool Poisoning and Adversarial Attacks [8.419049623790618]
  This work analyzes three classes of semantic attacks on MCP-integrated systems.<n>We introduce a layered security framework with three components: RSA-based manifest signing to enforce descriptor integrity, LLM-on-LLM semantic vetting to detect suspicious tool definitions, and lightweight guardrails that block anomalous tool behavior at runtime.<n>Our results show that the proposed framework reduces unsafe tool invocation rates without model fine-tuning or internal modification.
  arXiv  Detail & Related papers  (2025-12-06T20:07:58Z)
• LLM-Powered Detection of Price Manipulation in DeFi [12.59175486585742]
  Decentralized Finance (DeFi) smart contracts manage billions of dollars, making them a prime target for exploits.<n>Price manipulation vulnerabilities, often via flash loans, are a devastating class of attacks causing significant financial losses.<n>We propose PMDetector, a hybrid framework combining static analysis with Large Language Model (LLM)-based reasoning.
  arXiv  Detail & Related papers  (2025-10-24T09:13:30Z)
• Backdoor Collapse: Eliminating Unknown Threats via Known Backdoor Aggregation in Language Models [75.29749026964154]
  Ourmethod reduces the average Attack Success Rate to 4.41% across multiple benchmarks.<n>Clean accuracy and utility are preserved within 0.5% of the original model.<n>The defense generalizes across different types of backdoors, confirming its robustness in practical deployment scenarios.
  arXiv  Detail & Related papers  (2025-10-11T15:47:35Z)
• OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety [58.201189860217724]
  We introduce OpenAgentSafety, a comprehensive framework for evaluating agent behavior across eight critical risk categories.<n>Unlike prior work, our framework evaluates agents that interact with real tools, including web browsers, code execution environments, file systems, bash shells, and messaging platforms.<n>It combines rule-based analysis with LLM-as-judge assessments to detect both overt and subtle unsafe behaviors.
  arXiv  Detail & Related papers  (2025-07-08T16:18:54Z)
• BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems [62.17474934536671]
  We introduce the first framework to capture offensive and defensive cyber-capabilities in evolving real-world systems.<n>To capture the vulnerability lifecycle, we define three task types: Detect (detecting a new vulnerability), Exploit (exploiting a specific vulnerability), and Patch (patching a specific vulnerability)<n>We evaluate 8 agents: Claude Code, OpenAI Codex CLI with o3-high and o4-mini, and custom agents with o3-high, GPT-4.1, Gemini 2.5 Pro Preview, Claude 3.7 Sonnet Thinking, and DeepSeek-R1.
  arXiv  Detail & Related papers  (2025-05-21T07:44:52Z)
• $\ extit{Agents Under Siege}$: Breaking Pragmatic Multi-Agent LLM Systems with Optimized Prompt Attacks [45.74758377276353]
  Multi-agent Large Language Model (LLM) systems create novel adversarial risks because their behavior depends on communication between agents and decentralized reasoning.<n>In this work, we innovatively focus on attacking pragmatic systems that have constrains such as limited token bandwidth, latency between message delivery, and defense mechanisms.<n>We design a $textitpermutation-invariant adversarial attack$ that optimize prompt distribution across latency and bandwidth-constraint network topologies to bypass distributed safety mechanisms.
  arXiv  Detail & Related papers  (2025-03-31T20:43:56Z)
• SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM Agents [58.65256663334316]
  We present SafeAgentBench -- the first benchmark for safety-aware task planning of embodied LLM agents in interactive simulation environments.<n>SafeAgentBench includes: (1) an executable, diverse, and high-quality dataset of 750 tasks, rigorously curated to cover 10 potential hazards and 3 task types; (2) SafeAgentEnv, a universal embodied environment with a low-level controller, supporting multi-agent execution with 17 high-level actions for 9 state-of-the-art baselines; and (3) reliable evaluation methods from both execution and semantic perspectives.
  arXiv  Detail & Related papers  (2024-12-17T18:55:58Z)
• On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents [58.79302663733703]
  Large language model-based multi-agent systems have shown great abilities across various tasks due to the collaboration of expert agents.<n>The impact of clumsy or even malicious agents--those who frequently make errors in their tasks--on the overall performance of the system remains underexplored.<n>This paper investigates what is the resilience of various system structures under faulty agents on different downstream tasks.
  arXiv  Detail & Related papers  (2024-08-02T03:25:20Z)
• Dissecting Adversarial Robustness of Multimodal LM Agents [70.2077308846307]
  We manually create 200 targeted adversarial tasks and evaluation scripts in a realistic threat model on top of VisualWebArena.<n>We find that we can successfully break latest agents that use black-box frontier LMs, including those that perform reflection and tree search.<n>We also use ARE to rigorously evaluate how the robustness changes as new components are added.
  arXiv  Detail & Related papers  (2024-06-18T17:32:48Z)
• Malicious Agent Detection for Robust Multi-Agent Collaborative Perception [52.261231738242266]
  Multi-agent collaborative (MAC) perception is more vulnerable to adversarial attacks than single-agent perception. We propose Malicious Agent Detection (MADE), a reactive defense specific to MAC perception. We conduct comprehensive evaluations on a benchmark 3D dataset V2X-sim and a real-road dataset DAIR-V2X.
  arXiv  Detail & Related papers  (2023-10-18T11:36:42Z)

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.