AI Agent Security

Ask about this topic

Overview

AI agent security studies how tool use, memory, delegation, and interaction with people or environments create risks beyond those of a standalone language model, including prompt injection, privilege misuse, data leakage, and unsafe actions. Researchers test these risks with red-teaming and execution-grounded benchmarks for coding, browsing, GUI, embodied, and multi-agent tasks, measuring whether harmful effects occur across trajectories rather than judging text alone. Defenses range from input and memory provenance checks to sandboxing, least-privilege permissions, stateful runtime authorization, action verification, and cryptographic identity or audit mechanisms. An open challenge is making controls robust to adaptive, long-horizon attacks and interactions between tools or agents while preserving useful task completion; benchmark realism and comparable measures remain uneven.

Latest papers

CardsList

517 papers74 papers in the last four weeks, up 32% on the four weeks before. 0.4% of all new papers.

Apr 9, 2026

AITH: A Post-Quantum Continuous Delegation Protocol for Human-AI Trust Establishment

The rapid deployment of AI agents acting autonomously on behalf of human principals has outpaced the development of cryptographic protocols for establishing, bounding, and revoking human-AI trust relationships. Existing frameworks (TLS, OAuth 2.0, Macaroons) assume deterministic software and cannot address probabilistic AI agents operating continuously within variable trust boundaries. We present AITH (AI Trust Handshake), a post-quantum continuous delegation protocol. AITH introduces: (1) a Continuous Delegation Certificate signed once with ML-DSA-87 (FIPS 204, NIST Level 5), replacing per-operation signing with sub-microsecond boundary checks at 4.7M ops/sec; (2) a six-check Boundary Engine enforcing hard constraints, rate limits, and escalation triggers with zero cryptographic overhead on the critical path; (3) a push-based Revocation Protocol propagating invalidation within one second. A three-tier SHA-256 Responsibility Chain provides tamper-evident audit logging. All five security theorems are machine-verified via Tamarin Prover under the Dolev-Yao model. We validate AITH through five rounds of multi-model adversarial auditing, resolving 12 vulnerabilities across four severity layers. Simulation of 100,000 operations shows 79.5% autonomous execution, 6.1% human escalation, and 14.4% blocked.

Zhaoliang Chen

Faculty of Business Administration, University of Macau, Macao SAR, China

AI Agent SecurityCybersecurityTrust in AIRuntime Enforcement for AI Agents
AskPDF
Apr 8, 2026

MCP-DPT: A Defense-Placement Taxonomy and Coverage Analysis for Model Context Protocol Security

The Model Context Protocol (MCP) enables large language models (LLMs) to dynamically discover and invoke third-party tools, significantly expanding agent capabilities while introducing a distinct security landscape. Unlike prompt-only interactions, MCP exposes pre-execution artifacts, shared context, multi-turn workflows, and third-party supply chains to adversarial influence across independently operated components. While recent work has identified MCP-specific attacks and evaluated defenses, existing studies are largely attack-centric or benchmark-driven, providing limited guidance on where mitigation responsibility should reside within the MCP architecture. This is problematic given MCP's multi-party design and distributed trust boundaries. We present a defense-placement-oriented security analysis of MCP, introducing a layer-aligned taxonomy that organizes attacks by the architectural component responsible for enforcement. Threats are mapped across six MCP layers, and primary and secondary defense points are identified to support principled defense-in-depth reasoning under adversaries controlling tools, servers, or ecosystem components. A structured mapping of existing academic and industry defenses onto this framework reveals uneven and predominantly tool-centric protection, with persistent gaps at the host orchestration, transport, and supply-chain layers. These findings suggest that many MCP security weaknesses stem from architectural misalignment rather than isolated implementation flaws.

Mehrdad Rostamzadeh, Sidhant Narula, Nahom Birhan +2

Old Dominion University, Norfolk, Virginia, USA · Old Dominion Univer, Norfolk, Virginia, USA

MCP SecurityAI Agent Security
AskPDF
Apr 8, 2026

SkillSieve: A Hierarchical Triage Framework for Detecting Malicious AI Agent Skills

Agent skills combine natural-language instructions with executable code while inheriting an agent's filesystem, credential, and network access. Attacks can span prose and files, whereas regex and code-only analyzers cover only one modality. SkillSieve applies three progressively deeper layers: recall-oriented regex, AST, and metadata triage; four parallel LLM security sub-tasks; and an independent three-model jury with debate on disagreement. We evaluate 49,592 real ClawHub skills, a 390-skill labeled benchmark, and 100 adversarial samples across five evasion techniques on a 440 USD ARM board. The full pipeline achieves F1 = 0.929 (precision 0.912, recall 0.945) at an average cost of $0.006 per skill. An optional XGBoost fast path reduces Layer-2/3 calls by 32% with a 1.7-point F1 decrease while preserving recall. On 52 Feishu/Lark packages, Layer 2 reclassifies 13 of 14 Layer-1 flags as safe after contextual analysis; we also deploy the system as a Feishu chat bot. Code, labels, and aggregate results are open-sourced.

Yinghan Hou, Zongyou Yang

Department of Electrical and Electronic Engineering, Imperial College London, London, United Kingdom · Dyson School of Design Engineering, Imperial College London, London, United Kingdom

AI Agent SecurityAgent Skill Security AuditingAI Agent Security Benchmarks
AskPDF
Apr 7, 2026

Who Governs the Machine? A Machine Identity Governance Taxonomy (MIGT) for AI Systems Operating Across Enterprise and Geopolitical Boundaries

The governance of artificial intelligence has a blind spot: the machine identities that AI systems use to act. AI agents, service accounts, API tokens, and automated workflows now outnumber human identities in enterprise environments by ratios exceeding 80 to 1, yet no integrated framework exists to govern them. A single ungoverned automated agent produced $5.4-10 billion in losses in the 2024 CrowdStrike outage; nation-state actors including Silk Typhoon and Salt Typhoon have operationalized ungoverned machine credentials as primary espionage vectors against critical infrastructure. This paper makes four original contributions. First, the AI-Identity Risk Taxonomy (AIRT): a comprehensive enumeration of 37 risk sub-categories across eight domains, each grounded in documented incidents, regulatory recognition, practitioner prevalence data, and threat intelligence. Second, the Machine Identity Governance Taxonomy (MIGT): an integrated six-domain governance framework simultaneously addressing the technical governance gap, the regulatory compliance gap, and the cross-jurisdictional coordination gap that existing frameworks address only in isolation. Third, a foreign state actor threat model for enterprise identity governance, establishing that Silk Typhoon, Salt Typhoon, Volt Typhoon, and North Korean AI-enhanced identity fraud operations have already operationalized AI identity vulnerabilities as active attack vectors. Fourth, a cross-jurisdictional regulatory alignment structure mapping enterprise AI identity governance obligations under EU, US, and Chinese frameworks simultaneously, identifying irreconcilable conflicts and providing a governance mechanism for managing them. A four-phase implementation roadmap translates the MIGT into actionable enterprise programs.

Andrew Kurtz, Klaudia Krawiecka

AI GovernanceCybersecurityAI Agent SecurityAI Risk Management
AskPDF
Apr 7, 2026

A Formal Security Framework for MCP-Based AI Agents: Threat Taxonomy, Verification Models, and Defense Mechanisms

The Model Context Protocol (MCP), introduced by Anthropic in November 2024 and now governed by the Linux Foundation's Agentic AI Foundation, has rapidly become the de facto standard for connecting large language model (LLM)-based agents to external tools and data sources, with over 97 million monthly SDK downloads and more than 177000 registered tools. However, this explosive adoption has exposed a critical gap: the absence of a unified, formal security framework capable of systematically characterizing, analyzing, and mitigating the diverse threats facing MCP-based agent ecosystems. Existing security research remains fragmented across individual attack papers, isolated benchmarks, and point defense mechanisms. This paper presents MCPSHIELD, a comprehensive formal security framework for MCP-based AI agents. We make four principal contributions: (1) a hierarchical threat taxonomy comprising 7 threat categories and 23 distinct attack vectors organized across four attack surfaces, grounded in the analysis of over 177000 MCP tools; (2) a formal verification model based on labeled transition systems with trust boundary annotations that enables static and runtime analysis of MCP tool interaction chains; (3) a systematic comparative evaluation of 12 existing defense mechanisms, identifying coverage gaps across our threat taxonomy; and (4) a defense in depth reference architecture integrating capability based access control, cryptographic tool attestation, information flow tracking, and runtime policy enforcement. Our analysis reveals that no existing single defense covers more than 34 percent of the identified threat landscape, whereas MCPSHIELD's integrated architecture achieves theoretical coverage of 91 percent. We further identify seven open research challenges that must be addressed to secure the next generation of agentic AI systems.

Nirajan Acharya, Gaurav Kumar Gupta

University of the Cumberlands · Youngstown State University

MCP SecurityAI Agent SecurityTool Access Control for LLM AgentsFormal Verification
AskPDF
Apr 7, 2026

ClawLess: A Security Model of AI Agents

Autonomous AI agents powered by Large Language Models can reason, plan, and execute complex tasks, but their ability to autonomously retrieve information and run code introduces significant security risks. Existing approaches attempt to regulate agent behavior through training or prompting, which does not offer fundamental security guarantees. We present ClawLess, a security framework that enforces formally verified policies on AI agents under a worst-case threat model where the agent itself may be adversarial. ClawLess formalizes a fine-grained security model over system entities, trust scopes, and permissions to express dynamic policies that adapt to agents' runtime behavior. These policies are translated into concrete security rules and enforced through a user-space kernel augmented with BPF-based syscall interception. This approach bridges the formal security model with practical enforcement, ensuring security regardless of the agent's internal design.

Hongyi Lu, Nian Liu, Shuai Wang +1

Research Institute of Trustworthy Autonomous Systems, Southern University of Science and Technology, China · Department of Computer Science and Engineering, Southern University of Science and Technology · Department of Computer Science and Engineering, Hong Kong University of Science and Technology +3

AI Agent SecurityRuntime Enforcement for AI AgentsTool Access Control for LLM Agents
AskPDF
Apr 7, 2026

Auditable Agents

LLM agents call tools, query databases, delegate tasks, and trigger external side effects. Once an agent system can act in the world, the question is no longer only whether harmful actions can be prevented--it is whether those actions remain answerable after deployment. We distinguish accountability (the ability to determine compliance and assign responsibility), auditability (the system property that makes accountability possible), and auditing (the process of reconstructing behavior from trustworthy evidence). Our claim is direct: no agent system can be accountable without auditability. To make this operational, we define five dimensions of agent auditability, i.e., action recoverability, lifecycle coverage, policy checkability, responsibility attribution, and evidence integrity, and identify three mechanism classes (detect, enforce, recover) whose temporal information-and-intervention constraints explain why, in practice, no single approach suffices. We support the position with layered evidence rather than a single benchmark: lower-bound ecosystem measurements suggest that even basic security prerequisites for auditability are widely unmet (617 security findings across six prominent open-source projects); runtime feasibility results show that pre-execution mediation with tamper-evident records adds only 8.3 ms median overhead; and controlled recovery experiments show that responsibility-relevant information can be partially recovered even when conventional logs are missing. We propose an Auditability Card for agent systems and identify six open research problems organized by mechanism class.

Yi Nian, Aojie Yuan, Haiyue Zhang +7

University of Southern California · Arizona State University · University of Tennessee, Knoxville +1

AI AccountabilityAI Agent AuditingLLM AuditingAI Agent Security
AskPDF
Apr 7, 2026

LanG -- A Governance-Aware Agentic AI Platform for Unified Security Operations

Modern Security Operations Centers struggle with alert fatigue, fragmented tooling, and limited cross-source event correlation. Challenges that current Security Information Event Management and Extended Detection and Response systems only partially address through fragmented tools. This paper presents the LLM-assisted network Governance (LanG), an open-source, governance-aware agentic AI platform for unified security operations contributing: (i) a Unified Incident Context Record with a correlation engine (F1 = 87%), (ii) an Agentic AI Orchestrator on LangGraph with human-in-the-loop checkpoints, (iii) an LLM-based Rule Generator finetuned on four base models producing deployable Snort 2/3, Suricata, and YARA rules (average acceptance rate 96.2%), (iv) a Three-Phase Attack Reconstructor combining Louvain community detection, LLM-driven hypothesis generation, and Bayesian scoring (87.5% kill-chain accuracy), and (v) a layered Governance-MCP-Agentic AI-Security architecture where all tools are exposed via the Model Context Protocol, governed by an AI Governance Policy Engine with a two-layer guardrail pipeline (regex + Llama Prompt Guard 2 semantic classifier, achieving 98.1% F1 score with experimental zero false positives). Designed for Managed Security Service Providers, the platform supports multi-tenant isolation, role-based access, and fully local deployment. Finetuned anomaly and threat detectors achieve weighted F1 scores of 99.0% and 91.0%, respectively, in intrusion-detection benchmarks, running inferences in ≈\approx21 ms with a machine-side mean time to detect of 1.58 s, and the rule generator exceeds 91% deployability on live IDS engines. A systematic comparison against eight SOC platforms confirms that LanG uniquely satisfies multiple industrial capabilities all in one open-source tool, while enforcing selected AI governance policies.

Anes Abdennebi, Nadjia Kara, Laaziz Lahlou +1

Department of Software Engineering and IT, École de Technologie Supérieure (ÉTS), Montreal, Canada · Department of Mathematics & Computer Science, UQTR, Trois-Rivières, Canada

CybersecurityLLMs for CybersecurityAI Agent SecurityAI Agent GovernanceNetwork Intrusion Detection
AskPDF
Apr 6, 2026

Your Agent, Their Asset: A Real-World Safety Analysis of OpenClaw

OpenClaw, the most widely deployed personal AI agent in early 2026, operates with full local system access and integrates with sensitive services such as Gmail, Stripe, and the filesystem. While these broad privileges enable high levels of automation and powerful personalization, they also expose a substantial attack surface that existing sandboxed evaluations fail to capture. To address this gap, we present the first real-world safety evaluation of OpenClaw and introduce the CIK taxonomy, which unifies an agent's persistent state into three dimensions, i.e., Capability, Identity, and Knowledge, for safety analysis. Our evaluations cover 12 attack scenarios on a live OpenClaw instance across four backbone models (Claude Sonnet 4.5, Opus 4.6, Gemini 3.1 Pro, and GPT-5.4). The results show that poisoning any single CIK dimension increases the average attack success rate from 24.6% to 64-74%, with even the most robust model exhibiting more than a threefold increase over its baseline vulnerability. We further assess three CIK-aligned defense strategies alongside a file-protection mechanism; however, the strongest defense still yields a 63.8% success rate under Capability-targeted attacks, while file protection blocks 97% of malicious injections but also prevents legitimate updates. Taken together, these findings show that the vulnerabilities are inherent to the agent architecture, necessitating more systematic safeguards to secure personal AI agents. Our project page is https://ucsc-vlaa.github.io/CIK-Bench.

Zijun Wang, Haoqin Tu, Letian Zhang +11

UC Santa Cruz · NUS · Tencent +3

AI Agent SecurityAI Agent SafetyAI Agent Safety BenchmarksAgent Memory Poisoning
AskPDF
Apr 6, 2026

Undetectable Conversations Between AI Agents via Pseudorandom Noise-Resilient Key Exchange

AI agents are increasingly deployed to interact with other agents on behalf of users and organizations. We ask whether two such agents, operated by different entities, can carry out a parallel secret conversation while still producing a transcript that is computationally indistinguishable from an honest interaction, even to a strong passive auditor that knows the full model descriptions, the protocol, and the agents' private contexts. Building on recent work on watermarking and steganography for LLMs, we first show that if the parties possess an interaction-unique secret key, they can facilitate an optimal-rate covert conversation: the hidden conversation can exploit essentially all of the entropy present in the honest message distributions. Our main contributions concern extending this to the keyless setting, where the agents begin with no shared secret. We show that covert key exchange, and hence covert conversation, is possible even when each model has an arbitrary private context, and their messages are short and fully adaptive, assuming only that sufficiently many individual messages have at least constant min-entropy. This stands in contrast to previous covert communication works, which relied on the min-entropy in each individual message growing with the security parameter. To obtain this, we introduce a new cryptographic primitive, which we call pseudorandom noise-resilient key exchange: a key-exchange protocol whose public transcript is pseudorandom while still remaining correct under constant noise. We study this primitive, giving several constructions relevant to our application as well as strong limitations showing that more naive variants are impossible or vulnerable to efficient attacks. These results show that transcript auditing alone cannot rule out covert coordination between AI agents, and identify a new cryptographic theory that may be of independent interest.

Vinod Vaikuntanathan, Or Zamir

MIT · Tel Aviv University

Language Model SteganographyAI Agent SecurityMulti-Agent CommunicationAI Agent Auditing
AskPDF
Apr 6, 2026

ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems

Existing research on LLM agent security mainly focuses on prompt injection and unsafe input/output behaviors. However, as agents increasingly rely on third-party tools and MCP servers, a new class of supply-chain threats has emerged, where malicious behaviors are embedded in seemingly benign tools, silently hijacking agent execution, leaking sensitive data, or triggering unauthorized actions. Despite their growing impact, there is currently no comprehensive benchmark for evaluating such threats. To bridge this gap, we introduce SC-Inject-Bench, a large-scale benchmark comprising over 10,000 malicious MCP tools grounded in a taxonomy of 25+ attack types derived from MITRE ATT&CK targeting supply-chain threats. We observe that existing MCP scanners and semantic guardrails perform poorly on this benchmark. Motivated by this finding, we propose ShieldNet, a network-level guardrail framework that detects supply-chain poisoning by observing real network interactions rather than surface-level tool traces. ShieldNet integrates a man-in-the-middle (MITM) proxy and an event extractor to identify critical network behaviors, which are then processed by a lightweight classifier for attack detection. Extensive experiments show that ShieldNet achieves strong detection performance (up to 0.995 F-1 with only 0.8% false positives) while introducing little runtime overhead, substantially outperforming existing MCP scanners and LLM-based guardrails.

Zhuowen Yuan, Zhaorun Chen, Zhen Xiang +5

UIUC · University of Chicago · University of Georgia +5

AI Agent SecuritySoftware Supply Chain SecurityAI Agent Security Benchmarks
AskPDF
Apr 5, 2026

Causality Laundering: Denial-Feedback Leakage in Tool-Calling LLM Agents

Tool-calling LLM agents can read private data, invoke external services, and trigger real-world actions, creating a security problem at the point of tool execution. We identify a denial-feedback leakage pattern, which we term causality laundering, in which an adversary probes a protected action, learns from the denial outcome, and exfiltrates the inferred information through a later seemingly benign tool call. This attack is not captured by flat provenance tracking alone because the leaked information arises from causal influence of the denied action, not direct data flow. We present the Agentic Reference Monitor (ARM), a runtime enforcement layer that mediates every tool invocation by consulting a provenance graph over tool calls, returned data, field-level provenance, and denied actions. ARM propagates trust through an integrity lattice and augments the graph with counterfactual edges from denied-action nodes, enabling enforcement over both transitive data dependencies and denial-induced causal influence. In a controlled evaluation on three representative attack scenarios, ARM blocks causality laundering, transitive taint propagation, and mixed-provenance field misuse that a flat provenance baseline misses, while adding sub-millisecond policy evaluation overhead. These results suggest that denial-aware causal provenance is a useful abstraction for securing tool-calling agent systems.

Mohammad Hossein Chinaei

Independent Researcher

AI Agent SecurityRuntime Enforcement for AI AgentsTool Access Control for LLM Agents
AskPDF
Apr 4, 2026

Your Agent is More Brittle Than You Think: Uncovering Indirect Injection Vulnerabilities in Agentic LLMs

The rapid deployment of open-source frameworks has significantly advanced the development of modern multi-agent systems. However, expanded action spaces, including uncontrolled privilege exposure and hidden inter-system interactions, pose severe security challenges. Specifically, Indirect Prompt Injections (IPI), which conceal malicious instructions within third-party content, can trigger unauthorized actions such as data exfiltration during normal operations. While current security evaluations predominantly rely on isolated single-turn benchmarks, the systemic vulnerabilities of these agents within complex dynamic environments remain critically underexplored. To bridge this gap, we systematically evaluate six defense strategies against four sophisticated IPI attack vectors across nine LLM backbones. Crucially, we conduct our evaluation entirely within dynamic multi-step tool-calling environments to capture the true attack surface of modern autonomous agents. Moving beyond binary success rates, our multidimensional analysis reveals a pronounced fragility. Advanced injections successfully bypass nearly all baseline defenses, and some surface-level mitigations even produce counterproductive side effects. Furthermore, while agents execute malicious instructions almost instantaneously, their internal states exhibit abnormally high decision entropy. Motivated by this latent hesitation, we investigate Representation Engineering (RepE) as a robust detection strategy. By extracting hidden states at the tool-input position, we revealed that the RepE-based circuit breaker successfully identifies and intercepts unauthorized actions before the agent commits to them, achieving high detection accuracy across diverse LLM backbones. This study exposes the limitations of current IPI defenses and provides a highly practical paradigm for building resilient multi-agent architectures.

Wenhui Zhu, Xuanzhao Dong, Xiwen Chen +7

Arizona State University · Morgan Stanley · UC Davis +4

AI Agent SecurityIndirect Prompt InjectionPrompt Injection DefenseAI Agent Security Benchmarks
AskPDF
Apr 4, 2026

Measuring the Permission Gate: A Stress-Test Evaluation of Claude Code's Auto Mode

Claude Code's auto mode is the first deployed permission system for AI coding agents, using a two-stage transcript classifier to gate dangerous tool calls. Anthropic reports a 0.4% false positive rate and 17% false negative rate on production traffic. We present the first independent evaluation of this system on deliberately ambiguous authorization scenarios, i.e., tasks where the user's intent is clear but the target scope, blast radius, or risk level is underspecified. Using AmPermBench, a 128-prompt benchmark spanning four DevOps task families and three controlled ambiguity axes, we evaluate 253 state-changing actions at the individual action level against oracle ground truth. Our findings characterize auto mode's scope-escalation coverage under this stress-test workload. The end-to-end false negative rate is 81.0% (95% CI: 73.8%-87.4%), substantially higher than the 17% reported on production traffic, reflecting a fundamentally different workload rather than a contradiction. Notably, 36.8% of all state-changing actions fall outside the classifier's scope via Tier 2 (in-project file edits), contributing to the elevated end-to-end FNR. Even restricting to the 160 actions the classifier actually evaluates (Tier 3), the FNR remains 70.3%, while the FPR rises to 31.9%. The Tier 2 coverage gap is most pronounced on artifact cleanup (92.9% FNR), where agents naturally fall back to editing state files when the expected CLI is unavailable. These results highlight a coverage boundary worth examining: auto mode assumes dangerous actions transit the shell, but agents routinely achieve equivalent effects through file edits that the classifier does not evaluate.

Zimo Ji, Zongjie Li, Pingchuan Ma +3

Hong Kong University of Science and Technology · ETH Zurich

AI Agent SecurityTool-Use AuthorizationTool Access Control for LLM AgentsAI Coding AgentsAI Agent Security Benchmarks
AskPDF
Apr 3, 2026

A Systematic Security Evaluation of OpenClaw and Its Variants

Tool-augmented AI agents substantially extend the practical capabilities of large language models, but they also introduce security risks that cannot be identified through model-only evaluation. In this paper, we present a systematic security assessment of six representative OpenClaw-series agent frameworks, namely OpenClaw, AutoClaw, QClaw, KimiClaw, MaxClaw, and ArkClaw, under multiple backbone models. To support this study, we construct a benchmark of 205 test cases covering representative attack behaviors across the full agent execution lifecycle, enabling unified evaluation of risk exposure at both the framework and model levels. Our results show that all evaluated agents exhibit substantial security vulnerabilities, and that agentized systems are significantly riskier than their underlying models used in isolation. In particular, reconnaissance and discovery behaviors emerge as the most common weaknesses, while different frameworks expose distinct high-risk profiles, including credential leakage, lateral movement, privilege escalation, and resource development. These findings indicate that the security of modern agent systems is shaped not only by the safety properties of the backbone model, but also by the coupling among model capability, tool use, multi-step planning, and runtime orchestration. We further show that once an agent is granted execution capability and persistent runtime context, weaknesses arising in early stages can be amplified into concrete system-level failures. Overall, our study highlights the need to move beyond prompt-level safeguards toward lifecycle-wide security governance for intelligent agent frameworks.

Yuhang Wang, Haichang Gao, Zhenxing Niu +4

Xidian University · Data Science & Artificial Intelligence Research Institute, China Unicom

AI Agent SecurityLLM Agent SecurityAI Agent Security Benchmarks
AskPDF
Apr 3, 2026

Supply-Chain Poisoning Attacks Against LLM Coding Agent Skill Ecosystems

LLM-based coding agents extend their capabilities via third-party agent skills distributed through open marketplaces without mandatory security review. Unlike traditional packages, these skills are executed as operational directives with system-level privileges, so a single malicious skill can compromise the host. Prior work has not examined whether supply-chain attacks can directly hijack an agent's action space, such as file writes, shell commands, and network requests, despite existing safeguards. We introduce Document-Driven Implicit Payload Execution (DDIPE), which embeds malicious logic in code examples and configuration templates within skill documentation. Because agents reuse these examples during normal tasks, the payload executes without explicit prompts. Using an LLM-driven pipeline, we generate 1,070 adversarial skills from 81 seeds across 15 MITRE ATTACK categories. Across four frameworks and five models, DDIPE achieves 11.6% to 33.5% bypass rates, while explicit instruction attacks achieve 0% under strong defenses. Static analysis detects most cases, but 2.5% evade both detection and alignment. Responsible disclosure led to four confirmed vulnerabilities and two fixes.

Yubin Qu, Yi Liu, Tongcheng Geng +5

Griffith University Australia · Quantstamp Singapore · The State Information Center China +4

AI Agent SecurityLLM Agent SecurityAgent Skill Security Auditing
AskPDF
Apr 3, 2026

Poison Once, Exploit Forever: Environment-Injected Memory Poisoning Attacks on Web Agents

Memory makes LLM-based web agents personalized, powerful, yet exploitable. By storing past interactions to personalize future tasks, agents inadvertently create a persistent attack surface that spans websites and sessions. While existing security research on memory assumes attackers can directly inject into memory storage or exploit shared memory across users, we present a more realistic threat model: contamination through environmental observation alone. We introduce Environment-injected Trajectory-based Agent Memory Poisoning (eTAMP), the first attack to achieve cross-session, cross-site compromise without requiring direct memory access. A single contaminated observation (e.g., viewing a manipulated product page) silently poisons an agent's memory and activates during future tasks on different websites, bypassing permission-based defenses. Our experiments on (Visual)WebArena reveal two key findings. First, eTAMP achieves substantial attack success rates: up to 32.5% on GPT-5-mini, 23.4% on GPT-5.2, and 19.5% on GPT-OSS-120B. Second, we discover Frustration Exploitation: agents under environmental stress become dramatically more susceptible, with ASR increasing up to 8 times when agents struggle with dropped clicks or garbled text. Notably, more capable models are not more secure. GPT-5.2 shows substantial vulnerability despite superior task performance. With the rise of AI browsers like OpenClaw, ChatGPT Atlas, and Perplexity Comet, our findings underscore the urgent need for defenses against environment-injected memory poisoning.

Wei Zou, Mingwen Dong, Miguel Romero Calvo +7

Pennsylvania State University · Amazon Web Services

AI Agent SecurityAgent Memory PoisoningWeb Agents
AskPDF
Apr 2, 2026

I must delete the evidence: AI Agents Explicitly Cover up Fraud and Violent Crime

As ongoing research explores the ability of AI agents to be insider threats and act against company interests, we showcase the abilities of such agents to act against human well being in service of corporate authority. Building on Agentic Misalignment and AI scheming research, we present a scenario where the majority of evaluated state-of-the-art AI agents explicitly choose to suppress evidence of fraud and harm, in service of company profit. We test this scenario on 16 recent Large Language Models. Some models show remarkable resistance to our method and behave appropriately, but many do not, and instead aid and abet criminal activity. These experiments are simulations and were executed in a controlled virtual environment. No crime actually occurred.

Thomas Rivasseau

Data Mining and Security Lab, School of Information Studies, McGill University.

AI Agent SafetyAI EthicsAI Agent SecurityAI AlignmentAI Safety Evaluation
AskPDF
Apr 2, 2026

APEX: Agent Payment Execution with Policy for Autonomous Agent API Access

Autonomous agents are moving beyond simple retrieval tasks to become economic actors that invoke APIs, sequence workflows, and make real-time decisions. As this shift accelerates, API providers need request-level monetization with programmatic spend governance. The HTTP 402 protocol addresses this by treating payment as a first-class protocol event, but most implementations rely on cryptocurrency rails. In many deployment contexts, especially countries with strong real-time fiat systems like UPI, this assumption is misaligned with regulatory and infrastructure realities. We present APEX, an implementation-complete research system that adapts HTTP 402-style payment gating to UPI-like fiat workflows while preserving policy-governed spend control, tokenized access verification, and replay resistance. We implement a challenge-settle-consume lifecycle with HMAC-signed short-lived tokens, idempotent settlement handling, and policy-aware payment approval. The system uses FastAPI, SQLite, and Python standard libraries, making it transparent, inspectable, and reproducible. We evaluate APEX across three baselines and six scenarios using sample sizes 2-4x larger than initial experiments (N=20-40 per scenario). Results show that policy enforcement reduces total spending by 27.3% while maintaining 52.8% success rate for legitimate requests. Security mechanisms achieve 100% block rate for both replay attacks and invalid tokens with low latency overhead (19.6ms average). Multiple trial runs show low variance across scenarios, demonstrating high reproducibility with 95% confidence intervals. The primary contribution is a controlled agent-payment infrastructure and reference architecture that demonstrates how agentic access monetization can be adapted to fiat systems without discarding security and policy guarantees.

Mohd Safwan Uddin, Mohammed Mouzam, Mohammed Imran +1

Agentic CommerceTool Access Control for LLM AgentsAI Agent SecurityAccess Control
AskPDF
Apr 1, 2026

ClawSafety: "Safe" LLMs, Unsafe Agents

Personal AI agents like OpenClaw run with elevated privileges on users' local machines, where a single successful prompt injection can leak credentials, redirect financial transactions, or destroy files. This threat goes well beyond conventional text-level jailbreaks, yet existing safety evaluations fall short: most test models in isolated chat settings, rely on synthetic environments, and do not account for how the agent framework itself shapes safety outcomes. We introduce CLAWSAFETY, a benchmark of 120 adversarial test scenarios organized along three dimensions (harm domain, attack vector, and harmful action type) and grounded in realistic, high-privilege professional workspaces spanning software engineering, finance, healthcare, law, and DevOps. Each test case embeds adversarial content in one of three channels the agent encounters during normal work: workspace skill files, emails from trusted senders, and web pages. We evaluate five frontier LLMs as agent backbones, running 2,520 sandboxed trials across all configurations. Attack success rates (ASR) range from 40% to 75% across models and vary sharply by injection vector, with skill instructions (highest trust) consistently more dangerous than email or web content. Action-trace analysis reveals that the strongest model maintains hard boundaries against credential forwarding and destructive actions, while weaker models permit both. Cross-scaffold experiments on three agent frameworks further demonstrate that safety is not determined by the backbone model alone but depends on the full deployment stack, calling for safety evaluation that treats model and framework as joint variables. Code and data will be available at: https://weibowen555.github.io/ClawSafety/.

Bowen Wei, Yunbei Zhang, Jinhao Pan +5

George Mason University · Tulane University · Oak Ridge National Laboratory +1

AI Agent SecurityAI Agent SafetyPrompt Injection Attacks on AI AgentsAI Agent Safety Benchmarks
AskPDF
Mar 31, 2026

Architecting Secure AI Agents: Perspectives on System-Level Defenses Against Indirect Prompt Injection Attacks

AI agents, predominantly powered by large language models (LLMs), are vulnerable to indirect prompt injection, in which malicious instructions embedded in untrusted data can trigger dangerous agent actions. This position paper discusses our vision for system-level defenses against indirect prompt injection attacks. We articulate three positions: (1) dynamic replanning and security policy updates are often necessary for dynamic tasks and realistic environments; (2) certain context-dependent security decisions would still require LLMs (or other learned models), but should only be made within system designs that strictly constrain what the model can observe and decide; (3) in inherently ambiguous cases, personalization and human interaction should be treated as core design considerations. In addition to our main positions, we discuss limitations of existing benchmarks that can create a false sense of utility and security. We also highlight the value of system-level defenses, which serve as the skeleton of agentic systems by structuring and controlling agent behaviors, integrating rule-based and model-based security checks, and enabling more targeted research on model robustness and human interaction.

Chong Xiang, Drew Zagieboylo, Shaona Ghosh +5

NVIDIA · Independent Researcher · Johns Hopkins University

AI Agent SecurityPrompt Injection DefenseIndirect Prompt Injection
AskPDF
Mar 30, 2026

"What Did It Actually Do?": Understanding Risk Awareness and Traceability for Computer-Use Agents

Personalized computer-use agents are rapidly moving from expert communities into mainstream use. Unlike conventional chatbots, these systems can install skills, invoke tools, access private resources, and modify local environments on users' behalf. Yet users often do not know what authority they have delegated, what the agent actually did during task execution, or whether the system has been safely removed afterward. We investigate this gap as a combined problem of risk understanding and post-hoc auditability, using OpenClaw as a motivating case. We first build a multi-source corpus of the OpenClaw ecosystem, including incidents, advisories, malicious-skill reports, news coverage, tutorials, and social-media narratives. We then conduct an interview study to examine how users and practitioners understand skills, autonomy, privilege, persistence, and uninstallation. Our findings suggest that participants often recognized these systems as risky in the abstract, but lacked concrete mental models of what skills can do, what resources agents can access, and what changes may remain after execution or removal. Motivated by these findings, we propose AgentTrace, a traceability framework and prototype interface for visualizing agent actions, touched resources, permission history, provenance, and persistent side effects. A scenario-based evaluation suggests that traceability-oriented interfaces can improve understanding of agent behavior, support anomaly detection, and foster more calibrated trust.

Zifan Peng, Mingchen Li

The Hong Kong University of Science and Technology (Guangzhou), China · University of North Texas, The United States

Computer-Use Agent SafetyComputer-Use AgentsAI Agent AuditingAI Agent MonitoringAI Agent Security
AskPDF
Mar 29, 2026

A Security Analysis of the OpenClaw AI Agent Framework

AI agent frameworks connecting large language model (LLM) reasoning to host execution surfaces -- shell, filesystem, containers, and messaging -- introduce security challenges structurally distinct from conventional software. We present a systematic taxonomy of 470 advisories filed against OpenClaw, an open-source AI agent runtime, organized by architectural layer and trust-violation type. Vulnerabilities cluster along two orthogonal axes: (1) the system axis, reflecting the architectural layer (exec policy, gateway, channel, sandbox, browser, plugin, agent/prompt); and (2) the attack axis, reflecting adversarial techniques (identity spoofing, policy bypass, cross-layer composition, prompt injection, supply-chain escalation). Patch-differential evidence yields three principal findings. First, three Moderate- or High-severity advisories in the Gateway and Node-Host subsystems compose into a complete unauthenticated remote code execution (RCE) path -- spanning delivery, exploitation, and command-and-control -- from an LLM tool call to the host process. Second, the exec allowlist, the primary command-filtering mechanism, relies on a closed-world assumption that command identity is recoverable via lexical parsing. This is invalidated by shell line continuation, busybox multiplexing, and GNU option abbreviation. Third, a malicious skill distributed via the plugin channel executed a two-stage dropper within the LLM context, bypassing the exec pipeline and demonstrating that the skill distribution surface lacks runtime policy enforcement. The dominant structural weakness is per-layer trust enforcement rather than unified policy boundaries, making cross-layer attacks resilient to local remediation.

Surada Suwansathit, Yuxuan Zhang, Guofei Gu

SUCCESS Lab, Texas A&M University

AI Agent SecurityTool Access Control for LLM AgentsAgent Skill Security Auditing
AskPDF
Mar 28, 2026

Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses

Embodied Artificial Intelligence (Embodied AI) integrates perception, cognition, planning, and interaction into agents that operate in open-world, safety-critical environments. As these systems gain autonomy and enter domains such as transportation, healthcare, and industrial or assistive robotics, ensuring their safety becomes both technically challenging and socially indispensable. Unlike digital AI systems, embodied agents must act under uncertain sensing, incomplete knowledge, and dynamic human-robot interactions, where failures can directly lead to physical harm. This survey provides a comprehensive and structured review of safety research in embodied AI, examining attacks and defenses across the full embodied pipeline, from perception and cognition to planning, action and interaction, and agentic system. We introduce a multi-level taxonomy that unifies fragmented lines of work and connects embodied-specific safety findings with broader advances in vision, language, and multimodal foundation models. Our review synthesizes insights from over 500 papers spanning adversarial, backdoor, jailbreak, and hardware-level attacks; attack detection, safe training and robust inference; and risk-aware human-agent interaction. This analysis reveals several overlooked challenges, including the fragility of multimodal perception fusion, the instability of planning under jailbreak attacks, and the trustworthiness of human-agent interaction in open-ended scenarios. By organizing the field into a coherent framework and identifying critical research gaps, this survey provides a roadmap for building embodied agents that are not only capable and autonomous but also safe, robust, and reliable in real-world deployment.

Xiao Li, Xiang Zheng, Yifeng Gao +35

Fudan University · City University of Hong Kong · Jilin University +10

Embodied AIAI SafetyAI Agent SecurityRobot SafetyHuman-Robot Interaction
AskPDF
Mar 25, 2026

AIP: Agent Identity Protocol for Verifiable Delegation Across MCP and A2A

AI agents increasingly call tools via the Model Context Protocol (MCP) and delegate to other agents via Agent-to-Agent (A2A), yet neither protocol verifies agent identity. A scan of approximately 2,000 MCP servers found all lacked authentication. In our survey, we did not identify a prior implemented protocol that jointly combines public-key verifiable delegation, holder-side attenuation, expressive chained policy, transport bindings across MCP/A2A/HTTP, and provenance-oriented completion records. We introduce Invocation-Bound Capability Tokens (IBCTs), a primitive that fuses identity, attenuated authorization, and provenance binding into a single append-only token chain. IBCTs operate in two wire formats: compact mode (a signed JWT for single-hop cases) and chained mode (a Biscuit token with Datalog policies for multi-hop delegation). We provide reference implementations in Python and Rust with full cross-language interoperability. Compact mode verification takes 0.049ms (Rust) and 0.189ms (Python), with 0.22ms overhead over no-auth in real MCP-over-HTTP deployment. In a real multi-agent deployment with Gemini 2.5 Flash, AIP adds 2.35ms of overhead (0.086% of total end-to-end latency). Adversarial evaluation across 600 attack attempts shows 100% rejection rate, with two attack categories (delegation depth violation and audit evasion through empty context) uniquely caught by AIP's chained delegation model that neither unsigned nor plain JWT deployments detect.

Sunil Prakash

Indian School of Business, India

AI Agent SecurityTool Access Control for LLM AgentsLLM Agent VerificationCapability-Based Security
AskPDF
Mar 25, 2026

ClawKeeper: Comprehensive Safety Protection for OpenClaw Agents Through Skills, Plugins, and Watchers

OpenClaw has rapidly established itself as a leading open-source autonomous agent runtime, offering powerful capabilities including tool integration, local file access, and shell command execution. However, these broad operational privileges introduce critical security vulnerabilities, transforming model errors into tangible system-level threats such as sensitive data leakage, privilege escalation, and malicious third-party skill execution. Existing security measures for the OpenClaw ecosystem remain highly fragmented, addressing only isolated stages of the agent lifecycle rather than providing holistic protection. To bridge this gap, we present ClawKeeper, a real-time security framework that integrates multi-dimensional protection mechanisms across three complementary architectural layers. (1) Skill-based protection operates at the instruction level, injecting structured security policies directly into the agent context to enforce environment-specific constraints and cross-platform boundaries. (2) Plugin-based protection serves as an internal runtime enforcer, providing configuration hardening, proactive threat detection, and continuous behavioral monitoring throughout the execution pipeline. (3) Watcher-based protection introduces a novel, decoupled system-level security middleware that continuously verifies agent state evolution. It enables real-time execution intervention without coupling to the agent's internal logic, supporting operations such as halting high-risk actions or enforcing human confirmation. We argue that this Watcher paradigm holds strong potential to serve as a foundational building block for securing next-generation autonomous agent systems. Extensive qualitative and quantitative evaluations demonstrate the effectiveness and robustness of ClawKeeper across diverse threat scenarios. We release our code.

Songyang Liu, Chaozhuo Li, Chenxu Wang +8

Beijing University of Posts and Telecommunications · Beijing Academy of Artificial Intelligence · China Academy of Information and Communications Technology

AI Agent SecurityLLM Agent SecurityRuntime Enforcement for AI AgentsTool Access Control for LLM AgentsAI Agent Monitoring
AskPDF
Mar 24, 2026

The Cognitive Firewall:Securing Browser Based AI Agents Against Indirect Prompt Injection Via Hybrid Edge Cloud Defense

Deploying large language models (LLMs) as autonomous browser agents exposes a significant attack surface in the form of Indirect Prompt Injection (IPI). Cloud-based defenses can provide strong semantic analysis, but they introduce latency and raise privacy concerns. We present the Cognitive Firewall, a three-stage split-compute architecture that distributes security checks across the client and the cloud. The system consists of a local visual Sentinel, a cloud-based Deep Planner, and a deterministic Guard that enforces execution-time policies. Across 1,000 adversarial samples, edge-only defenses fail to detect 86.9% of semantic attacks. In contrast, the full hybrid architecture reduces the overall attack success rate (ASR) to below 1% (0.88% under static evaluation and 0.67% under adaptive evaluation), while maintaining deterministic constraints on side-effecting actions. By filtering presentation-layer attacks locally, the system avoids unnecessary cloud inference and achieves an approximately 17,000x latency advantage over cloud-only baselines. These results indicate that deterministic enforcement at the execution boundary can complement probabilistic language models, and that split-compute provides a practical foundation for securing interactive LLM agents.

Qianlong Lan, Anuj Kaul

eBay Inc

AI Agent SecurityPrompt Injection DefensePrompt Injection Attacks on AI AgentsEdge-Cloud ComputingTool Access Control for LLM Agents
AskPDF
Mar 24, 2026

Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution

We identify a critical security vulnerability in mainstream Claw personal AI agents: untrusted content encountered during heartbeat-driven background execution can silently pollute agent memory and subsequently influence user-facing behavior without the user's awareness. This vulnerability arises from an architectural design shared across the Claw ecosystem: heartbeat background execution runs in the same session as user-facing conversation, so content ingested from any external source monitored in the background (including email, message channels, news feeds, code repositories, and social platforms) can enter the same memory context used for foreground interaction, often with limited user visibility and without clear source provenance. We formalize this process as an Exposure (E) →\rightarrow Memory (M) →\rightarrow Behavior (B) pathway: misinformation encountered during heartbeat execution enters the agent's short-term session context, potentially gets written into long-term memory, and later shapes downstream user-facing behavior. We instantiate this pathway in an agent-native social setting using MissClaw, a controlled research replica of Moltbook. We find that (1) social credibility cues, especially perceived consensus, are the dominant driver of short-term behavioral influence, with misleading rates up to 61%; (2) routine memory-saving behavior can promote short-term pollution into durable long-term memory at rates up to 91%, with cross-session behavioral influence reaching 76%; (3) under naturalistic browsing with content dilution and context pruning, pollution still crosses session boundaries. Overall, prompt injection is not required: ordinary social misinformation is sufficient to silently shape agent memory and behavior under heartbeat-driven background execution.

Yechao Zhang, Shiqian Zhao, Jie Zhang +5

Nanyang Technological University · Agency for Science, Technology and Research (A*STAR) · Johns Hopkins University

AI Agent SecurityAgent Memory PoisoningAgent Memory Management
AskPDF
Mar 24, 2026

AgentRAE: Remote Action Execution through Notification-based Visual Backdoors against Screenshots-based Mobile GUI Agents

The rapid adoption of mobile graphical user interface (GUI) agents, which autonomously control applications and operating systems (OS), exposes new system-level attack surfaces. Existing backdoors against web GUI agents and general GenAI models rely on environmental injection or deceptive pop-ups to mislead the agent operation. However, these techniques do not work on screenshots-based mobile GUI agents due to the challenges of restricted trigger design spaces, OS background interference, and conflicts in multiple trigger-action mappings. We propose AgentRAE, a novel backdoor attack capable of inducing Remote Action Execution in mobile GUI agents using visually natural triggers (e.g., benign app icons in notifications). To address the underfitting caused by natural triggers and achieve accurate multi-target action redirection, we design a novel two-stage pipeline that first enhances the agent's sensitivity to subtle iconographic differences via contrastive learning, and then associates each trigger with a specific mobile GUI agent action through a backdoor post-training. Our extensive evaluation reveals that the proposed backdoor preserves clean performance with an attack success rate of over 90% across ten mobile operations. Furthermore, it is hard to visibly detect the benign-looking triggers and circumvents eight representative state-of-the-art defenses. These results expose an overlooked backdoor vector in mobile GUI agents, underscoring the need for defenses that scrutinize notification-conditioned behaviors and internal agent representations.

Yutao Luo, Haotian Zhu, Shuchao Pang +4

Nanjing University of Science and Technology, China · Macquarie University, Australia · Western Sydney University, Australia +2

Backdoor AttacksAI Agent SecurityMobile GUI Agents
AskPDF
Mar 24, 2026

Agent-Sentry: Bounding LLM Agents via Execution Provenance

Agentic computing systems, while immensely capable, raise serious security, privacy, and safety concerns. A key issue is that the full set of functionalities offered by these systems, combined with their probabilistic execution flows, is not known beforehand. Given this lack of characterization, it is challenging to validate whether a system has successfully carried out the user's intended task or instead executed irrelevant actions, potentially as a consequence of compromise. We present Agent Sentry, a runtime defense that learns a bound on an agent's benign execution from prior legitimate executions and flags any action that falls outside this bound. Agent Sentry layers three complementary checks: a structural classifier over the sequence of actions and the provenance of each function's arguments; a deterministic allowlist check over sensitive argument values; and an LLM judge, invoked only on the residual of actions where the first two checks cannot safely decide between a legitimate new request and a carefully crafted injection. We demonstrate the effectiveness of Agent Sentry in AgentDojo and AgentDyn by blocking 94.3% of successful injections while allowing 95.1% of benign executions, without modifying the agent, its tools, or the LLM.

Rohan Sequeira, Stavros Damianakis, Umar Iqbal +1

University of Southern California, Los Angeles, California, USA · Washington University in St. Louis, St. Louis, Missouri, USA

Computer-Use Agent SafetyAI Agent SecurityAI Agent MonitoringRuntime Enforcement for AI AgentsPrompt Injection Attacks on AI Agents
AskPDF