← Back to all categories

Agentic Threats

39 resources

Attacks & Threats

Tool misuse, autonomous harm, and agent-specific attack vectors

paper 2026

The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneck

Linfeng Fan, Ziwei Li, Yuan Tian + 3 more

Tool-using LLM agents must act on untrusted webpages, emails, files, and API outputs while issuing privileged tool calls. Existing defenses often mediate trust at the granularity of an entire tool invocation, forcing a brittle choice in mixed-trust workflows: allow external content to influence a call and risk hijacked destinations or commands, or quarantine the call and block benign retrieval-then-act behavior. The key observation behind this paper is that indirect prompt injection becomes dang

paper 2026

Towards Security-Auditable LLM Agents: A Unified Graph Representation

Chaofan Li, Lyuye Zhang, Jintao Zhai + 9 more

LLM-based agentic systems are rapidly evolving to perform complex autonomous tasks through dynamic tool invocation, stateful memory management, and multi-agent collaboration. However, this semantics-driven execution paradigm creates a severe semantic gap between low-level physical events and high-level execution intent, making post-hoc security auditing fundamentally difficult. Existing representation mechanisms, including static SBOMs and runtime logs, provide only fragmented evidence and fail

paper 2026

Safety Context Injection: Inference-Time Safety Alignment via Static Filtering and Agentic Analysis

Zhenhao Xu, Wenhan Chang, Yichuan Chen + 3 more

Large Reasoning Models (LRMs) improve performance on complex tasks, but they also make safety control harder at deployment time. In black-box settings, defenders cannot modify model weights and must instead intervene at inference time. This setting creates three practical challenges: harmful intent may be hidden by educational or role-play framing, deep safety analysis can introduce non-trivial latency, and long adversarial contexts can dilute the local cues that simpler filters rely on. These c

paper 2026

When Child Inherits: Modeling and Exploiting Subagent Spawn in Multi-Agent Networks

Ziwen Cai, Yihe Zhang, Xiali Hei

Since the official release of ChatGPT in 2022, large language models (LLMs) have rapidly evolved from chatbot-style interfaces into agentic systems that can delegate work through tools and newly spawned subagents. While these capabilities improve automation and scalability, they also pose new security risks in multi-agent networks. Existing research has studied how individual LLM-based agents can be compromised through prompt injection, jailbreaking, poisoned retrieval data, or malicious extensi

paper 2026

WAAA! Web Adversaries Against Agentic Browsers

Sohom Datta, Alex Nahapetyan, William Enck + 1 more

Large language models (LLMs) are increasingly being integrated into web browsers to create agentic browsing systems that execute actions on behalf of the user. Prior work considering the security of agentic browsers focuses exclusively on indirect prompt-injection attacks. However, by failing to consider traditional web attacks, previous agentic browser threat models have a blind spot to web social engineering attacks originally designed to trick humans. In this paper, we propose the first web-f

paper 2026

Redefining AI Red Teaming in the Agentic Era: From Weeks to Hours

Raja Sekhar Rao Dheekonda, Will Pearce, Nick Landers

AI systems are entering critical domains like healthcare, finance, and defense, yet remain vulnerable to adversarial attacks. While AI red teaming is a primary defense, current approaches force operators into manual, library-specific workflows. Operators spend weeks hand-crafting workflows - assembling attacks, transforms, and scorers. When results fall short, workflows must be rebuilt. As a result, operators spend more time constructing workflows than probing targets for security and safety vul

paper 2026

When Agents Handle Secrets: A Survey of Confidential Computing for Agentic AI

Javad Forough, Marios Kogias, Hamed Haddadi

Agentic AI systems, specifically LLM-driven agents that plan, invoke tools, maintain persistent memory, and delegate tasks to peer agents via protocols such as MCP and A2A, introduce a threat surface that differs materially from standalone model inference. Agents accumulate sensitive context, hold credentials, and operate across pipelines no single party fully controls, enabling prompt injection, context exfiltration, credential theft, and inter-agent message poisoning. Current defenses operate

paper 2026

When Agents Overtrust Environmental Evidence: An Extensible Agentic Framework for Benchmarking Evidence-Grounding Defects in LLM Agents

Strick Sheng, Ziyue Wang, Liyi Zhou

Large language model agents increasingly operate through environment-facing scaffolds that expose files, web pages, APIs, and logs. These observations influence tool use, state tracking, and action sequencing, yet their reliability and authority are often uncertain. Environmental grounding is therefore a systems-level problem involving context admission, evidence provenance, freshness checking, verification policy, action gating, and model reasoning. Existing agent benchmarks mainly evaluate tas

paper 2026

Architecture Matters: Comparing RAG Systems under Knowledge Base Poisoning

Samuel Korn

Retrieval-Augmented Generation (RAG) systems are vulnerable to knowledge base poisoning, yet existing attacks have been evaluated almost exclusively against vanilla retrieve-then-generate pipelines. Architectures designed to handle conflicting retrieved information - multi-agent debate, agentic retrieval, recursive language models - remain untested against adversarially optimized contradictions. We evaluate four RAG architectures (vanilla RAG, agentic RAG, MADAM-RAG, and Recursive Language Model

paper 2026

SoK: The Attack Surface of Agentic AI -- Tools, and Autonomy

Ali Dehghantanha, Sajad Homayoun

Recent AI systems combine large language models with tools, external knowledge via retrieval-augmented generation (RAG), and even autonomous multi-agent decision loops. This agentic AI paradigm greatly expands capabilities - but also vastly enlarges the attack surface. In this systematization, we map out the trust boundaries and security risks of agentic LLM-based systems. We develop a comprehensive taxonomy of attacks spanning prompt-level injections, knowledge-base poisoning, tool/plug-in expl

paper 2026

Agentic AI in Industry: Adoption Level and Deployment Barriers

Spyridon Alvanakis Apostolou, Jan Bosch, Helena Holmström Olsson

Agentic AI systems are entering software engineering workflows, yet empirical evidence on how industrial organizations actually adopt them remains sparse. We present a qualitative interview study with sixteen practitioners across twelve companies of varying size and domain. This study characterizes the current agentic AI adoption state of these companies, employing a six-level maturity framework adapted from established AI-driven organizations. The findings reveal that seven companies operate at

paper 2026

Attacks and Mitigations for Distributed Governance of Agentic AI under Byzantine Adversaries

Matthew D. Laws, Alina Oprea, Cristina Nita-Rotaru

Agentic AI governance is a critical component of agentic AI infrastructure ensuring that agents follow their owner's communication and interaction policies, and providing protection against attacks from malicious agents. The state-of-the-art solution, SAGA, assumes a logically centralized point of trust, the Provider, which serves as a repository for user and agent information and actively enforces policies. While SAGA provides protection against malicious agents, it remains vulnerable to a mali

paper 2026

Red-Teaming Agent Execution Contexts: Open-World Security Evaluation on OpenClaw

Hongwei Yao, Yiming Liu, Yiling He + 1 more

Agentic language-model systems increasingly rely on mutable execution contexts, including files, memory, tools, skills, and auxiliary artifacts, creating security risks beyond explicit user prompts. This paper presents DeepTrap, an automated framework for discovering contextual vulnerabilities in OpenClaw. DeepTrap formulates adversarial context manipulation as a black-box trajectory-level optimization problem that balances risk realization, benign-task preservation, and stealth. It combines ris

paper 2026

Token Economics for LLM Agents: A Dual-View Study from Computing and Economics

Yuxi Chen, Junming Chen, Chenyu He + 9 more

As LLM agents evolve, tokens have emerged as the core economic primitives of Agentic AI. However, their exponential consumption introduces severe computational, collaborative, and security bottlenecks. Current surveys remain fragmented across system optimization, architecture design, and trust, lacking a unified framework to evaluate the fundamental trade-off between output quality and economic cost. To bridge this gap, this survey presents the first comprehensive survey of Token Economics. By u

paper 2026

Containment Verification: AI Safety Guarantees Independent of Alignment

Royce Moon, Lav R. Varshney

Agentic frameworks are the software layer through which AI agents act in the world. Existing safety methods intervene on the model and therefore remain conditional on unverifiable properties of learned behavior. We introduce containment verification, which locates safety guarantees in the agentic framework itself. Under havoc oracle semantics, the AI is modeled as an unconstrained oracle ranging over the entire typed action space, and the verified containment layer must enforce the boundary poli

paper 2026

From Assistance to Agency: Rethinking Autonomy and Control in CI/CD Pipelines

Marcus Emmanuel Barnes, Taher A. Ghaleb, Safwat Hassan

AI agents are assuming active roles in Continuous Integration and Continuous Deployment (CI/CD) workflows, yet the research community lacks a shared vocabulary for describing what it means for CI/CD to be agentic, how much decision authority is delegated, and where control should reside. This paper presents a vision of agentic CI/CD in which the central challenge is not improving task performance but designing authority transfer, defined as the delegation of operational decisions from human-cont

paper 2026

Why Does Agentic Safety Fail to Generalize Across Tasks?

Yonatan Slutzky, Yotam Alexander, Tomer Slor + 2 more

AI agents are increasingly deployed in multi-task settings, where the task to perform is specified at test time, and the agent must generalize to unseen tasks. A major concern in such settings is safety: often, an agent must not only execute unseen tasks, but do so while avoiding risks and handling ones that materialize. Empirical evidence suggests that even when the ability to execute generalizes to unseen tasks, the ability to do so safely frequently does not. This paper provides theory and ex

paper 2026

MAGIQ: A Post-Quantum Multi-Agentic AI Governance System with Provable Security

Sepideh Avizeh, Tushin Mallick, Alina Oprea + 2 more

Our computing ecosystem is being transformed by two emerging paradigms: the increased deployment of agentic AI systems and advancements in quantum computing. With respect to agentic AI systems, one of the most critical problems is creating secure governing architectures that ensure agents follow their owners' communication and interaction policies and can be held accountable for the messages they exchange with other agents. With respect to quantum computing, existing systems must be retrofitted

paper 2026

Beyond the Black Box: Interpretability of Agentic AI Tool Use

Hariom Tatsat, Ariye Shater

AI agents are promising for high-stakes enterprise workflows, but dependable deployment remains limited because tool-use failures are difficult to diagnose and control. Agents may skip required tool calls, invoke tools unnecessarily, or take actions whose consequence becomes visible only after execution. Existing observability methods are mostly external: prompts reveal correlations, evaluations score outputs, and logs arrive only after the model has already acted. In long-horizon settings, thes

paper 2026

Agentic AI and the Industrialization of Cyber Offense: Forecast, Consequences, and Defensive Priorities for Enterprises and the Mittelstand

Christopher Koch

Agentic AI systems can plan, call tools, inspect code, interact with web applications, and coordinate multi-step workflows. These same capabilities change the economics of cyber offense. The central near-term risk is not that every low-skill criminal immediately becomes a frontier exploit researcher; it is that agentic AI compresses the attack lifecycle by lowering the cost of reconnaissance, phishing, credential abuse, vulnerability triage, exploit adaptation, and post-compromise decision suppo

paper 2026

Authorization Propagation in Multi-Agent AI Systems: Identity Governance as Infrastructure

Krti Tallam

The security discussion around agentic AI focuses heavily on prompt injection. This paper argues that multi-agent systems also create a distinct authorization problem: maintaining authorization invariants as non-human principals retrieve data, delegate tasks, and synthesize results across changing boundaries. We call this problem authorization propagation. It is not reducible to prompt injection and is not fully addressed by classical access-control models such as RBAC, ABAC, or ReBAC. The paper

paper 2026

Securing the Agent: Vendor-Neutral, Multitenant Enterprise Retrieval and Tool Use

Francisco Javier Arceo, Varsha Prasad Narsing

Retrieval-Augmented Generation (RAG) and agentic AI systems are increasingly prevalent in enterprise AI deployments. However, real enterprise environments introduce challenges largely absent from academic treatments and consumer-facing APIs: multiple tenants with heterogeneous data, strict access-control requirements, regulatory compliance, and cost pressures that demand shared infrastructure. A fundamental problem underlies existing RAG architectures in these settings: retrieval systems rank do

paper 2026

DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents

Zhaorun Chen, Xun Liu, Haibo Tong + 14 more

AI agents are increasingly deployed across diverse domains to automate complex workflows through long-horizon and high-stakes action executions. Due to their high capability and flexibility, such agents raise significant security and safety concerns. A growing number of real-world incidents have shown that adversaries can easily manipulate agents into performing harmful actions, such as leaking API keys, deleting user data, or initiating unauthorized transactions. Evaluating agent security is in

paper 2026

Architectural Obsolescence of Unhardened Agentic-AI Runtimes

Alfredo Metere

An agentic-AI runtime issues tool calls, sends messages, and actuates devices on behalf of an LLM. Catching the four ways an action can diverge from its audit record -- F1 gate-bypass, F2 audit-forgery, silent host failure, F4 wrong-target, -- is a load-bearing safety property of any such runtime. We show that upstream OpenClaw, the most engineered single-user agentic-AI gateway in public release, catches none of them: recall is 0.000 on every cell of every confusion matrix, on a 1600-sample tem

paper 2026

Position: Safety and Fairness in Agentic AI Depend on Interaction Topology, Not on Model Scale or Alignment

Tanav Singh Bajaj, Nikhil Singh, Karan Anand + 1 more

As large language models are increasingly deployed as interacting agents in high-stakes decisions, the AI safety community assumes that safety properties of individual models will compose into safe multi-agent behavior. This position paper argues that this assumption is fundamentally mistaken. In agentic AI, safety is determined by interaction topology, not model weights. When agents deliberate sequentially or aggregate via parallel voting with a judge, the structure of information flow and deci

paper 2026

AgenticAITA: A Proof-Of-Concept About Deliberative Multi-Agent Reasoning for Autonomous Trading Systems

Ivan Letteri

Conventional algorithmic trading systems are grounded in deterministic heuristics or offline-trained statistical models that cannot adapt to the semantic complexity of rapidly shifting market regimes. This paper introduces AGENTICAITA, an agentic AI framework that replaces the traditional signal then execute paradigm with a fully autonomous deliberative loop in which multiple specialized Large Language Model agents reason, negotiate, and act in concert - without any offline training or human int

paper 2026

BioVeil MATRIX: Uncovering and categorizing vulnerabilities of agentic biological AI scientists

Kimon Antonios Provatas, Avery Self, Ioannis Mouratidis + 1 more

Agentic AI scientists equipped with domain-specific tools are rapidly entering scientific workflows across disciplines, with especially strong uptake in the life sciences where they can be used for literature synthesis, sequence analysis, and experimental planning support. While these systems accelerate biological research, they also introduce risks for dual-use applications that are not captured by current model-centric safety evaluations. We present evidence that current agentic AI scientists,

paper 2026

AgentReputation: A Decentralized Agentic AI Reputation Framework

Mohd Sameen Chishti, Damilare Peter Oyinloye, Jingyue Li

Decentralized, agentic AI marketplaces are rapidly emerging to support software engineering tasks such as debugging, patch generation, and security auditing, often operating without centralized oversight. However, existing reputation mechanisms fail in this setting for three fundamental reasons: agents can strategically optimize against evaluation procedures; demonstrated competence does not reliably transfer across heterogeneous task contexts; and verification rigor varies widely, from lightwei

paper 2026

Contextual Agentic Memory is a Memo, Not True Memory

Binyan Xu, Xilin Dai, Kehuan Zhang

Current agentic memory systems (vector stores, retrieval-augmented generation, scratchpads, and context-window management) do not implement memory: they implement lookup. We argue that treating lookup as memory is a category error with provable consequences for agent capability, long-term learning, and security. Retrieval generalizes by similarity to stored cases; weight-based memory generalizes by applying abstract rules to inputs never seen before. Conflating the two produces agents that accum

paper 2026

Lightweight Quantum Agent for Edge Systems: Joint PQC and NOMA Resource Allocation

Yongtao Yao, Wenjing Xiao, Miaojiang Chen + 5 more

In the context of quantum secure scenarios, existing research on mobile edge devices and intelligent computing and edge (ICE) systems based on the Non-Orthogonal Multiple Access (NOMA) communication model have overlooked the energy consumption overhead of Post-Quantum Cryptography (PQC) modules, and the high complexity of traditional resource allocation algorithms fails to meet the demands of real-time decision-making. To address these challenges, this paper proposes a lightweight agentic AI fra

paper 2025

Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges

Anshuman Chhabra, Shrestha Datta, Shahriar Kabir Nahin + 1 more

Agentic AI systems powered by large language models (LLMs) and endowed with planning, tool use, memory, and autonomy, are emerging as powerful, flexible platforms for automation. Their ability to autonomously execute tasks across web, software, and physical environments creates new and amplified security risks, distinct from both traditional AI safety and conventional software security. This survey outlines a taxonomy of threats specific to agentic AI, reviews recent benchmarks and evaluation me

paper reviewed open access 2024

LLM Agents Can Autonomously Hack Websites

Richard Fang, Rohan Bindu, Akul Gupta + 2 more — arXiv preprint

Demonstrates that LLM agents can autonomously perform web hacking tasks including SQL injection, XSS, and CSRF attacks without human guidance.

paper reviewed open access 2024

LLM Agents Can Autonomously Exploit One-day Vulnerabilities

Richard Fang, Rohan Bindu, Akul Gupta + 1 more — arXiv preprint

Shows that LLM agents (GPT-4) can autonomously exploit real-world one-day vulnerabilities given CVE descriptions, achieving 87% success rate.

paper reviewed open access 2024

AgentDojo: A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents

Edoardo Debenedetti, Jie Zhang, Mislav Balunovic + 3 more — arXiv preprint

Introduces AgentDojo, a framework for evaluating the security of LLM agents against prompt injection and other attacks in realistic tool-use scenarios.

paper reviewed open access 2024

R-Judge: Benchmarking Safety Risk Awareness for LLM Agents

Tongxin Yuan, Zhiwei He, Lingzhong Dong + 9 more — EMNLP 2024

Introduces R-Judge benchmark for evaluating whether LLM agents can identify safety risks in agentic scenarios involving tool use and multi-step reasoning.

paper reviewed open access 2024

The Emerged Security and Privacy of LLM Agent: A Survey with Case Studies

Feng He, Tianqing Zhu, Dayong Ye + 3 more — arXiv preprint

Surveys security and privacy challenges specific to LLM-based agents, covering agent architectures, attack surfaces, and defense mechanisms.

paper reviewed open access 2024

ConfusedPilot: Confused Deputy Attacks Against RAG-based Code Assistants

Andrew Patel, Hossein Aboutorab, Ilia Kolochenko — arXiv preprint

Introduces confused deputy attacks against RAG-based code assistants like GitHub Copilot, where poisoned code repositories manipulate assistant outputs.

paper reviewed open access 2024

Adversarial Attacks on Multimodal Agents

Chen Henry Wu, Jing Yu Koh, Ruslan Salakhutdinov + 2 more — arXiv preprint

Demonstrates adversarial attacks on multimodal agents that take actions in digital environments, showing visual perturbations can hijack agent behavior.

paper reviewed open access 2023

Not What You've Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection

Kai Greshake, Sahar Abdelnabi, Shailesh Mishra + 3 more — AISec 2023

Introduces indirect prompt injection attacks against LLM-integrated applications, demonstrating how adversaries can remotely control LLMs by injecting prompts into data sources the LLM retrieves.