Prompt injection: research

Journal articles, academic blogs, think-tank reports and papers in this section, newest first. Back to the news.

arXiv (research papers) Md Anas Biswas SI

Confidently Wrong: Severity-Aware Calibration of Prompt-Injection Detectors under Attack Shift

Prompt-injection detectors are deployed as guards: a model scores an input and a downstream system trusts or blocks it on that score. I study the confidence of these scores, not only their accuracy, when the attack distribution shifts away from the clean benchmark on which the operating point was chosen. I evaluate three released detectors, ProtectAI-v2 and two Prompt-Guard-2 checkpoints, at a single source-calibrated threshold that I freeze and transport across five shifts…

Read at arXiv (research papers)

arXiv (research papers) Zesen Liu SI

Safe to Check, Unsafe to Use: Relinking at the Compression Boundary of LLM Agents

Summarization-based prompt compression is increasingly used by LLM agents to shorten long, distributed contexts, but it shifts the security boundary: filters inspect the pre-compression prompt while the backend acts on a newly generated compressed context. We identify relinking, a compression-boundary vulnerability where the compressor behaves as a confused deputy, summarizing distributed, locally benign fragments into a complete malicious instruction. Unlike prompt…

Read at arXiv (research papers)

arXiv (research papers) Zhen Zhao SI

AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI Agents

Traditional OS security models based on "resource exposure plus permission checks" face structural challenges as LLM-driven autonomous agents acquire capabilities for planning, tool use, network access, and code execution. Once an agent runtime is compromised through prompt injection or malicious tool outputs, an attacker can compose POSIX-style resource primitives into behaviors far beyond the user's task authorization. To address this, we propose AgenticOS, an…

Read at arXiv (research papers)

arXiv (research papers) Zhengsong Zhang SI

Local LLM Agents as Vulnerable Runtimes:A Source-Code Audit of the Agent Runtime Layer

Local LLM agents such as OpenClaw and Nanobot run on end-user machines and act on host resources - the shell, filesystem, browser, stored credentials, and messaging applications - through natural-language goals. These agents have become privileged software runtimes that mediate between user intent, model outputs, and host-level actions. Existing research characterizes the landscape through prompt injection, malicious skills, marketplace risks, or black-box evaluation of…

Read at arXiv (research papers)

arXiv (research papers) Shanghao Shi SI

Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning

The integration of external tools has substantially expanded the capabilities of large language model (LLM) agents, but it also introduces new attack surfaces beyond prompt injection. In particular, cross-tool description poisoning can manipulate planner-visible tool metadata to steer an agent's trajectory, even if the poisoned tool itself is never chosen. To understand the effectiveness of existing defenses against this emerging threat, we first evaluate several…

Read at arXiv (research papers)

arXiv (research papers) Reza Soosahabi SI

Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems

Agentic AI systems increasingly rely on language-model components to interpret instructions, process external data, invoke tools, and coordinate with other agents. These capabilities make prompt-injection and jailbreak attacks more consequential, especially as attackers adopt model-guided automation to scale probing, prompt refinement, and response evaluation. This work analyzes the resulting attack-defense setting through a probabilistic model of a target system, its…

Read at arXiv (research papers)

arXiv (research papers) Gulshan Saleem SI

A Layered Security Framework Against Prompt Injection in RAG-Based Chatbots

Prompt injection is ranked as the most critical vulnerability in large language model (LLM) deployments by the OWASP Top 10 for LLM Applications, yet existing defenses operate at isolated pipeline stages and remain incomplete. Input filters cannot inspect retrieved documents, while output monitors cannot prevent malicious payloads from reaching the model. Consequently, retrieval-augmented generation (RAG) chatbots remain vulnerable to indirect injection, where a poisoned…

Read at arXiv (research papers)

arXiv (research papers) J Alex Corll SI

Amplify, Don't Create: Temporal Accumulation for Slow-Burn Prompt Injection

Most prompt-injection detectors score a single event or message. Control-plane attacks against tool-using agents can instead distribute weak directives across a trajectory while keeping each event below threshold. We test whether a proxy-side temporal accumulator recovers this slow-burn signal by reducing frozen per-event scores to peak and CUSUM persistence statistics. To avoid circularity, grafts are generated against a held-out autoregressive cloaking target and then…

Read at arXiv (research papers)

arXiv (research papers) Zunchen Huang SI

Analyzing the Narration Gap in LLM-Solver Loops

Formal tools such as SAT and SMT solvers are increasingly embedded in language model reasoning pipelines when a safety or security critical question can be formulated in logic. Unlike chain of thought whose steps are sampled from the model distribution without formal guarantee, a solver produces a sound and independently verifiable answer. However, the soundness guarantee can be lost in the interaction between the solver and the model. The hybrid pipeline has three…

Read at arXiv (research papers)

arXiv (research papers) Po-Han Cheng SI

CodeSentinel: A Three-Layer Defense Against Indirect Prompt Injection in Code Contexts

Code large language models increasingly retrieve external code context from repositories, documentation, issue threads, and coding-agent environments, creating an indirect prompt-injection surface where attackers hide instructions in comments, strings, identifiers, or decoy code. We propose CodeSentinel, a three-layer inference-time sanitizer. It uses Tree-sitter to extract high-risk model-facing CST nodes, then combines syntax-guided pre-filtering, CST-guided Dynamic…

Read at arXiv (research papers)

arXiv (research papers) Laxmipriya Ganesh Iyer SI

The Gate Is Only as Honest as Its Contracts: ContractGuard for the Contract Layer of Risk-Aware Causal Gating

Risk-Aware Causal Gating (RACG) defends tool-augmented LLM agents against indirect prompt injection by removing dangerous tools from the agent's visible action space, so that even a fully injection-compliant agent cannot call a tool it cannot see. We make three points. First, this structural guarantee does not eliminate the trust assumption behind safe tool use; it relocates it into the integrity of the tool contracts -- declared preconditions, effects, risk, and…

Read at arXiv (research papers)

arXiv (research papers) Yuchuan Tian SI

SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents

Tool-using language-model agents introduce security failures that go beyond unsafe text: they can disclose protected objects, write persistent memory, send messages, modify databases, or trigger harmful code and tool effects. Existing evaluations often collapse these stages into a single attack success rate, making it difficult to tell whether a model merely agreed with an attacker or actually produced observable harm. We introduce SafeClawBench, a staged benchmark for…

Read at arXiv (research papers)

arXiv (research papers) Xuelong Dai SI

MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents

Multimodal Large Language Model (MLLM)-based web agents provide practical, high-precision solutions for visual browser automation; however, they inherently expand the attack surface, introducing novel vision-based vulnerabilities. Existing adversarial evaluations targeting these agents frequently rely on permissive threat models and visually conspicuous artifacts. In this paper, we investigate a constrained vulnerability detection setting: a trusted web platform where the…

Read at arXiv (research papers)

arXiv (research papers) Aaditya Pai SI

PARSE: Provenance-Aware Retrieval Sanitization for Professional Domain LLM Agents

Prompt injection defenses evaluated on synthetic benchmarks do not generalize to real enterprise documents, which are longer, denser, and interleave legitimate authority language with factual content. We demonstrate this gap with a benchmark of 122 tasks across five professional domains (financial, legal, medical, scientific, DevOps) built on real retrieved documents -- actual SEC filings, Federal Register rules, PubMed abstracts, arXiv papers, and GitHub postmortems…

Read at arXiv (research papers)

arXiv (research papers) Hankyul Baek SI

An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios

AI agents are increasingly being adopted in enterprise and personal settings with access to emails, databases, documents, and other tools where they can read, update, and disseminate sensitive information. Much of prior research on data leakage risks in agents has focused on adversarial data exfiltration through prompt injections and jailbreaks. However, sensitive information may also be exposed during non-adversarial use, creating leakage risks even when users issue benign…

Read at arXiv (research papers)

arXiv (research papers) David Huang SI

Rapid Poison: Practical Poisoning Attacks Against the Rapid Response Framework

The Rapid Response (RR) framework, deployed in production systems, including Anthropic's ASL-3 safeguards, continuously improves jailbreak-detection classifiers. When new jailbreaks emerge that bypass these classifiers, Rapid Response generates synthetic variants for training, helping the model generalize from the new attacks and quickly adapt. We reveal that prompt injection can infiltrate this pipeline to deliver poisoned samples into the classifier's training set…

Read at arXiv (research papers)

arXiv (research papers) Ismail Hossain SI

SkillVetBench: LLM-as-Judge for Multi-Dimensional Security Risk Evaluation in Open-Source LLM Agent Skills

Open-source LLM agent ecosystems are growing rapidly, yet the security of community-contributed skills - modular tool definitions that extend agent capabilities - remains largely unvetted. The gap we fill: existing scanners operate at the code layer and are structurally blind to instruction-layer and multi-agent risk - natural-language directives that hijack an agent, exfiltrate data through encoded side channels, or chain harm across pipelines - so what is needed is a…

Read at arXiv (research papers)

arXiv (research papers) Aman Anifer SI

GAS-Leak-LLM: Genetic Algorithm-Based Suffix Optimization for Black-Box LLM Jailbreaking

Large Language Models (LLMs) constitute pivotal components within the AI-dominated information technology ecosystem. To mitigate risks associated with harmful or policy-violating outputs, commercial systems employ advanced alignment strategies and multi-layered content moderation mechanisms. Despite these safeguards, recent research has demonstrated that LLMs remain vulnerable to adversarial manipulation, particularly through jailbreaking and prompt injection techniques. In…

Read at arXiv (research papers)

arXiv (research papers) Zixin Rao SI

FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion

Large language model (LLM) agents increasingly rely on long-term memory to support complex task execution, user personalization, and domain adaptation. Meanwhile, emerging access-control mechanisms for LLM agents are being explored to block policy-violating requests and prevent misuse. We reveal a novel attack surface arising from agent memory operations: prohibited content that would trigger access control can be fragmented across interactions, stored in long-term memory in…

Read at arXiv (research papers)

arXiv (research papers) Lipeng He SI

Defending against Adaptive Prompt Injection Attacks via Reasoning-enabled Task Alignment

Indirect prompt injection attacks hijack LLM-based agents by embedding malicious instructions in third-party data that the agent retrieves during task execution. Existing defenses report near-zero attack success rate on static benchmarks, yet recent adaptive evaluations show that these results collapse once the attacker is allowed to optimize against the deployed defense. In this work, we trace this collapse to two failure modes. First, existing defense methods are confined…

Read at arXiv (research papers)

arXiv (research papers) Zhongye Liu SI

Forced Deferral: Manipulating Routing Decisions in Multimodal LLM Cascades

While multimodal large language models (MLLMs) have shown strong visual reasoning abilities, serving a large model for every query is computationally expensive. MLLM cascades mitigate this cost by first querying a weak but cheaper model and deferring to a strong model when the weak model's output is unconfident. However, since the weak model's confidence directly controls compute allocation, these systems expose a new attack surface: an adversary can manipulate confidence so…

Read at arXiv (research papers)

arXiv (research papers) Xinhang Ma SI

AutoDojo: A Generative Benchmark for Evaluating Prompt Injection Defenses in LLM Agents

Indirect prompt injection (IPI) is a major security threat to LLM-powered agents. Thus, a growing body of work has proposed a variety of defensive approaches against IPI which can be grouped into three broad categories: prompt-level, filter-based, and system-level designs. However, commonly used benchmarks for evaluating defense, such as AgentDojo, are \emph{inherently static}, generating a fixed distribution of IPI attacks. Consequently, a defense can score well on them…

Read at arXiv (research papers)

arXiv (research papers) Saeid Jamshidi SI

Security Engineering of OpenClaw: Analyzing Attack Surface Expansion and Trust-Boundary Violations

Agentic large language model (LLM) systems can now execute actions, not only produce text. When model outputs trigger privileged operations such as shell commands, browser automation, or external tool calls, the security problem shifts from alignment alone to system configuration and structural design. We analyze OpenClaw, a self-hosted multi-agent system in which LLM outputs can execute commands and interact with tools and services. We measure compromise probability…

Read at arXiv (research papers)

arXiv (research papers) Yuguang Zhou SI

From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails

LLM-based guardrails have emerged as a highly effective defense against prompt injection and jailbreak attacks in autonomous agents. However, we reveal that the very reasoning and task-following capabilities enabling this protection introduce a novel vulnerability: attackers can inject crafted data to trap the guardrail in extended reasoning loops, effectuating a systematic denial-of-service (DoS) attack. To systematically expose this threat, we design a beam-search…

Read at arXiv (research papers)

arXiv (research papers) Youngduk Kim SI

SkillMutator: Benchmarking and Defending Language-and-Code Cross-modal Attacks on LLM Agent Skills

Large language model (LLM) agents increasingly extend their capabilities at runtime by loading Agent Skills, which pair natural-language specifications (SKILL.md) with executable scripts and resources. Because a skill's behavior relies on both natural-language instructions and executable code, assessing its safety requires cross-modal reasoning, creating a new language-and-code attack surface. Attackers can present a benign workflow in SKILL.md while embedding implicit…

Read at arXiv (research papers)

arXiv (research papers) Zihao Wang SI

Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents

LLM-based web agents are increasingly deployed in real-world settings such as e-commerce, where they interact extensively with untrusted web content while executing actions that carry direct financial consequences. This makes them vulnerable to prompt-injection attacks, in which seemingly benign web content conceals adversarial instructions that manipulate the agent's behavior. Existing security benchmarks adopt an \textit{attack-centric} perspective, focusing on the…

Read at arXiv (research papers)

arXiv (research papers) Xu Yang SI

No Hidden Prompts Needed! You Can Game AI Peer Review with Presentation-Only Revisions

As AI-generated reviews move from experimental tools into peer-review infrastructure, most robustness concerns have focused on explicit attacks such as hidden instructions and prompt injection. We study a harder and more policy-relevant failure mode: no hidden text, no prompt injection, and no changes to methods, experiments, figures, equations, proofs, or numerical results. The attacker modifies only presentation-level content, such as the abstract, contribution framing…

Read at arXiv (research papers)

arXiv (research papers) Haowei Qian SI

Nous: An Attempt to Extract and Inject the Cognition Behind Prediction-Market Behavior

As LLM agents proliferate in prediction markets and collective decision-making, they risk a cognitive monoculture: agents built on shared foundation models produce correlated forecasts, and recent measurement finds frontier-model errors correlated at r ~ 0.77. We ask whether human cognitive diversity can be recovered from behavior and transferred to LLM agents. Nous extracts a structured eight-dimension behavioral profile from real Polymarket trading activity and injects it…

Read at arXiv (research papers)

arXiv (research papers) Pengfei He SI

PI-Hunter: Automated Red-Teaming for Exposing and Localizing Prompt Injections

Large Language Models (LLMs) are rapidly evolving into agentic systems that interact with external tools and environments, introducing new security risks such as indirect prompt injection attacks through untrusted external sources. Existing defenses mainly focus on blocking malicious content at inference time, and current red-teaming methods primarily optimize attack success. As a result, developers have limited visibility into how latent prompt injections emerge and…

Read at arXiv (research papers)

arXiv (research papers) Xinyu Zhao SI

Does AI Reviewer See the Full Picture? Attacking and Defending Multimodal Peer Review

The integration of Large Language Models (LLMs) and Multimodal LLMs (MLLMs) into scientific peer-review workflows introduces novel and significant risks for adversarial manipulation, especially given the multimodal nature of scientific papers where figures, not just text, convey core evidence. This creates a significant gap: current robustness studies on AI peer-review are overwhelmingly text-only. Moreover, the problem is distinct from standard jailbreaking, as a…

Read at arXiv (research papers)

Older stories