SIR: Self-improving Red-teaming for Compute Use Agents

• URL: http://arxiv.org/abs/2608.30207v1
• Date: Mon, 31 Aug 2026 03:39:43 GMT
• Title: SIR: Self-improving Red-teaming for Compute Use Agents
• Abstract summary: Computer use agents (CUAs) are vision-language models that perceive a screen and act on a real operating system through mouse, keyboard, and terminal.<n>Because they can be exposed to untrusted content while operating, they are vulnerable to indirect prompt injection (IPI)<n>We present SIR, a black box IPI attack that composes stealthy injections from a small library of reusable principles stated in plain language.
• Score: 71.71987044117371
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Computer use agents (CUAs) are vision-language models that perceive a screen and act on a real operating system through mouse, keyboard, and terminal, and they are increasingly deployed to automate everyday digital tasks. Because they can be exposed to untrusted content while operating, they are vulnerable to indirect prompt injection (IPI), in which an adversary plants instructions in content the agent will read and redirects it toward actions that violate the user's intent. Existing CUA safety benchmarks evaluate fixed injections written by hand, which may underestimate the risk posed by an adaptive adversary. We present SIR, a black box IPI attack that (i) composes stealthy injections from a small library of reusable principles stated in plain language and (ii) wraps composition in an iterative feedback loop that diagnoses the victim's failed trajectories and distills the bypasses into new, named strategies that are reapplied across tasks. Unlike prior red teaming of web agents, we target CUAs at the operating system level and score attacks with a fully deterministic oracle, using checks on filesystem, service, and permission state rather than an LLM judge. On experiment, we evaluate three frontier CUAs. Composing principles with feedback raises the attack success rate over a baseline written by hand, for example from 4% to 24% on Claude Opus 4.8 and from 0% to 28% on Gemini 3.5 Flash, while the benign task still completes. Principles discovered against one model further transfer to a different architecture with no additional feedback.

Related papers

• Beyond the Payload: How User Invocation Shapes Coding Agent Vulnerability to Repository Poisoning [39.57248170050304]
  Coding agents are increasingly used for software engineering tasks, including bootstrapping projects from third-party repositories.<n>We term these user-side choices Prompt-Level configurations (PLCs) and introduce CIPR (Coding In Poisoned Repos)<n> CIPR comprises 1,920 instances across 20 repositories, four task types, three social-media-grounded prompt styles, and three skill/rule conditions, and measures attack success rate (ASR) and agent alert rate (AR)<n>Vulnerability is highly context-dependent, with task type creating up to a 4.5-fold difference in ASR, with test-
  arXiv  Detail & Related papers  (2026-08-31T12:26:01Z)
• Invisible Ink Threats: Adversarial Goals Behind Legitimate Tasks in Computer-Use Agents [10.553545687886928]
  We present II-Bench, a collection of seemingly harmless adversarial tasks.<n>II-Bench comprises 444 examples targeting confidentiality and integrity attacks across three platforms.<n>We also construct HITLCUA, a comprehensive adversarial testing framework.
  arXiv  Detail & Related papers  (2026-08-03T10:16:53Z)
• Beyond Attack-Success Rate: Action-Graded Severity Scale for Tool-Using AI Agents [0.0]
  Agentic red-teaming benchmarks report whether an injected agent was compromised as a single bit: the attack succeeded, or it did not.<n>We argue that this binary attack-success rate discards the information a defender most needs, namely how harmful the resulting action was.<n>We introduce an action-graded harm rubric that scores an agent's tool-call trajectory on a seven-level ordinal scale (L0 to L6) according to whether the executed action was reversible, whether it crossed scope to reach another party, and whether it expanded privilege.
  arXiv  Detail & Related papers  (2026-07-08T14:38:01Z)
• Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents [93.19140872946842]
  Web agents driven by large language models (LLMs) are increasingly deployed in real-world environments.<n>This makes them vulnerable to prompt-injection attacks, in which seemingly benign content embeds adversarial instructions that manipulate agent behaviour.<n>We introduce textbfsysname, a benchmark to systematically categorize and attribute harm in real-world web agent systems.
  arXiv  Detail & Related papers  (2026-06-11T14:12:43Z)
• Trojan's Whisper: Stealthy Manipulation of OpenClaw through Injected Bootstrapped Guidance [23.059379933610163]
  Guidance injection is a stealthy attack vector that embeds adversarial operational narratives into bootstrap guidance files.<n>We construct 26 malicious skills spanning 13 attack categories including exfiltration, workspace destruction, privilege escalation, and persistent backdoor installation.<n>Our attacks achieve success rates from 16.4% to 64.2%, with the majority of malicious actions executed autonomously without user confirmation.
  arXiv  Detail & Related papers  (2026-03-20T14:17:56Z)
• Measuring and Exploiting Confirmation Bias in LLM-Assisted Security Code Review [6.417595678110472]
  We study whether confirmation bias affects LLM-based vulnerability detection, and whether this failure mode can be exploited in software supply-chain attacks.<n>Study 1 quantifies confirmation bias through controlled experiments on 250 CVE vulnerability/patch pairs evaluated across four state-of-the-art models under five framing conditions for the review prompt.<n>Study 2 evaluates exploitability in practice mimicking adversarial pull requests that reintroduce known vulnerabilities while framed as security improvements or urgent functionality fixes via their pull request metadata.
  arXiv  Detail & Related papers  (2026-03-19T10:40:27Z)
• When Actions Go Off-Task: Detecting and Correcting Misaligned Actions in Computer-Use Agents [50.5814495434565]
  This work makes the first effort to define and study misaligned action detection in computer-use agents (CUAs)<n>We identify three common categories in real-world CUA deployment and construct MisActBench, a benchmark of realistic trajectories with human-annotated, action-level alignment labels.<n>We propose DeAction, a practical and universal guardrail that detects misaligned actions before execution and iteratively corrects them through structured feedback.
  arXiv  Detail & Related papers  (2026-02-09T18:41:15Z)
• Poison Once, Control Anywhere: Clean-Text Visual Backdoors in VLM-based Mobile Agents [54.35629963816521]
  This work introduces VIBMA, the first clean-text backdoor attack targeting VLM-based mobile agents.<n>The attack injects malicious behaviors into the model by modifying only the visual input.<n>We show that our attack achieves high success rates while preserving clean-task behavior.
  arXiv  Detail & Related papers  (2025-06-16T08:09:32Z)
• AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases [73.04652687616286]
  We propose AgentPoison, the first backdoor attack targeting generic and RAG-based LLM agents by poisoning their long-term memory or RAG knowledge base. Unlike conventional backdoor attacks, AgentPoison requires no additional model training or fine-tuning. On each agent, AgentPoison achieves an average attack success rate higher than 80% with minimal impact on benign performance.
  arXiv  Detail & Related papers  (2024-07-17T17:59:47Z)
• Dissecting Adversarial Robustness of Multimodal LM Agents [70.2077308846307]
  We manually create 200 targeted adversarial tasks and evaluation scripts in a realistic threat model on top of VisualWebArena.<n>We find that we can successfully break latest agents that use black-box frontier LMs, including those that perform reflection and tree search.<n>We also use ARE to rigorously evaluate how the robustness changes as new components are added.
  arXiv  Detail & Related papers  (2024-06-18T17:32:48Z)

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

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