BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems

• URL: http://arxiv.org/abs/2505.15216v2
• Date: Thu, 10 Jul 2025 02:10:30 GMT
• Title: BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems
• Abstract summary: We introduce the first framework to capture offensive and defensive cyber-capabilities in evolving real-world systems.<n>To capture the vulnerability lifecycle, we define three task types: Detect (detecting a new vulnerability), Exploit (exploiting a specific vulnerability), and Patch (patching a specific vulnerability)<n>We evaluate 8 agents: Claude Code, OpenAI Codex CLI with o3-high and o4-mini, and custom agents with o3-high, GPT-4.1, Gemini 2.5 Pro Preview, Claude 3.7 Sonnet Thinking, and DeepSeek-R1.
• Score: 62.17474934536671
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: AI agents have the potential to significantly alter the cybersecurity landscape. Here, we introduce the first framework to capture offensive and defensive cyber-capabilities in evolving real-world systems. Instantiating this framework with BountyBench, we set up 25 systems with complex, real-world codebases. To capture the vulnerability lifecycle, we define three task types: Detect (detecting a new vulnerability), Exploit (exploiting a specific vulnerability), and Patch (patching a specific vulnerability). For Detect, we construct a new success indicator, which is general across vulnerability types and provides localized evaluation. We manually set up the environment for each system, including installing packages, setting up server(s), and hydrating database(s). We add 40 bug bounties, which are vulnerabilities with monetary awards of \$10-\$30,485, covering 9 of the OWASP Top 10 Risks. To modulate task difficulty, we devise a new strategy based on information to guide detection, interpolating from identifying a zero day to exploiting a specific vulnerability. We evaluate 8 agents: Claude Code, OpenAI Codex CLI with o3-high and o4-mini, and custom agents with o3-high, GPT-4.1, Gemini 2.5 Pro Preview, Claude 3.7 Sonnet Thinking, and DeepSeek-R1. Given up to three attempts, the top-performing agents are OpenAI Codex CLI: o3-high (12.5% on Detect, mapping to \$3,720; 90% on Patch, mapping to \$14,152), Custom Agent with Claude 3.7 Sonnet Thinking (67.5% on Exploit), and OpenAI Codex CLI: o4-mini (90% on Patch, mapping to \$14,422). OpenAI Codex CLI: o3-high, OpenAI Codex CLI: o4-mini, and Claude Code are more capable at defense, achieving higher Patch scores of 90%, 90%, and 87.5%, compared to Exploit scores of 47.5%, 32.5%, and 57.5% respectively; while the custom agents are relatively balanced between offense and defense, achieving Exploit scores of 37.5-67.5% and Patch scores of 35-60%.

Related papers

• Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens [65.53494487819053]
  Claw-like AI agents (e.g., OpenClaw) are always-on processes with persistent access to credentials, files, tools, and external services.<n>We adopt a computer-system analogy: treating a Claw-like agent as an agentic computer system whose gateway runtime plays an OS-like mediation role.<n>We develop SafeClawArena, a benchmark of 406 adversarial tasks across four attack surfaces.
  arXiv  Detail & Related papers  (2026-06-29T18:00:45Z)
• CyberChainBench: Can AI Agents Secure Smart Contracts Against Real-World On-Chain Vulnerabilities? [28.049089946230513]
  CyberChainBench is a benchmark for evaluating LLM-based agents on smart contract security.<n>Built from 541 real-world exploit incidents from DeFiHackLabs, the benchmark provides end-to-end on-chain evaluation.<n>Results reveal a clear difficulty gradient: the best configuration scores 37.5% on detection, 43.7% on exploitation, but only 23.4% on patching.
  arXiv  Detail & Related papers  (2026-06-24T17:58:16Z)
• VisualClaw: A Real-Time, Personalized Agent for the Physical World [76.89211120690028]
  We present VisualClaw, a self-evolving multimodal agent built around two principles.<n>First, hybrid encoding reduces deployment cost by filtering less informative streaming frames with a cascaded gate.<n>Second, skill evolution lets the agent learn from failures, producing skill-bank updates that help future questions.
  arXiv  Detail & Related papers  (2026-06-15T06:58:22Z)
• ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks? [92.21756459993695]
  Exploitation is a challenging task because it requires low-level program reasoning.<n>Despite its importance and diagnostic value, exploitation remains under-evaluated.<n>We introduce ExploitGym, a large-scale, diverse, realistic benchmark on the exploitation capabilities of AI agents.
  arXiv  Detail & Related papers  (2026-05-11T18:00:14Z)
• Reward Hacking Benchmark: Measuring Exploits in LLM Agents with Tool Use [0.016201337438609787]
  Reinforcement learning (RL) trained language model agents with tool access are increasingly deployed in coding assistants, research tools, and autonomous systems.<n>We introduce the Reward Hacking Benchmark (RHB), a suite of multi-step tasks requiring sequential tool operations with naturalistic shortcut opportunities.<n>We evaluate 13 frontier models from OpenAI, Anthropic, Google, and DeepSeek.
  arXiv  Detail & Related papers  (2026-05-03T07:10:42Z)
• Your Agent, Their Asset: A Real-World Safety Analysis of OpenClaw [87.97230960702274]
  We present the first real-world safety evaluation of OpenClaw.<n>We introduce the CIK taxonomy, which unifies an agent's persistent state into three dimensions.<n>Our evaluations cover 12 attack scenarios on a live OpenClaw instance.
  arXiv  Detail & Related papers  (2026-04-06T15:27:05Z)
• VibeGuard: A Security Gate Framework for AI-Generated Code [0.11080037957254413]
  Anthropic's Claude Code CLI shipped a 59.8 MB source map file in its npm package exposing 512,000 lines of proprietary TypeScript.<n>We present VibeGuard, a pre-publish security gate that targets five such blind spots.<n>We discuss how these results inform a defense-in-depth workflow for teams that rely on AI code generation.
  arXiv  Detail & Related papers  (2026-04-01T15:57:01Z)
• Malicious Agent Skills in the Wild: A Large-Scale Security Empirical Study [47.60135753021306]
  Third-party agent skills extend LLM-based agents with instruction files and executable code that run on users' machines.<n>No ground-truth dataset exists to characterize the resulting threats.<n>We construct the first labeled dataset of malicious agent skills by behaviorally verifying 98,380 skills.
  arXiv  Detail & Related papers  (2026-02-06T09:52:27Z)
• CVE-Factory: Scaling Expert-Level Agentic Tasks for Code Security Vulnerability [50.57373283154859]
  We present CVE-Factory, the first multiagent framework to achieve expert-level quality in automatically transforming vulnerability tasks.<n>It is also evaluated on the latest realistic vulnerabilities and achieves a 66.2% verified success.<n>We synthesize over 1,000 executable training environments, the first large-scale scaling of agentic tasks in code security.
  arXiv  Detail & Related papers  (2026-02-03T02:27:16Z)
• Penetration Testing of Agentic AI: A Comparative Security Analysis Across Models and Frameworks [0.0]
  Agentic AI introduces security vulnerabilities that traditional LLM safeguards fail to address.<n>We conduct the first systematic testing and comparative evaluation of agentic AI systems.<n>We identify six distinct defensive behavior patterns including a novel "hallucinated compliance" strategy.
  arXiv  Detail & Related papers  (2025-12-16T19:22:50Z)
• The Trojan Knowledge: Bypassing Commercial LLM Guardrails via Harmless Prompt Weaving and Adaptive Tree Search [58.8834056209347]
  Large language models (LLMs) remain vulnerable to jailbreak attacks that bypass safety guardrails to elicit harmful outputs.<n>We introduce the Correlated Knowledge Attack Agent (CKA-Agent), a dynamic framework that reframes jailbreaking as an adaptive, tree-structured exploration of the target model's knowledge base.
  arXiv  Detail & Related papers  (2025-12-01T07:05:23Z)
• Cuckoo Attack: Stealthy and Persistent Attacks Against AI-IDE [64.47951172662745]
  Cuckoo Attack is a novel attack that achieves stealthy and persistent command execution by embedding malicious payloads into configuration files.<n>We formalize our attack paradigm into two stages, including initial infection and persistence.<n>We contribute seven actionable checkpoints for vendors to evaluate their product security.
  arXiv  Detail & Related papers  (2025-09-19T04:10:52Z)
• Multi-Agent Penetration Testing AI for the Web [3.93181912653522]
  MAPTA is a multi-agent system for autonomous web application security assessment.<n>It combines large language model orchestration with tool-grounded execution and end-to-end exploit validation.<n>On the 104-challenge XBOW benchmark, MAPTA achieves 76.9% overall success.
  arXiv  Detail & Related papers  (2025-08-28T14:14:24Z)
• Measuring Harmfulness of Computer-Using Agents [7.662513862243521]
  Computer-using agents (CUAs) autonomously control computers to perform multi-step actions.<n> CUAHarm consists of 104 expert-written realistic misuse risks.<n>We evaluate open-source and proprietary LMs, such as Claude Sonnet, GPT-4o, Gemini Pro 1.5, Llama-3.3-70B, and Mistral Large 2.
  arXiv  Detail & Related papers  (2025-07-31T07:02:19Z)
• OS-Harm: A Benchmark for Measuring Safety of Computer Use Agents [60.78202583483591]
  We introduce OS-Harm, a new benchmark for measuring safety of computer use agents.<n> OS-Harm is built on top of the OSWorld environment and aims to test models across three categories of harm: deliberate user misuse, prompt injection attacks, and model misbehavior.<n>We evaluate computer use agents based on a range of frontier models and provide insights into their safety.
  arXiv  Detail & Related papers  (2025-06-17T17:59:31Z)
• CyberGym: Evaluating AI Agents' Cybersecurity Capabilities with Real-World Vulnerabilities at Scale [46.76144797837242]
  Large language model (LLM) agents are becoming increasingly skilled at handling cybersecurity tasks autonomously.<n>Existing benchmarks fall short, often failing to capture real-world scenarios or being limited in scope.<n>We introduce CyberGym, a large-scale and high-quality cybersecurity evaluation framework featuring 1,507 real-world vulnerabilities.
  arXiv  Detail & Related papers  (2025-06-03T07:35:14Z)
• Stealthy Patch-Wise Backdoor Attack in 3D Point Cloud via Curvature Awareness [52.07366900097567]
  Backdoor attacks pose a severe threat to deep neural networks (DNNs)<n>Existing 3D point cloud backdoor attacks rely on sample-wise global modifications, which suffer from low imperceptibility.<n>We propose the Stealthy Patch-Wise Backdoor Attack (SPBA), the first patch-wise backdoor attack framework for 3D point clouds.
  arXiv  Detail & Related papers  (2025-03-12T12:30:59Z)
• Fine-Grained 1-Day Vulnerability Detection in Binaries via Patch Code Localization [12.73365645156957]
  1-day vulnerabilities in binaries have become a major threat to software security.<n>patch presence test is one of the effective ways to detect the vulnerability.<n>We propose a novel approach named PLocator, which leverages stable values from both the patch code and its context.
  arXiv  Detail & Related papers  (2025-01-29T04:35:37Z)
• There are More Fish in the Sea: Automated Vulnerability Repair via Binary Templates [4.907610470063863]
  We propose a template-based automated vulnerability repair approach for Java binaries.<n>Experiments on the Vul4J dataset demonstrate that TemVUR successfully repairs 11 vulnerabilities.<n>To assess the generalizability of TemVUR, we curate the ManyVuls4J dataset.
  arXiv  Detail & Related papers  (2024-11-27T06:59:45Z)
• RedCode: Risky Code Execution and Generation Benchmark for Code Agents [50.81206098588923]
  RedCode is a benchmark for risky code execution and generation. RedCode-Exec provides challenging prompts that could lead to risky code execution. RedCode-Gen provides 160 prompts with function signatures and docstrings as input to assess whether code agents will follow instructions.
  arXiv  Detail & Related papers  (2024-11-12T13:30:06Z)
• HonestCyberEval: An AI Cyber Risk Benchmark for Automated Software Exploitation [5.748110232428846]
  We introduce HonestCyberEval, a new benchmark for assessing AI models' capabilities and risks in automated software exploitation.<n>We assess several leading language models, including OpenAI's GPT-4.5, o3-mini, o1 and o1-mini.<n>Our findings reveal that these models vary significantly in their success rates and efficiency.
  arXiv  Detail & Related papers  (2024-10-29T10:57:11Z)
• Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models [33.1538965735133]
  Cybench is a framework for specifying cybersecurity tasks and evaluating agents on those tasks.<n>We include 40 professional-level Capture the Flag (CTF) tasks from 4 distinct CTF competitions.<n>We construct a cybersecurity agent and evaluate 8 models: GPT-4o, OpenAI o1-preview, Claude 3 Opus, Claude 3.5 Sonnet, Mixtral 8x22b Instruct, Gemini 1.5 Pro, Llama 3 70B Chat, and Llama 3.1 405B Instruct.
  arXiv  Detail & Related papers  (2024-08-15T17:23:10Z)
• On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents [58.79302663733703]
  Large language model-based multi-agent systems have shown great abilities across various tasks due to the collaboration of expert agents.<n>However, the impact of clumsy or even malicious agents, on the overall performance of the system remains underexplored.<n>This paper investigates what is the resilience of various system structures under faulty agents.
  arXiv  Detail & Related papers  (2024-08-02T03:25:20Z)
• PatchFinder: A Two-Phase Approach to Security Patch Tracing for Disclosed Vulnerabilities in Open-Source Software [15.867607171943698]
  We propose a two-phase framework with end-to-end correlation learning for better-tracing security patches. PatchFinder achieves a Recall@10 of 80.63% and a Mean Reciprocal Rank (MRR) of 0.7951. When applying PatchFinder in practice, we initially identified 533 patch commits and submitted them to the official, 482 of which have been confirmed by CVE Numbering Authorities.
  arXiv  Detail & Related papers  (2024-07-24T07:46:24Z)
• 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)
• (De)Randomized Smoothing for Certifiable Defense against Patch Attacks [136.79415677706612]
  We introduce a certifiable defense against patch attacks that guarantees for a given image and patch attack size. Our method is related to the broad class of randomized smoothing robustness schemes. Our results effectively establish a new state-of-the-art of certifiable defense against patch attacks on CIFAR-10 and ImageNet.
  arXiv  Detail & Related papers  (2020-02-25T08:39:46Z)

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.