Evaluating LLMs for Real-World Web Vulnerability Detection

• URL: http://arxiv.org/abs/2606.21397v1
• Date: Fri, 19 Jun 2026 13:02:47 GMT
• Title: Evaluating LLMs for Real-World Web Vulnerability Detection
• Abstract summary: Large Language Models (LLMs) have emerged as a promising tool for automated vulnerability detection.<n>This work benchmarks six frontier (Claude Opus 4.6, Codex GPT-5.4, Gemini 3.1-pro-preview) and open-weight models (Qwen 3.5, Qwen 3 Coder Next, MiniMax M2.5) on their ability to detect real-world web vulnerabilities.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) have emerged as a promising tool for automated vulnerability detection, yet their effectiveness on web-specific vulnerabilities remains to be explored. This work benchmarks six frontier (Claude Opus 4.6, Codex GPT-5.4, Gemini 3.1-pro-preview) and open-weight models (Qwen 3.5, Qwen 3 Coder Next, MiniMax M2.5) on their ability to detect real-world web vulnerabilities using static analysis in WordPress plugins, including SQL injection, stored cross-site scripting, path traversal, and remote code execution. Using five prompt designs of varying structure, scope, and complexity across three experiment iterations, we aim to answer how model and prompt choice affects vulnerability detection. Our results show that all models are capable of detecting valid security issues, but the detection rate varies depending on the model and prompt. For example, Claude Opus 4.6 achieved the highest web vulnerability detection rate (63%), while open-weight MiniMax M2.5 performs on par with other frontier models (48%), and self-hosted Qwen 3.5 only achieved 35%. We show that scoped prompts that narrow the vulnerability scope outperform open-ended ones, whereas the prompt complexity has little impact. Surprisingly, no model achieved full reporting consistency across three experiment iterations, with some as low as 50%. Our experiments demonstrate the opportunities and limits of LLM-based vulnerability detection, as no model correctly identified one baseline vulnerability in one of the plugins. Additionally, we derive practical lessons learned for security practitioners and publish all code and data to support future research.

Related papers

• Exploring Automated Vulnerability Identification in JavaScript Code Using Large Language Models [9.103541014019232]
  JavaScript powers approximately 98.8% of all websites.<n> Static Application Security Testing tools often fail to identify many real-world vulnerabilities when applied to isolated code snippets.<n>This paper presents an empirical study of Large Language Model (LLM)-based vulnerability identification for JavaScript programs.
  arXiv  Detail & Related papers  (2026-09-12T08:59:33Z)
• Are Frontier LLMs Ready for Cybersecurity? Evidence for Vertical Foundation Models from Dual-Mode Vulnerability Benchmarks [0.3303672705634661]
  We evaluate whether frontier LLMs are ready for cybersecurity through a dual-mode benchmark.<n>We test six frontier models (GPT-5.4, Codex5.3, Claude Opus4.6, Sonnet4.6, Gemini3.1Pro and Gemini3Flash) and two domain-specialized models across four testing paradigms.
  arXiv  Detail & Related papers  (2026-05-22T05:24:43Z)
• Vulnerability Detection with Interprocedural Context in Multiple Languages: Assessing Effectiveness and Cost of Modern LLMs [4.077363992854019]
  Large Language Models (LLMs) have been a promising way for automated vulnerability detection.<n>This study investigates the effectiveness of detection, the inference cost, and the quality of explanations in detecting vulnerabilities related to interprocedural dependencies.<n>The results show that Gemini 3 Flash offers the best cost-effectiveness trade-off for C vulnerabilities.
  arXiv  Detail & Related papers  (2026-04-09T16:17:58Z)
• T-MAP: Red-Teaming LLM Agents with Trajectory-aware Evolutionary Search [49.99459363244884]
  We propose a trajectory-aware evolutionary search method, T-MAP, which leverages execution trajectories to guide the discovery of adversarial prompts.<n>Our approach enables the automatic generation of attacks that not only bypass safety guardrails but also reliably realize harmful objectives through actual tool interactions.
  arXiv  Detail & Related papers  (2026-03-21T12:33:34Z)
• AWE: Adaptive Agents for Dynamic Web Penetration Testing [0.0]
  AWE is a memory-augmented multi-agent framework for autonomous web penetration testing.<n>It embeds structured, vulnerability-specific analysis pipelines within a lightweight LLM orchestration layer.<n>AWE achieves substantial gains on injection-class vulnerabilities.
  arXiv  Detail & Related papers  (2026-03-01T07:32:42Z)
• Beyond Single Bugs: Benchmarking Large Language Models for Multi-Vulnerability Detection [1.2802720336459552]
  We introduce a benchmark for Multi-Vulnerability Detection across four major languages: C, C++, Python, and JavaScript.<n>We construct a dataset of 40,000 files by injecting controlled counts of vulnerabilities into long-context code samples.<n>Our results reveal a sharp degradation in performance as vulnerability density increases.
  arXiv  Detail & Related papers  (2025-12-26T05:43:35Z)
• Ensembling Large Language Models for Code Vulnerability Detection: An Empirical Evaluation [69.8237598448941]
  This study investigates the potential of ensemble learning to enhance the performance of Large Language Models (LLMs) in source code vulnerability detection.<n>We propose Dynamic Gated Stacking (DGS), a Stacking variant tailored for vulnerability detection.
  arXiv  Detail & Related papers  (2025-09-16T03:48:22Z)
• VulAgent: Hypothesis-Validation based Multi-Agent Vulnerability Detection [55.957275374847484]
  VulAgent is a multi-agent vulnerability detection framework based on hypothesis validation.<n>It implements a semantics-sensitive, multi-view detection pipeline, each aligned to a specific analysis perspective.<n>On average, VulAgent improves overall accuracy by 6.6%, increases the correct identification rate of vulnerable--fixed code pairs by up to 450%, and reduces the false positive rate by about 36%.
  arXiv  Detail & Related papers  (2025-09-15T02:25:38Z)
• OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety [58.201189860217724]
  We introduce OpenAgentSafety, a comprehensive framework for evaluating agent behavior across eight critical risk categories.<n>Unlike prior work, our framework evaluates agents that interact with real tools, including web browsers, code execution environments, file systems, bash shells, and messaging platforms.<n>It combines rule-based analysis with LLM-as-judge assessments to detect both overt and subtle unsafe behaviors.
  arXiv  Detail & Related papers  (2025-07-08T16:18:54Z)
• Backdoor Cleaning without External Guidance in MLLM Fine-tuning [76.82121084745785]
  Believe Your Eyes (BYE) is a data filtering framework that leverages attention entropy patterns as self-supervised signals to identify and filter backdoor samples.<n>It achieves near-zero attack success rates while maintaining clean-task performance.
  arXiv  Detail & Related papers  (2025-05-22T17:11:58Z)
• Large Language Models for In-File Vulnerability Localization Can Be "Lost in the End" [6.6389862916575275]
  New development practice requires researchers to investigate whether commonly used LLMs can effectively analyze large file-sized inputs.<n>This paper is to evaluate the effectiveness of several state-of-the-art chat-based LLMs, including the GPT models, in detecting in-file vulnerabilities.
  arXiv  Detail & Related papers  (2025-02-09T14:51:15Z)
• VELVET: a noVel Ensemble Learning approach to automatically locate VulnErable sTatements [62.93814803258067]
  This paper presents VELVET, a novel ensemble learning approach to locate vulnerable statements in source code. Our model combines graph-based and sequence-based neural networks to successfully capture the local and global context of a program graph. VELVET achieves 99.6% and 43.6% top-1 accuracy over synthetic data and real-world data, respectively.
  arXiv  Detail & Related papers  (2021-12-20T22:45:27Z)

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.