Web application fuzzing faces significant challenges in achieving comprehensive test interface (attack surface) coverage, primarily due to the complexity of user interactions and dynamic website architectures. While web crawlers can automatically access and extract critical website information—including form fields and request parameters—which are essential for generating effective fuzzing test cases, current crawler technologies exhibit three primary limitations: (i) insufficient capabilities in analyzing page relationships and determining page states; (ii) lack of functionality-aware exploration capabilities, resulting in generated inputs with poor contextual relevance; (iii) generation of unstructured operation sequences that fail to execute effectively due to their incompatibility with state-based testing logic. To address these challenges, we propose CrawlMLLM, a framework using multi-modal large language models to simulate human web browsing. It includes three core components: page state mining, functionality analysis, and automatic operation generation. Evaluations show 163% code coverage improvements over SOTA work. When integrated with vulnerability audit tools, CrawlMLLM found 44 vulnerabilities in three vulnerable web applications versus 34 by the baseline. In six real-world applications, CrawlMLLM detected 20 vulnerabilities while the next best method found six.
Broken access control vulnerabilities pose significant security risks to the protected web interfaces of IoT devices, enabling adversaries to gain unauthorized access to sensitive configurations and even use them as stepping stones for attacking the intranet. Despite its ranking as the first in the latest OWASP Top 10, there remains a lack of effective methodologies to detect these vulnerabilities systematically. We present ACBreaker, a novel methodology powered by a large language model (LLM), to effectively identify broken access control vulnerabilities in the protected web interfaces of IoT devices. Our methodology consists of three stages. The initial stage transforms firmware code that exceeds the LLM context window into semantically intact code snippets. The second stage involves using an LLM to extract device-specific information from firmware code. The final stage integrates this information into the mutation-based fuzzer to improve fuzzing effectiveness and employ differential analysis to identify vulnerabilities. We evaluated ACBreaker across 11 IoT devices, analyzing 1,274,646 lines of code and discovering 39 previously unknown vulnerabilities. We further analyzed these vulnerabilities, categorizing them into three types that contribute to protected interface evasion, and provided mitigation suggestions. These vulnerabilities were responsibly disclosed to vendors, with CVE IDs assigned to those in six IoT devices.
Cross-site scripting attacks represent one of the major security threats facing web applications, with Stored XSS attacks becoming the predominant form. Compared to reflected XSS, stored XSS attack payloads exhibit temporal and spatial asynchrony between injection and execution, rendering traditional browserside defenses based on request–response differential analysis ineffective. This paper presents XSShield, the first detection framework that leverages a Large Language Model to understand JavaScript semantics to defend against Stored XSS attacks. Through a Prompt Optimizer based on gradient descent and UCB-R selection algorithms, and a Data Adaptor based on program dependence graphs, the framework achieves real-time and fine-grained code processing. Experimental evaluation shows that XSShield achieves 93% accuracy and an F1 score of 0.9266 on the GPT-4 model, improving accuracy by an average of 88.8% compared to existing solutions. The processing time, excluding model communication overhead, averages only 0.205 s, demonstrating practical deployability without significantly impacting user experience.