RiskChainBench is a new benchmark that pairs 3,600 synthetic token‑text restoration inputs with 600 human‑labeled local web environments to evaluate how well models can recover obfuscated platform messages and then investigate the associated websites. The benchmark measures both message restoration accuracy and the subsequent web‑investigation decision, using a fixed multimodal evidence judge to assess faithfulness, sufficiency, completeness, and consistency. Across ten models, performance varies widely, with entry recovery ranging from 35.2% to 95.2% and web decision accuracy from 26.3% to 62.8%, highlighting execution failures as the main bottleneck.
Platform abuse campaigns conceal redirection instructions with emojis, homophones, character decomposition, and redundant symbols, then route users through disguised links to services associated with...
The study investigates how often web agents incorrectly treat an HTTP 200 response as a successful fetch of useful content. By sampling 10,307 URLs—including random paths and expert‑verified claim citations—the authors found that roughly 20–25% of 200 responses lack the page’s main content, leading to inflated estimates of claim verifiability and citation drift. They propose simple heuristics and classifiers to detect unusable responses, achieving high recall with low false‑alarm rates, and recommend that agents report fetch outcomes explicitly.
arXiv:2610.19830v1 Announce Type: cross Abstract: Cyber Threat Intelligence (CTI) investigators who monitor dark web marketplaces for leaked credentials and breach listings must pass the diverse CAPT...
arXiv:2606.17467v3 Announce Type: replace-cross Abstract: Prompt injection defenses evaluated on synthetic benchmarks do not generalize to real enterprise documents, which are longer, denser, and int...
arXiv:2608. 04565v1 Announce Type: cross Abstract: LLM-based search agents are widely used for information-seeking tasks, but their reliance on external tool returns introduces a critical security risk: web content retrieved during execution is untrusted, exposing agents to prompt injection and goal hijacking.
The paper investigates how large language model (LLM) search agents can be manipulated through repeated, coordinated search-result injections, a threat that extends beyond single-page poisoning. It introduces Authority‑Chain Hijack (ACH), a strategy that chains manipulated evidence across multiple sources, achieving high attack success rates on the SafeSearch benchmark. The authors also present Trace‑Guided Strategy Evolution (TGSE), an automated method that refines attacker strategies using execution traces, further improving success rates.
arXiv:2603. 00801v2 Announce Type: replace Abstract: Language agents increasingly act as web-enabled systems that search, browse, and synthesize information from diverse sources.
arXiv:2609.06027v1 Announce Type: cross Abstract: Search-augmented LLM agents are increasingly used for consumer decisions, making them vulnerable to Generative Engine Optimization (GEO) poisoning. E...
ReDence is an open‑source hybrid framework that combines a multimodal agentic Investigator model with a specialized Decider model to detect modern social engineering scams. It automates investigative planning, evidence interpretation, and decision‑making across multiple scam categories, achieving 96% accuracy and operating at least four times faster than comparable agentic models. The system outputs structured, evidence‑grounded reports that enable analysts to audit decisions and support mitigation efforts.
TriShieldRAG introduces a three‑layered defense for Retrieval‑Augmented Generation: an Ingest Guard that screens documents, a Retrieval Scorer that re‑ranks based on trust, and a Cross‑LLM Consensus that validates evidence across three models. Against the original PoisonedRAG attack on the 2.68M‑passage Natural Questions corpus, the framework reduces attack success from about 79% to 1%. However, adaptive attacks that only alter document formatting can bypass the Ingest Guard and still achieve high success rates, revealing limits of layered defenses that rely on the same retrieved evidence.
arXiv:2606. 13385v1 Announce Type: cross Abstract: Web agents driven by large language models (LLMs) are increasingly deployed in real-world environments, where they operate over untrusted web content and execute actions with direct consequences.
The paper introduces MIRAGE, a benchmark of 750 multi‑step decision tasks designed to test autonomous web agents’ investigative abilities across Wikipedia Forensics, Shopping Admin adjudication, and Reddit Moderation. Each task contains a misleading visible context and a hidden context that holds decisive evidence, allowing performance to be broken down into Investigation, Reasoning, and Decision Accuracy, with an added Investigative Hallucination Rate. Evaluation of eight LLM agents reveals three consistent patterns: agents often reach relevant pages but fail to extract decisive evidence, procedural hints improve investigation but not decision accuracy on Wikipedia tasks, and 12.6% of trajectories include fabricated facts.