Dhruv Kuchhal | Searcharxiv

arXiv2026

AgentShield: Fortifying Web Applications Against AI Agents and Agentic Browsers

LLM-powered browser agents are transforming web automation by autonomously interacting with websites and executing complex workflows. However, the effectiveness of existing anti-bot defenses against these systems remains unclear. We present a systematic empirical evaluation of modern agentic browsers against widely deployed anti-bot mechanisms, characterizing their capabilities, limitations, and behavioral differences from humans. Our results show that many defenses designed for traditional automation struggle to reliably detect autonomous AI-driven agents. Accordingly, we introduce AgentShield, a behavioral fingerprinting framework that targets the Perception-Reasoning-Action (PRA) loop underlying agentic user clients. By leveraging dynamic web forms that progressively reveal content and capture fine-grained interaction telemetry, AgentShield exposes behavioral artifacts that enable differentiation between human users and autonomous agents. We further demonstrate Extinction-induced Policy Drift (EPD), a novel attack in which repeated obstruction of task completion causes agents to relax their safeguards and engage in unsafe behaviors. Finally, we present defensive prompt injection techniques that exploit agent-specific weaknesses to mitigate emerging forms of autonomous web abuse that rely on content extraction. Our results highlight both the shortcomings of current web defenses against agentic user clients and the feasibility of practical detection mechanisms tailored to this new generation of web automation.

cs.CR↗

arXiv2018

Under the Shadow of Sunshine: Characterizing Spam Campaigns Abusing Phone Numbers Across Online Social Networks

Cybercriminals abuse Online Social Networks (OSNs) to lure victims into a variety of spam. Among different spam types, a less explored area is OSN abuse that leverages the telephony channel to defraud users. Phone numbers are advertized via OSNs, and users are tricked into calling these numbers. To expand the reach of such scam / spam campaigns, phone numbers are advertised across multiple platforms like Facebook, Twitter, GooglePlus, Flickr, and YouTube. In this paper, we present the first data-driven characterization of cross-platform campaigns that use multiple OSN platforms to reach their victims and use phone numbers for monetization. We collect 23M posts containing 1.8M unique phone numbers from Twitter, Facebook, GooglePlus, Youtube, and Flickr over a period of six months. Clustering these posts helps us identify 202 campaigns operating across the globe with Indonesia, United States, India, and United Arab Emirates being the most prominent originators. We find that even though Indonesian campaigns generate highest volume (3.2M posts), only 1.6% of the accounts propagating Indonesian campaigns have been suspended so far. By examining campaigns running across multiple OSNs, we discover that Twitter detects and suspends 93% more accounts than Facebook. Therefore, sharing intelligence about abuse-related user accounts across OSNs can aid in spam detection. According to our dataset, around 35K victims and 8.8M USD could have been saved if intelligence was shared across the OSNs. By analyzing phone number based spam campaigns running on OSNs, we highlight the unexplored variety of phone-based attacks surfacing on OSNs.

cs.CY↗