MemoPhishAgent: Memory-Augmented Multi-Modal LLM Agent for Phishing URL Detection

TL;DR

MemoPhishAgent is a memory-augmented multi-modal LLM system that dynamically uses episodic memories and tools to improve phishing URL detection, outperforming state-of-the-art methods on multiple datasets and real-world deployment.

Contribution

It introduces a novel memory-augmented multi-modal LLM agent that leverages episodic memories and tool orchestration for enhanced phishing detection.

Findings

MPA improves recall by 13.6% on public datasets.

On real-world URL benchmark, MPA improves recall by 20%.

Episodic memory contributes up to 27% recall gain without extra computation.

Abstract

Traditional phishing website detection relies on static heuristics or reference lists, which lag behind rapidly evolving attacks. While recent systems incorporate large language models (LLMs), they are still prompt-based, deterministic pipelines that underutilize reasoning capability. We present MemoPhishAgent (MPA), a memory-augmented multi-modal LLM agent that dynamically orchestrates phishing-specific tools and leverages episodic memories of past reasoning trajectories to guide decisions on recurring and novel threats. On two public datasets, MPA outperforms three state-of-the-art (SOTA) baselines, improving recall by 13.6%. To better reflect realistic, user-facing phishing detection performance, we further evaluate MPA on a benchmark of real-world suspicious URLs actively crawled from five social media platforms, where it improves recall by 20%. Detailed analysis shows episodic memory contributes up to 27% recall gain without introducing additional computational overhead. The ablation study confirms the necessity of the agent-based approach compared to prompt-based baselines and validates the effectiveness of our tool design. Finally, MPA is deployed in production, processing 60K targeted high-risk URLs weekly, and achieving 91.44% recall, providing proactive protection for millions of customers. Together, our results show that combining multi-modal reasoning with episodic memory yields robust phishing detection in realistic user-exposure settings. Our implementation is available at https://github.com/XuanChen-xc/MemoPhishAgent.git.