RefPentester: A Knowledge-Informed Self-Reflective Penetration Testing Framework Based on Large Language Models

TL;DR

RefPentester is a knowledge-informed, self-reflective penetration testing framework using LLMs that improves automation, adaptability, and success rates in identifying vulnerabilities, surpassing baseline models like GPT-4o.

Contribution

The paper introduces RefPentester, a novel LLM-based framework that incorporates self-reflection and knowledge guidance to enhance automated penetration testing.

Findings

RefPentester successfully revealed credentials on Hack The Box's Sau machine.

It outperformed the baseline GPT-4o model by 16.7% in success rate.

Demonstrated superior success rates across different PT stages.

Abstract

Automated penetration testing (AutoPT) powered by large language models (LLMs) has gained attention for its ability to automate ethical hacking processes and identify vulnerabilities in target systems by leveraging the inherent knowledge of LLMs. However, existing LLM-based AutoPT frameworks often underperform compared to human experts in challenging tasks for several reasons: the imbalanced knowledge used in LLM training, short-sightedness in the planning process, and hallucinations during command generation. Moreover, the trial-and-error nature of the PT process is constrained by existing frameworks lacking mechanisms to learn from previous failures, restricting adaptive improvement of PT strategies. To address these limitations, we propose a knowledge-informed, self-reflective PT framework powered by LLMs, called RefPentester. This AutoPT framework is designed to assist human operators in identifying the current stage of the PT process, selecting appropriate tactics and techniques for each stage, choosing suggested actions, providing step-by-step operational guidance, and reflecting on and learning from previous failed operations. We also modeled the PT process as a seven-state Stage Machine to integrate the proposed framework effectively. The evaluation shows that RefPentester can successfully reveal credentials on Hack The Box's Sau machine, outperforming the baseline GPT-4o model by 16.7%. Across PT stages, RefPentester also demonstrates superior success rates on PT stage transitions.