Construction and Evaluation of LLM-based agents for Semi-Autonomous penetration testing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kobayashi, Masaya, Fuchi, Masane, Zanashir, Amar, Yoneda, Tomonori, Takagi, Tomohiro
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909504135757824
author Kobayashi, Masaya
Fuchi, Masane
Zanashir, Amar
Yoneda, Tomonori
Takagi, Tomohiro
author_facet Kobayashi, Masaya
Fuchi, Masane
Zanashir, Amar
Yoneda, Tomonori
Takagi, Tomohiro
contents With the emergence of high-performance large language models (LLMs) such as GPT, Claude, and Gemini, the autonomous and semi-autonomous execution of tasks has significantly advanced across various domains. However, in highly specialized fields such as cybersecurity, full autonomy remains a challenge. This difficulty primarily stems from the limitations of LLMs in reasoning capabilities and domain-specific knowledge. We propose a system that semi-autonomously executes complex cybersecurity workflows by employing multiple LLMs modules to formulate attack strategies, generate commands, and analyze results, thereby addressing the aforementioned challenges. In our experiments using Hack The Box virtual machines, we confirmed that our system can autonomously construct attack strategies, issue appropriate commands, and automate certain processes, thereby reducing the need for manual intervention.
format Preprint
id arxiv_https___arxiv_org_abs_2502_15506
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Construction and Evaluation of LLM-based agents for Semi-Autonomous penetration testing
Kobayashi, Masaya
Fuchi, Masane
Zanashir, Amar
Yoneda, Tomonori
Takagi, Tomohiro
Cryptography and Security
With the emergence of high-performance large language models (LLMs) such as GPT, Claude, and Gemini, the autonomous and semi-autonomous execution of tasks has significantly advanced across various domains. However, in highly specialized fields such as cybersecurity, full autonomy remains a challenge. This difficulty primarily stems from the limitations of LLMs in reasoning capabilities and domain-specific knowledge. We propose a system that semi-autonomously executes complex cybersecurity workflows by employing multiple LLMs modules to formulate attack strategies, generate commands, and analyze results, thereby addressing the aforementioned challenges. In our experiments using Hack The Box virtual machines, we confirmed that our system can autonomously construct attack strategies, issue appropriate commands, and automate certain processes, thereby reducing the need for manual intervention.
title Construction and Evaluation of LLM-based agents for Semi-Autonomous penetration testing
topic Cryptography and Security
url https://arxiv.org/abs/2502.15506