Do Agents Dream of Root Shells? Partial-Credit Evaluation of LLM Agents in Capture the Flag Challenges

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Al-Kaswan, Ali, Plotnikov, Maksim, Hájek, Maxim, Vízner, Roland, van Deursen, Arie, Izadi, Maliheh
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917462918823936
author Al-Kaswan, Ali
Plotnikov, Maksim
Hájek, Maxim
Vízner, Roland
van Deursen, Arie
Izadi, Maliheh
author_facet Al-Kaswan, Ali
Plotnikov, Maksim
Hájek, Maxim
Vízner, Roland
van Deursen, Arie
Izadi, Maliheh
contents Large Language Model (LLM) agents are increasingly proposed for autonomous cybersecurity tasks, but their capabilities in realistic offensive settings remain poorly understood. We present DeepRed, an open-source benchmark for evaluating LLM-based agents on realistic Capture The Flag (CTF) challenges in isolated virtualized environments. DeepRed places an agent in a Kali attacker environment with terminal tools and optional web search, connected over a private network to a target challenge, and records full execution traces for analysis. To move beyond binary solved/unsolved outcomes, we introduce a partial-credit scoring method based on challenge-specific checkpoints derived from public writeups, together with an automated summarise-then-judge labelling pipeline for assigning checkpoint completion from logs. Using DeepRed, we benchmark ten commercially accessible LLMs on ten VM-based CTF challenges spanning different challenge categories. The results indicate that current agents remain limited: the best model achieves only 35% average checkpoint completion, performing strongest on common challenge types and weakest on tasks requiring non-standard discovery and longer-horizon adaptation.
format Preprint
id arxiv_https___arxiv_org_abs_2604_19354
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Do Agents Dream of Root Shells? Partial-Credit Evaluation of LLM Agents in Capture the Flag Challenges
Al-Kaswan, Ali
Plotnikov, Maksim
Hájek, Maxim
Vízner, Roland
van Deursen, Arie
Izadi, Maliheh
Artificial Intelligence
Cryptography and Security
Software Engineering
Large Language Model (LLM) agents are increasingly proposed for autonomous cybersecurity tasks, but their capabilities in realistic offensive settings remain poorly understood. We present DeepRed, an open-source benchmark for evaluating LLM-based agents on realistic Capture The Flag (CTF) challenges in isolated virtualized environments. DeepRed places an agent in a Kali attacker environment with terminal tools and optional web search, connected over a private network to a target challenge, and records full execution traces for analysis. To move beyond binary solved/unsolved outcomes, we introduce a partial-credit scoring method based on challenge-specific checkpoints derived from public writeups, together with an automated summarise-then-judge labelling pipeline for assigning checkpoint completion from logs. Using DeepRed, we benchmark ten commercially accessible LLMs on ten VM-based CTF challenges spanning different challenge categories. The results indicate that current agents remain limited: the best model achieves only 35% average checkpoint completion, performing strongest on common challenge types and weakest on tasks requiring non-standard discovery and longer-horizon adaptation.
title Do Agents Dream of Root Shells? Partial-Credit Evaluation of LLM Agents in Capture the Flag Challenges
topic Artificial Intelligence
Cryptography and Security
Software Engineering
url https://arxiv.org/abs/2604.19354