Newest papers

How Effectively Do Security Skills Guide Coding Agents in Vulnerability Detection in Linux Kernel?

Haoran Xue, Nikta Akbarpour, Gias Uddin, Song Wang · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.25705 · cited by 0 works (OpenAlex) · open-access PDF

German

Große potenzielle Schwachstellenmodelle (LLM) Codierungsagenten, wie Claude Code, werden zunehmend für Sicherheitsaudits eingesetzt, wobei Aufgabenspezifische Fähigkeiten wiederverwendbare Schwachstellenanalyseverfahren bieten. Allerdings sind ihre Wirksamkeit bei der Erkennung tatsächlicher Schwachstellen und die Gründe für ihre Ausfälle unklar. Wir führen die erste groß angelegte empirische Studie von 37 Sicherheitsfähigkeiten unter Verwendung von Claude Code mit vier zugrunde liegenden LLMs über 100 echte Linux-Kernel-Schwachstellen in zehn Klassen durch. Wir bewerten die Schwachstellenerkennung auf Funktion, Statement und Schwachstellenklasse-Ebenen. Die Ergebnisse zeigen, dass 24,4 Prozent der Läufer die korrekte Funktion lokalisieren, 18,3 Prozent sowohl die Funktion als auch die Sicherheits

Machine translation (Meta M2M100); the English original follows.

Abstract

Large language model (LLM) coding agents, such as Claude Code, are increasingly used for security auditing, with task-specific skills providing reusable vulnerability-analysis procedures. However, their effectiveness in detecting real vulnerabilities and the reasons for their failures remain unclear. We conduct the first large-scale empirical study of 37 security skills using Claude Code with four underlying LLMs on 100 real Linux kernel vulnerabilities across ten classes, totaling 14800 runs. We evaluate vulnerability detection at the function, statement, and vulnerability-class levels. Results show that 24.4 percent of runs locate the correct function, 18.3 percent identify both the function and responsible statement, and 15.7 percent correctly classify the vulnerability. Our failure analysis reveals limitations in evidence acquisition, state-relation reasoning, code interpretation, candidate verification, investigation coverage, and reporting. Based on these findings, we propose SKILL-SIF, a self-improving framework that iteratively evolves security skills by learning from missed vulnerabilities and failure patterns. We evolve five top-performing skills using the original benchmark and evaluate them on 100 previously unseen vulnerabilities. SKILL-SIF improves fully correct vulnerability detection by up to eight times (from 2 percent to 16 percent) on unseen vulnerabilities, while aggregate accuracy across four LLMs increases by 1.85 times (from 6.05 percent to 11.20 percent). These findings highlight the limitations of existing security skills and demonstrate the potential of failure-driven skill evolution to improve vulnerability detection in coding agents.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Vulnerability (computing), Coding (social sciences), Benchmark (surveying), Computer security, Secure coding

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange