LLM Security Guard for Code

TL;DR

LLMSecGuard is a framework combining static analyzers and LLMs to improve code security and benchmark the security evolution of LLMs in software development.

Contribution

It introduces LLMSecGuard, an open-source framework that enhances code security and provides security benchmarking for LLMs.

Findings

Enhanced security of code generated by LLMs

Open source framework available for developers

Benchmarking feature tracks security improvements over time

Abstract

Many developers rely on Large Language Models (LLMs) to facilitate software development. Nevertheless, these models have exhibited limited capabilities in the security domain. We introduce LLMSecGuard, a framework to offer enhanced code security through the synergy between static code analyzers and LLMs. LLMSecGuard is open source and aims to equip developers with code solutions that are more secure than the code initially generated by LLMs. This framework also has a benchmarking feature, aimed at providing insights into the evolving security attributes of these models.