Softfreak, an API governance specialist, said on Tuesday it will launch F-Avata Shell, a solution that automatically generates vulnerability assessment scripts using AI, within this year.

The company said F-Avata Shell creates and manages assessment scripts used in Softfreak's vulnerability assessment automation solution F-Avata.

It applies a patent-pending technology that combines a large language model (LLM) and a domain-specific language (DSL) to automate script writing and maintenance tasks previously carried out directly by security experts.

Vulnerability assessment scripts automatically check whether security settings for IT assets such as servers, PCs, DBMS, web, WAS and networks meet vulnerability analysis and evaluation standards for major information and communications infrastructure and electronic financial infrastructure.

F-Avata Shell generates assessment scripts tailored to target environments by converting and combining derived diagnostic rules into DSL-based standardised rules and templates after an LLM analyses inspection criteria in security guidelines and derives diagnostic rules.

Softfreak said its development goals for F-Avata Shell include cutting script creation time by up to 90 percent compared with manual work and reducing script maintenance work stemming from guideline changes by up to 70 percent. It also stressed that, because it manages inspection rules and scripts in a linked way, when guidelines change it can review the revised criteria and related scripts together and quickly reflect the changes.

Softfreak CEO Dong-cheol Kim (김동철) said, "A lot of time and manpower are投入 to modify and verify assessment scripts every time security guidelines change." He said, "F-Avata Shell will support security professionals by reducing the burden of these repetitive tasks so they can raise productivity and focus on analysing assessment results and making practical security improvements."