LLM-Powered Secure Code Review Assistant for Student Projects: A Hybrid Approach Combining Static Analysis and Local Large Language Models

This research presents an LLM-powered secure code review assistant designed to help student developers identify and understand potential security vulnerabilities in software projects. The proposed system combines static analysis tools, including Semgrep, Bandit, ESLint, SpotBugs, and Trivy, with a locally hosted Qwen large language model through Ollama. The platform supports ZIP file uploads and public GitHub repositories, coordinates security scanning, consolidates findings, and generates contextual explanations and potential remediation suggestions. It also provides a dashboard, scan history, and PDF and JSON reports. The paper describes the system architecture, workflow, implementation approach, security considerations, limitations, and proposed evaluation methodology. It distinguishes scanner-reported findings from AI-generated observations to avoid presenting unverified AI suggestions as confirmed vulnerabilities. Quantitative performance claims are not made because completed benchmark and user-study results are not reported.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-16
DOI
https://doi.org/10.5281/zenodo.24268520
Primary Topic
Web Application Security Vulnerabilities
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

LLM-Powered Secure Code Review Assistant for Student Projects: A Hybrid Approach Combining Static Analysis and Local Large Language Models

Chandrashekhar Kamargaonkar, Akash Kurane, Shubham Phad, Kaustubh Manjarekar
4 links
Zenodo (CERN European Organization for Nuclear Research)
Web Application Security Vulnerabilities
preprint

LLM-Powered Secure Code Review Assistant for Student Projects: A Hybrid Approach Combining Static Analysis and Local Large Language Models

Chandrashekhar Kamargaonkar, Akash Kurane, Shubham Phad, Kaustubh Manjarekar
preprint en

Abstract

This research presents an LLM-powered secure code review assistant designed to help student developers identify and understand potential security vulnerabilities in software projects. The proposed system combines static analysis tools, including Semgrep, Bandit, ESLint, SpotBugs, and Trivy, with a locally hosted Qwen large language model through Ollama. The platform supports ZIP file uploads and public GitHub repositories, coordinates security scanning, consolidates findings, and generates contextual explanations and potential remediation suggestions. It also provides a dashboard, scan history, and PDF and JSON reports. The paper describes the system architecture, workflow, implementation approach, security considerations, limitations, and proposed evaluation methodology. It distinguishes scanner-reported findings from AI-generated observations to avoid presenting unverified AI suggestions as confirmed vulnerabilities. Quantitative performance claims are not made because completed benchmark and user-study results are not reported.

Zenodo (CERN European Organization for Nuclear Research)
MIT Art, Design and Technology University (IN)
Web Application Security Vulnerabilities
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon