security-patch

Detect Treble, GSI, Mainline, APEX, system-as-root(SAR), A/B, etc.

The official repository of "GraphSPD: Graph-Based 安全 Patch Detection with Enriched Code Semantics". The paper will appear in the IEEE Symposium on 安全 and Privacy (S&P), San Francisco, CA, May 22-26, 2023.

The public dataset in the paper "PatchDB: A Large-Scale 安全 Patch Dataset". This paper appears in the 51st Annual IEEE/IFIP International Conference on Dependable Systems and 网络s (DSN), Online, June 21-24, 2021.

[ISSTA 2024] PatchFinder: A Two-Phase Approach to 安全 Patch Tracing for Disclosed Vulnerabilities in Open Source Software

QualvoSec is a minimalistic security patch management tool for unattended upgrades on Linux and BSD based systems that are obtaining their packages from a distribution based repository.

Composer meta-package that automatically applies critical and isolated security patches for Magento 2 / Adobe Commerce, released by Adobe

A demo program of security patch identification using the RNN model, which is demonstrated in the paper "PatchRNN: A Deep Learning-Based System for 安全 Patch Identification". This paper appears in the 2021 IEEE/AFCEA Military Communications Conference (MILCOM 2021), San Diego, USA, November 29–December 2, 2021.

Hackathor 2020 Contest Project

This is A CLI based simple credit card manager based on Beginner to intermediate OOP fundamentals. This code is written in python and is under rigorous development and new features will be updated regularly.

A Demo Program of 安全 Patch Identification with Graph Neural 网络s.

安全 Patch Identification with Classical Machine Learning Algorithms.

Paimon: Patch Identification Monster (extended version of GraphSPD)

Patch oversampling (synthesis) with direct patch analysis. This is an alternative solution to the PatchOversampling repository, providing a simpler and more direct way to synthesize patches. The original oversampling method is described in the DSN'21 paper "PatchDB: A Large-Scale 安全 Patch Dataset".

Data clearance for security patches and non-security patches. This method is described as Nearest Link 搜索 in the paper "PatchDB: A Large-Scale 安全 Patch Dataset", which appears in 2021 51st Annual IEEE/IFIP International Conference on Dependable Systems and 网络s (DSN 2021), Taipei, June 21-24, 2021, pp. 149-160.

Oversampling operations on security/non-security patches. This method is described in the paper "PatchDB: A Large-Scale 安全 Patch Dataset", which appears in 2021 51st Annual IEEE/IFIP International Conference on Dependable Systems and 网络s (DSN 2021), Taipei, June 21-24, 2021, pp. 149-160.

The source code in the paper "PatchRNN: A Deep Learning-Based System for 安全 Patch Identification". This paper appears in the 2021 IEEE/AFCEA Military Communications Conference (MILCOM 2021), San Diego, USA, November 29–December 2, 2021.

A security-hardened fork of "Simply Show Hooks". Replaces the compromised original (CVE-2024-6297) and patches unlisted Cross-Site Scripting (XSS) vulnerabilities.

RNN-based security patch identification with oversampling samples. This is an extension code in the MILCOM'21 paper "PatchRNN: A Deep Learning-Based System for 安全 Patch Identification".

Binary 安全 Patch Identification with Graph Neural 网络s.

A security-hardened fork of the abandoned "PostGallery" WordPress plugin. Fixes critical Arbitrary File Upload (CVE-2025-13543) and Guest Access vulnerabilities.