OSPTrack
收藏资源简介:
OSPTrack是由格拉斯哥大学创建的一个标签化数据集,专注于模拟开源软件包的执行过程。该数据集涵盖了多个生态系统,包括npm、pypi、crates.io、nuget和packagist,共包含9,461个软件包报告,其中1,962个为恶意软件包。数据集通过在隔离环境中捕获软件包和库执行期间生成的特征,如文件、套接字、命令和DNS记录,来帮助识别恶意指示器。数据集的创建过程包括多进程分析、报告解析和特征提取,旨在解决开源软件供应链安全中的漏洞检测问题,特别是在源代码访问受限的情况下。
OSPTrack is a labeled dataset created by the University of Glasgow, focusing on simulating the execution processes of open-source software packages. This dataset covers multiple package ecosystems including npm, pypi, crates.io, nuget, and packagist, containing a total of 9,461 software package reports, among which 1,962 are malicious packages. It helps identify malicious indicators by capturing features generated during the execution of packages and libraries in isolated environments, such as files, sockets, commands, and DNS records. The dataset's creation workflow includes multiprocess analysis, report parsing, and feature extraction, aiming to address vulnerability detection issues in open-source software supply chain security, particularly when access to source code is restricted.




