SCRUBD
收藏资源简介:
SCRUBD是由印度理工学院坎普尔分校开发的智能合约漏洞数据集,旨在评估智能合约中的重入(RE)和未处理异常(UX)漏洞。数据集包含两个子集:SCRUBD/CD(众包数据集)和SCRUBD/SD(合成数据集)。SCRUBD/CD通过众包和专家手动检查标注了469个合约中的746个RE函数和566个UX函数,而SCRUBD/SD则通过手动合成的方式覆盖了多种RE漏洞场景,包含239个函数。该数据集主要用于评估智能合约漏洞检测工具的性能,旨在解决智能合约安全性评估中的标准化数据集缺失问题。
SCRUBD is a smart contract vulnerability dataset developed by the Indian Institute of Technology Kanpur, designed to evaluate reentrancy (RE) and unhandled exception (UX) vulnerabilities in smart contracts. The dataset comprises two subsets: SCRUBD/CD (Crowdsourced Dataset) and SCRUBD/SD (Synthetic Dataset). SCRUBD/CD annotates 746 RE functions and 566 UX functions across 469 contracts through crowdsourcing and manual expert inspection. SCRUBD/SD covers various RE vulnerability scenarios via manual synthesis, containing 239 functions. This dataset is primarily used to evaluate the performance of smart contract vulnerability detection tools, aiming to address the shortage of standardized datasets in smart contract security assessment.




