Our paper suggests an adapted algorithm that can report error chains between API misuses. The empirical study onn 471 GitHub repositories showed that 50% of projects are affected by connected cryptographic API misuses. Further, the runtime overhead is minimal, and developers appreciate the adaption.
Empirial study of cryptographic misuses on enterprise-driven applications that identified several potential effective false positives, such as the use of hash algorithms in a non-security context. Further, we introduced a theoretical model of vulnerabilities caused by API misuses.
So far, benchmarks for cryptographic API misuses only focused on a subset of issues or tools. To drive future development in this domain, we will openly generate a benchmark. We will derive the generation of this novel benchmark from best practices.
Our study analyzes cryptographic API misuses in over 900 Python and MicroPython projects, revealing that 52% of the projects have at least one misuse. The findings indicate a positive impact of good API design in reducing misuses compared to Java and C.