CKChad Kimes

Chad Kimes

*This is an algorithmically generated author record based on publicly available information.

Chad Kimes has a total of 1 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Auditing the CI/CD Platform: Reproducible Builds vs. Hardware-Attested Build Environments, Which is Right for You?

    by Marcela S. Melara, Chad Kimes

    Recent software supply chain attacks, such as the SolarWinds hack and xz-utils backdoor, demonstrate the widespread impact of tampering with a package's build environment to subvert the final software product. To detect compromises in the systems and services that underlie the supply chain with higher confidence, software vendors today may choose between a few promising approaches: Reproducible Builds, or hardware-attested build environments, a proposed combination of CNCF in-toto, OpenSSF Supplychain Levels for Software Artifacts (SLSA) and trusted hardware. At first glance, Reproducible Builds and hardware-attested build environments address the same challenge of auditing the build platform for tampering. The former establishes practices for achieving consistent build process results, making divergences detectable, while the latter captures verifiable evidence about properties of the compute environment that executed a build, making changes to that environment detectable. Yet, despite relying largely on existing techniques, neither approach has seen broad adoption by CI/CD platforms or enterprise software vendors. To understand these trends, we seek to answer two central questions: (1) What security problems does each approach address? (2) What are the practical adoption requirements of each approach? In this security-in-practice talk, we examine the similarities and differences between Reproducible Builds and hardware-attested build environments, and discuss challenges of implementation in practice. We will address a very common question about the tradeoffs between adopting only one or both approaches, and conclude with adoption recommendations for several types of software development settings.