Context. Software Performance Engineering (SPE) and Chaos Engineering (CE) are crucial parts of the Software Development Lifecycle (SDLC) to test and improve a system’s quality attributes such as performance and reliability. Problem. Existing CE tools are either tightly coupled to specific infrastructures, support only a limited range of failure scenarios, or require extensive manual integration with load testing. As a result, researchers face significant overhead when attempting to conduct comprehensive and reproducible chaos experiments in combination with load tests. Objective. The objective of this thesis is to design and implement a generalized, infrastructure-agnostic framework that allows the automated execution of combined chaos experiments and load tests using e.g. MiSArch as System under Test (SuT), while allowing the derivation of service-level objectives (SLOs). Method. To achieve this, we follow the Design Science Research (DSR) methodology, where we first conduct interviews to elicit requirements, then design and implement the system architecture and finally evaluate the tool in a user study. In addition, we use the resulting software artifact to generate SLOs for MiSArch. Result. The resulting MiSArch Experiment Tool fulfills the majority of the requirements elicited and enables researchers to design, configure, and execute coordinated chaos and load experiments. However, while providing functional usefulness, usability still has room for improvement. Finally, through experimentation, SLOs for MiSArch were derived. Conclusion. This thesis contributes a novel framework that reduces the barrier to conducting chaos-based SPE experiments on microservice reference architectures. Although subject to limitations with respect to infrastructure access, Chaos Toolkit extension support, and workload modeling assumptions, the MiSArch Experiment Tool provides a valuable step toward more systematic, reproducible, and accessible SPE research.