During December 2024, Cyni enhanced the PurpleLlama repository by extending its test case data model in Python to improve prompt guard coverage. She introduced a new field, caught_by_promptguard, within query_llm.py, enabling more granular tracking of prompt guard scenarios during testing. This update focused on data structure design and careful codebase instrumentation, ensuring backward compatibility with existing test cases to minimize integration risk. By refining the test-data model and leveraging Python programming and Git-based change tracking, Cyni improved data fidelity and enabled faster iteration for prompt guard validation, ultimately reducing the risk of undetected issues in production prompt handling workflows.