In February 2025, Kyoon focused on enhancing secret management for LLM integration within the mit-submit/A2rchi repository. Kyoon developed a conditional API key preparation system in Python, leveraging configuration management and CLI development skills. This approach introduced a model-to-secret mapping and usage guards, ensuring that API keys are only provisioned for LLM models actively used by the user. By making API keys optional for unused models, Kyoon reduced credential exposure and simplified onboarding for diverse model configurations. The work improved security hygiene and maintainability, laying a foundation for scalable secret governance and more flexible, error-resistant API integration workflows.