Worked on stabilizing streaming tool call responses for local LLMs in the 0xPlaygrounds/rig repository, focusing on backend development and API integration using Rust. Addressed a parsing edge case where empty or missing function names in OpenAI provider configurations could lead to misclassification of partial tool calls during streaming. Improved the detection logic to ensure correct handling of these cases, reducing downstream errors and enhancing the reliability of streaming tool call responses. This targeted fix contributed to smoother developer workflows by maintaining the stability of local LLM tooling and ensuring robust integration with the OpenAI provider in production environments.