A study tested whether a small prompt edit improved tool-calling accuracy on 100 test cases from BFCL V4, a UC Berkeley benchmark. While the revised prompt successfully reduced invented optional arguments, the overall score gain was too small to be considered significant, and results remained inconsistent across runs even with fixed settings and providers.