skill-creatorpar anthropic

Create, test, and iteratively improve AI agent skills with structured evaluation and benchmarking. Guides you through the full skill development lifecycle: intent capture, draft writing, test case creation, evaluation, and iteration based on user feedback Runs quantitative benchmarks comparing skill-enabled vs. baseline performance, with timing and token usage analysis across multiple test cases Generates an interactive browser-based review tool for qualitative evaluation — side-by-side...