Generates pytest test cases for Python functions using an LLM, then validates them by actually running pytest — retrying with the failure context fed back in when a generated test breaks. The retry-with-error-context loop here is the same shape you’ll see again in this path’s SQL agent — once you’ve built it once, you recognize it everywhere.
Test generation that actually runs and retries is a meaningfully different (and more hireable) claim than 'I used Copilot to write tests' — shows you can validate AI output programmatically.