Regex mentions false positives
You ask an AI for a Python regex to extract @mentions from chat messages so you can notify users. It returns `re.findall(r'@(\w+)', text)`. It pulls names fine until QA reports notifications firing from email addresses (`@gmail` in `a@gmail.com`), code blocks (`@property`), and it misses `@jean-luc` and `@O'Brien`. You re-prompt 'be more accurate' and it adds `\b`, which fixes nothing. How do you re-steer this to actually match your product's mention rules?
extract_mentions(text: str) → list[str]["hey @alice and @bob_smith"]out["alice","bob_smith"]["mail a@gmail.com now"]out[]["see ```@property``` but @data_team wins"]out["data_team"]Treat the AI’s output as a draft to verify, not an answer to trust. Name the specific flaw and the input that triggers it, say how you’d catch it (tests, edge cases, reading critically), and how you’d re-prompt or decompose to get it right.
Vibe & agentic: describe the solution in plain language (or narrate it) and the coach grades your approach.