Question
You asked an AI assistant for a Python function that reads a JSON config file and returns the parsed dict. It gave you a clean three-liner that opens the file, calls json.load, and returns the result. It works perfectly when you point it at a good file. What did the AI likely leave out, and how do you decide whether that omission matters before shipping it?
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 coding: describe the solution in plain language (or narrate it) and the coach grades your approach. Generating runnable code from your description is coming next.