Timezone change ripple risk
An AI agent changed the default timezone handling in a shared `formatTimestamp` utility to fix a bug where one customer saw times in UTC instead of their local zone. The targeted bug is fixed. This utility is imported by 40+ modules across billing, notifications, audit logs, and analytics. How do you confirm the change didn't silently break adjacent features — and what's the trap of relying on the test suite passing?
consumers_broken_by_default_change(consumer_lines: list[str], new_default_zone: str) → list[str][["billing|UTC|default","notifications|America/New_York|explicit","audit_log|UTC|default","analytics|UTC|explicit"],"America/Los_Angeles"]out["audit_log","billing"][["billing|UTC|default","notifications|America/New_York|explicit","audit_log|UTC|default","analytics|UTC|explicit"],"UTC"]out[][["support_inbox|America/New_York|default","billing|UTC|default"],"America/New_York"]out["billing"]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.