multica-ai

Applies lightweight heuristics—avoid overcomplication, make surgical edits, surface unstated assumptions, define testable success—to catch the sneaky mistakes LLMs make most often.