概要
The Prompt Improver skill optimizes the interaction between users and AI by ensuring that vague requests are never executed based on assumptions. When a prompt lacks context, this skill triggers a structured four-phase workflow: it researches the codebase and conversation history, generates grounded multiple-choice questions for the user, gathers specific clarifications, and finally executes the task with full context. This process significantly reduces errors, minimizes hallucinations, and ensures that complex technical tasks are aligned with the project's actual architecture and requirements.
