clarification

What problem does it solve? AI agents often interrupt users with excessive clarification questions or, conversely, make wrong assumptions on genuinely ambiguous requests. This Skill defines clear rules for when to ask questions versus when to proceed with reasonable assumptions, preventing question spam while avoiding costly misinterpretations. ## Core Features & Use Cases - Decision Framework: Provides explicit criteria for when to ask questions (high-impact architectural, security, or performance decisions) versus when to use reasonable defaults (standard patterns like pytest, black, or logging). - Question Formatting Rules: Enforces a single-question pattern with 2-4 options and brief tradeoff descriptions, avoiding open-ended or stacked questions. - Context-Aware Behavior: Adjusts questioning behavior by phase—planning, execution, investigation, and revision—so exploration is never interrupted mid-flow. - Use Case: When a user says "add caching", the agent recognizes this as ambiguous (in-memory vs Redis vs Memcached) and asks one structured question; when a user says "add tests", it proceeds directly with pytest. ## Quick Start Activate this skill so the agent only asks clarifying questions for genuinely ambiguous, high-impact decisions and uses stated assumptions otherwise.