What problem does it solve? Turning a vague task or ticket into an actionable implementation plan often skips critical research and decision-making, leading to costly rework. This Skill guides an AI through a rigorous, interactive planning process that grounds every plan in actual codebase research and explicit user decisions. ## Core Features & Use Cases - Adaptive Questioning: Scales the number and focus of clarifying questions based on task complexity (LOW/MEDIUM/HIGH) and upstream artifacts like frame briefs and research docs, avoiding redundant questions. - Research-Driven Planning: Spawns parallel exploration agents to map the codebase, trace data flow, and surface conventions before any plan is written. - Structured Plan Output: Produces a phased plan.md with success criteria, a mechanical Progress tracking section, and a concise two-page plan brief stored under context/changes/<change-id>/. - Use Case: A developer receives a ticket to add OAuth login. They invoke the skill with the change ID; it reads prior research, asks targeted architecture and edge-case questions with recommended options, then writes a phased implementation plan with automated and manual verification steps. ## Quick Start Ask the AI to create a detailed implementation plan for your task or ticket, optionally passing a change ID or path to an existing research or frame document.
