What problem does it solve?
This Skill helps product teams turn scattered requirement inputs, research notes, code scan findings, and conflicting stakeholder statements into a single structured PMContext. It reduces ambiguity by forcing every important point into a clear status: fact, assumption, conflict, or pending confirmation.
Core Features & Use Cases
- 8-dimension requirement refinement: Systematically covers user scenarios, boundary conditions, conflict detection, prioritization, terminology, current workaround friction, technical constraints, and success metrics.
- Dual operating modes: Supports guided questioning for collaborative clarification with a PM, or autonomous inference with the --auto mode when fast end-to-end refinement is needed.
- Traceable PMContext output: Produces a single self-contained PMContext file with confidence marking, information gaps, decision logs, and audit trails for downstream PRD, stories, and prototype generation.
- Use Case: After collecting inputs from user interviews, support tickets, internal docs, and codebase scans, use this Skill to consolidate them into a reviewable PMContext before generating a PRD.
Quick Start
Ask the agent to run pm-refine on your collected materials to clarify the requirement, label confidence and gaps, and write a PMContext file.