prd

What problem does it solve? Writing a PRD from scratch often leads to inconsistent structure, invented baselines, and targets that do not map to company OKRs. This Skill produces a PRD that follows the MeetFlow template section for section, with every number and persona cited to a file in the repository, so reviewers can verify each claim. ## Core Features & Use Cases - Template-driven structure: Fills all 8 required sections (Problem, Who It's For, What It Does, How It Works, Success Metrics, What We're NOT Building, Dependencies, Timeline) in the exact order of assets/template-prd.md. - Evidence gathering before drafting: Reads company baselines, product vision, Q2 2026 OKRs, persona files, user feedback, and interview transcripts so every metric has a source. - Mechanical self-checks: Runs grep-based validation for leftover placeholders, missing sections, wrong date formats, and em dashes before handing over the document. - Use Case: Ask for a PRD for Transcript Accuracy v2 and receive a complete draft saved to the current cohort outputs folder, with baselines from company.md, targets from okrs-q2-2026.md, and persona pointers to 05-user-personas/. ## Quick Start Ask the assistant to write a PRD for your feature idea, for example: write a PRD for Transcript Accuracy v2 targeting June 2026 and save it under the current cohort outputs folder.