recipe-definelisted

**Context**: Transform validated hypotheses into a PRD with 4 Risks confidence scores, hypothesis traceability, and user stories. The PRD follows a standard structure that can be consumed by downstream implementation workflows (e.g., claude-code-workflows). ## Orchestrator Definition **Execution Protocol**: 1. **Delegate review** to prd-reviewer (via Agent tool, subagent_type: "discover:prd-reviewer") for bias-free quality assessment 2. **Follow the definition flow** defined below 3. **Stop at every `[STOP — BLOCKING]` marker** — present findings and CANNOT proceed until user explicitly confirms ## Workflow Overview ``` Input (validated hypotheses + Opportunity context) ↓ 1. Readiness Assessment → Check hypothesis confidence levels ↓ 2. PRD Drafting → Using prd-template.md with discovery extensions ↓ 3. User Story Generation → 4 Risks per story [Stop: User reviews PRD draft] ↓ 4. Quality Review → prd-reviewer assesses the PRD [Stop: User reviews feedback] ↓ Output: docs/prd/[feature-name]-prd.md ``` ## Execution Decision Flow ### 1. Readiness Assessment Input: $ARGUMENTS Assess whether hypotheses are "validated enough" for PRD creation: 1. Read relevant Opportunity and hypothesis files 2. For each hypothesis intended for the PRD: - Check confidence scores against thresholds (see prd-standards skill `references/user-story-guide.md`) - Assess cost x risk x reversibility - Determine: validated enough / needs more validation 3. See product-pr