/review-prd

name: review-prd
description: Comprehensive PRD review with product and technical analysis
category: PRD Management
version: 0.4.2

Review PRD Command

Comprehensive PRD review following product-first, then technical approach.

Purpose

Review and validate PRDs before development with:

• **Product validation**: Right problem, right solution, right scope
• **Technical feasibility**: Can build it, how to build it, risks
• **Clear decision**: GO/ITERATE/KILL with honest assessment
• **Grade assignment**: A-F rating for quality tracking

Workflow

1. List & Select PRD

Scan `product/prds/01-draft/` and display selection menu with PRD name, priority, and current grade.

2. Complete Review Analysis

**Output comprehensive review in one response** covering:

A. Product Review

**Problem validation**:

• User persona and pain point
• Frequency and severity
• Evidence (support tickets, user research, churn data)
• Challenge: Is this real and urgent?

**Solution assessment**:

• Proposed approach summary
• Scope (what's in, what's out, why)
• Alternative approaches considered
• Simplification opportunities

**Component challenge**:

• Each major component: Essential vs nice-to-have vs bloat
• Usage estimate and effort
• Impact if removed
• Verdict: Keep/Defer/Remove

**User experience**:

• Current vs proposed flow
• Friction points identified
• Journey improvements

**Success metrics**:

• How success is measured
• Targets with baselines
• Acceptance criteria completeness

**Strategic context**:

• Company strategy alignment
• Competitive landscape
• Timing rationale

**Product Score**: X/10 with strengths and weaknesses

B. Technical Review

**Architecture & design**:

• Proposed architecture fit
• Design patterns used
• Integration points
• Technical debt impact

**Implementation breakdown**:

• Major work areas with realistic estimates
• Critical paths identified
• Testing strategy

**Dependencies & blockers**:

• Technical dependencies (APIs, services, data)
• Team dependencies
• Hard blockers vs soft dependencies

**Tech stack**:

• New dependencies and their risks
• Breaking changes
• Tech debt created vs paid

**Quality & performance**:

• Performance impact and SLA targets
• Security attack surface
• Scalability considerations
• Monitoring requirements

**Risk assessment**:

• Technical risks with likelihood and impact
• Mitigation strategies
• Known unknowns and de-risking approaches

**Technical Score**: X/10 with technical feeling

C. Synthesis & Decision

**Final Grade**: A-F (Product score + Technical score)

**Global feeling**: Honest assessment of PRD quality and readiness

**3 Critical Challenges**:

1. **KILL Challenge**: Should we build this?

• Problem validated?
• Better alternatives exist?
• Strategic fit?
• Verdict: Build / Don't build

2. **TIMING Challenge**: Now or later?

• Urgency level
• Team capacity
• Dependencies ready?
• Opportunity cost
• Verdict: Now / Defer (when?)

3. **SCOPE Challenge**: All or part?

• MVP viable?
• Effort saved by reducing
• Value preserved
• Verdict: Full scope / MVP first / Phased

**Blockers** (must fix before /setup-prd):

• Critical issues preventing development start
• Owner assignments

**Recommendations** (to improve grade):

• Specific improvements with effort/impact
• Priority ranking

**ROI Summary**:

• Effort estimate with confidence
• Value assessment
• Risk level
• Overall ROI: Excellent/Good/Questionable/Poor

**Next Steps**:

• ✅ Ready for development OR
• ⚠️ Needs iteration (what specifically)
• Recommended priority
• Action items with owners

Grading System

**A (90-100)**: Exceptional PRD, ready for immediate development **B (80-89)**: Strong PRD, minor improvements recommended **C (70-79)**: Solid PRD, some concerns need addressing **D (60-69)**: Weak PRD, significant gaps to fill **F (<60)**: Not ready, major rework needed

Principles

**Product-first thinking**:

• Most PRD failures = building wrong thing
• Challenge assumptions aggressively
• Demand evidence, not opinions
• Think ROI: Effort vs Value vs Risk

**Technical realism**:

• Research actual codebase (real files, patterns, gaps)
• Realistic estimates (not ideal case)
• Surface unknowns honestly
• Identify dependencies early

**Honest feedback**:

• Share real feeling, don't sugarcoat
• Grade reflects actual quality
• Suggest optimizations (smaller scope, phased, alternatives)
• Call out blockers clearly

**Decision clarity**:

• Clear GO/ITERATE/KILL verdict
• Specific next actions
• Owner assignments
• Timeline expectations

Configuration

{
  "prd_workflow": {
    "directories": {
      "draft": "product/prds/01-draft",
      "ready": "product/prds/02-ready"
    },
    "review": {
      "grading_enabled": true,
      "minimum_grade": "C"
    }
  }
}

Success Criteria

• ✅ Product validation complete (problem, solution, scope)
• ✅ Technical feasibility assessed (architecture, risks, estimates)
• ✅ Clear grade (A-F) with justification
• ✅ 3 critical challenges answered (KILL/TIMING/SCOPE)
• ✅ Actionable next steps with owners
• ✅ User understands PRD quality and readiness

Example Output

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📋 PRD-007: OAuth2 Integration | P0 | 9-day effort
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

🎨 PRODUCT REVIEW

Problem: 12% signup drop-off, users want one-click auth
Evidence: 47 support tickets, top-3 in user research
Challenge: ✅ Validated (strong evidence)

Solution: Google + GitHub OAuth
Alternatives: Auth0 ($300/month), magic links (doesn't solve one-click)
Recommendation: Build in-house (lower long-term cost)

Scope optimization: Remove Microsoft OAuth (2% of signups)
Impact: 64% effort saved, 95% value preserved

Product Score: 9/10 (strong validation, clear scope)

🔧 TECHNICAL REVIEW

Architecture: Passport.js (already in stack)
Estimate: 5 days MVP (Google only) + 1 day GitHub
Risk: OAuth security (state param, credent