afrexai-prd-engine

What this skill does

# PRD Engine — Product Requirements That Ship

Complete product requirements methodology: from idea to spec to shipped feature. Not just a JSON template — a full system for writing PRDs that developers actually follow and stakeholders actually approve.

## When to Use This Skill

- Turning a vague idea into a buildable specification
- Writing PRDs for new features, products, or major refactors
- Reviewing/improving existing PRDs before sprint planning
- Breaking epics into right-sized user stories
- Creating technical design documents alongside product specs
- Preparing specs for AI coding agents (Claude Code, Cursor, Copilot)

---

## Phase 1: Discovery Brief

Before writing a single requirement, answer these questions. Skip this and you'll rewrite the PRD 3 times.

### Problem Validation Checklist

```yaml
discovery_brief:
  problem:
    statement: "" # One sentence. If you need two, you don't understand it yet.
    who_has_it: "" # Specific persona, not "users"
    frequency: "" # Daily? Weekly? Once? (daily problems > occasional ones)
    current_workaround: "" # What do they do today? (no workaround = maybe not a real problem)
    evidence:
      - type: "" # support_ticket | user_interview | analytics | churned_user | sales_objection
        detail: ""
        date: ""

  impact:
    users_affected: "" # Number or percentage
    revenue_impact: "" # $ at risk or $ opportunity
    strategic_alignment: "" # Which company goal does this serve?

  constraints:
    deadline: "" # Hard date or flexible?
    budget: "" # Engineering weeks available
    dependencies: "" # What must exist first?
    regulatory: "" # Any compliance requirements?

  success_metrics:
    primary: "" # ONE metric that defines success
    secondary: [] # 2-3 supporting metrics
    measurement_method: "" # How will you actually measure this?
    target: "" # Specific number, not "improve"
    timeframe: "" # When do you expect to see results?
```

### Problem Statement Formula

**[Persona] needs [capability] because [reason], but currently [blocker], which causes [measurable impact].**

Examples:
- ❌ "Users need better onboarding" (vague, unmeasurable)
- ✅ "New free-trial users (500/month) need to reach their first 'aha moment' within 10 minutes because 73% who don't will churn within 48 hours, but currently the average time-to-value is 34 minutes due to a 12-step setup wizard, which costs us ~$18K/month in lost conversions."

### Kill Criteria

Before proceeding, check these. If any are true, STOP and push back:

| Signal | Action |
|--------|--------|
| No evidence of the problem (just someone's opinion) | Demand evidence. Opinions aren't requirements. |
| Solution already decided ("just build X") | Rewind to the problem. Solutions without problems = features nobody uses. |
| Success metric is unmeasurable | Define how you'll measure it or don't build it. |
| Affects <1% of users with no revenue impact | Deprioritize. Small problems with small impact = small returns. |
| Scope keeps expanding during discovery | Scope lock. If everything is in scope, nothing is. |

---

## Phase 2: PRD Document

### PRD Template

```markdown
# PRD: [Feature Name]

**Author:** [Name]
**Status:** Draft | In Review | Approved | In Progress | Shipped
**Created:** YYYY-MM-DD
**Last Updated:** YYYY-MM-DD
**Approvers:** [Names + roles]

## 1. Problem & Opportunity

[Problem statement from discovery brief — one paragraph max]

### Evidence
- [Evidence point 1 — with data]
- [Evidence point 2 — with data]

### Impact
- Users affected: [number]
- Revenue impact: [$ amount or % change]
- Strategic goal: [which one]

## 2. Solution Overview

[2-3 paragraphs max. What are we building and why this approach?]

### What This Is
- [Bullet list of what's in scope]

### What This Is NOT
- [Explicit exclusions — this prevents scope creep]

### Key Decisions Made
| Decision | Options Considered | Chosen | Rationale |
|----------|-------------------|--------|-----------|
| [Decision 1] | A, B, C | B | [Why] |

## 3. User Stories

[See Phase 3 below for story writing methodology]

## 4. Design & UX

### User Flow
1. User [action] →
2. System [response] →
3. User sees [outcome]

### Wireframes/Mockups
[Link to Figma/screenshots or describe key screens]

### Edge Cases
| Scenario | Expected Behavior |
|----------|------------------|
| [Edge case 1] | [What happens] |
| [Edge case 2] | [What happens] |
| Empty state | [What user sees with no data] |
| Error state | [What user sees on failure] |
| Slow connection | [Loading behavior] |

## 5. Technical Considerations

### Architecture Notes
- [Key technical decisions]
- [New services/APIs needed]
- [Database changes]

### Dependencies
- [External service X]
- [Team Y's API]
- [Library Z]

### Performance Requirements
- Page load: <[X]ms
- API response: <[X]ms
- Concurrent users: [X]

### Security & Privacy
- [Data handling requirements]
- [Auth/permissions needed]
- [PII considerations]

## 6. Release Plan

### Rollout Strategy
- [ ] Feature flag: [flag name]
- [ ] Beta group: [who]
- [ ] % rollout: [10% → 50% → 100%]
- [ ] Rollback plan: [how]

### Launch Checklist
- [ ] QA sign-off
- [ ] Analytics events implemented
- [ ] Monitoring/alerts configured
- [ ] Documentation updated
- [ ] Support team briefed
- [ ] Stakeholders notified

## 7. Success Criteria

| Metric | Current | Target | Timeframe |
|--------|---------|--------|-----------|
| [Primary metric] | [X] | [Y] | [Z weeks] |
| [Secondary metric] | [X] | [Y] | [Z weeks] |

### Post-Launch Review
- **1-week check:** [What to look at]
- **1-month review:** [What to measure]
- **Kill/iterate decision:** [Criteria for each]
```

### PRD Quality Rubric (score before sharing)

| Dimension | 0-2 (Weak) | 3-4 (Adequate) | 5 (Strong) | Weight |
|-----------|-----------|----------------|------------|--------|
| **Problem clarity** | Vague, no data | Clear but thin evidence | Sharp statement + multiple evidence points | x4 |
| **Scope discipline** | Everything in scope | Some boundaries | Explicit in/out + "what this is NOT" | x3 |
| **Story quality** | Vague tasks | Stories with some criteria | INVEST stories + verifiable acceptance criteria | x4 |
| **Edge cases** | None listed | Happy path + 1-2 edges | Comprehensive: empty, error, slow, permissions, concurrent | x3 |
| **Success metrics** | "Improve X" | Metric + target | Metric + baseline + target + timeframe + measurement method | x3 |
| **Technical feasibility** | No tech section | Architecture notes | Dependencies, performance, security, migration plan | x2 |
| **Release plan** | None | "Ship it" | Feature flag + rollout % + rollback + launch checklist | x1 |

**Scoring:** Sum (score × weight). Max = 100.
- **80-100:** Ship-ready. Get approvals.
- **60-79:** Solid but missing pieces. Fill gaps before review.
- **40-59:** Needs work. Major sections incomplete.
- **<40:** Start over or go back to discovery.

---

## Phase 3: User Story Methodology

### Story Format

```yaml
story:
  id: "US-001"
  title: "" # Action-oriented: "Add priority field to tasks table"
  persona: "" # Who benefits
  narrative: "As a [persona], I want [capability] so that [benefit]"
  acceptance_criteria:
    - criterion: "" # Verifiable statement
      type: "functional" # functional | performance | security | ux
  priority: 1 # Execution order (dependencies first)
  size: "" # XS | S | M | L | XL
  status: "todo" # todo | in_progress | review | done
  notes: "" # Runtime observations
  depends_on: [] # Story IDs this depends on
  blocked_by: [] # External blockers
```

### INVEST Checklist (every story must pass)

| Letter | Criterion | Test |
|--------|-----------|------|
| **I** — Independent | Can be built without other incomplete stories | No circular dependencies |
| **N** — Negotiable | Details can flex (the "what" is fixed, the "how" is flexible) | Multiple implementation approaches exist |
| **V** — Valuable | Delivers user or business value on its own | Would a user/stakeholde