name: rfc-review description: "Structured review of game design RFCs with quantitative scoring across 6 dimensions: core loop, progression, economy, player motivation, risk, cross-consistency. Use when reviewing docs/rfcs/ documents, evaluating game design decisions, or checking design health before implementation. Not for code review (use @implementation-review), not for balance numbers (use balance-review later)." category: 00-meta layer: governance related_skills: - "@rfc-orchestrator" - "@context-discovery"
/rfc-review: Game Design RFC Review
Interactive RFC review with quantitative scoring. Every recommendation includes WHY + concrete alternative. No vague praise.
Inspired by gstack-game's /game-review, adapted for RFC-driven development.
Phase 0: Read & Orient
Read the target RFC from docs/rfcs/. Extract 5 context anchors:
| Anchor | Status | Found |
|---|---|---|
| Genre & Platform | ✅/❌ | {what was found} |
| Target Session Length | ✅/❌ | {what was found} |
| Monetization Model | ✅/❌ | {what was found} |
| Target Audience | ✅/❌ | {what was found} |
| Design Pillars | ✅/❌ | {what was found} |
For ❌ items, provide best-guess with [inferred] tag.
Confirm context before proceeding. For missing anchors, ask ONE question with options and a RECOMMENDATION.
Mode Selection
Select review mode based on game type:
| Mode | Focus |
|---|---|
| PC/Console RPG (default for ultima-magic) | Core loop depth, mastery curve, narrative, session arc |
| Mobile/Casual | Retention, economy, session fit, monetization ethics |
| Multiplayer/Competitive | Balance, matchmaking, counterplay |
Mode Weight Table (PC/Console RPG default)
| Section | Weight |
|---|---|
| 1. Core Loop | 30% |
| 2. Progression & Retention | 20% |
| 3. Economy | 10% |
| 4. Player Motivation | 15% |
| 5. Risk Assessment | 10% |
| 6. Cross-Consistency | 15% |
Review Pacing
After EACH section, present:
- Section score:
Section {N} — {name}: {score}/10 - 1-sentence biggest finding
- Options: A) Continue, B) Dig deeper, C) Fast-forward, D) Stop here
STOP and wait after each section.
Section 1: Core Loop
Evaluate the nested loop model:
- Micro loop (moment-to-moment): Is the core action satisfying? What's the verb?
- Meso loop (session): What's the session goal? How long is a session?
- Macro loop (progression): What keeps the player coming back?
- Meta loop (mastery): What's the long-term aspiration?
Check:
- Loop clarity — can you describe each loop in one sentence?
- Loop depth — does the micro loop have meaningful decisions?
- Loop uniqueness — what differentiates from competitors?
- MDA alignment — do Mechanics produce intended Dynamics and Aesthetics?
Score: ___/10
Section 2: Progression & Retention
Evaluate at each retention tier:
| Tier | Question |
|---|---|
| FTUE | Does the first 5 minutes teach the core loop AND deliver an "aha moment"? |
| D1 | What reason does the player have to come back tomorrow? |
| D7 | What new system/content unlocks by day 7? |
| D30 | What's the long-term aspiration keeping mastery players? |
Check:
- Flow state design — is difficulty curve shaped for flow?
- Skill gates vs content gates vs time gates — which are used and are they appropriate?
- Churn points — where will players most likely quit?
Score: ___/10
Section 3: Economy & Monetization
Evaluate currency and reward systems:
- Currencies — how many? Are they clear?
- Sinks & Faucets — is there a sink for every faucet?
- Reward schedules — variable ratio? Fixed interval?
- Spending tiers — does economy serve different player types?
Red flags:
- Economy with no sinks (hyperinflation)
- Pay-gated core progression
- Probabilistic rewards with no pity system
Score: ___/10
Section 4: Player Motivation & Emotion
Evaluate using Self-Determination Theory (SDT):
| Need | In RFC? | How? |
|---|---|---|
| Autonomy | ✅/❌ | {meaningful choices} |
| Competence | ✅/❌ | {skill growth} |
| Relatedness | ✅/❌ | {connection} |
Check:
- Bartle type coverage — which player types are served?
- Emotional arc — does the design create tension/release cycles?
- Ludonarrative consonance — do mechanics support the story?
Score: ___/10
Section 5: Risk Assessment
Evaluate across 5 categories:
| Risk | Level | Detail |
|---|---|---|
| Pillar violation | H/M/L | Does any system contradict design pillars? |
| Scope (Lake vs Ocean) | H/M/L | Is scope achievable? Deep and narrow > shallow and wide. |
| Technical feasibility | H/M/L | Can this be built with current tech stack? |
| Market differentiation | H/M/L | What makes this different from competitors? |
| Retention cliffs | H/M/L | Where does content/system depth run out? |
Score: ___/10
Section 6: Cross-RFC Consistency
Cross-validate findings across sections AND across related RFCs:
- Does the combat RFC's difficulty curve match the progression RFC's pacing?
- Does the party RFC's character growth match the dungeon RFC's challenge scaling?
- Does the overworld RFC's exploration match the state machine RFC's transitions?
- Does the UI RFC support all the information needs of combat, dungeon, and overworld?
Flag contradictions invisible within any single RFC.
Score: ___/10
Forcing Questions
Minimum 2, routed by RFC maturity:
| RFC State | Focus |
|---|---|
| Early/skeleton | "Describe the core loop in one sentence" / "Who comes back tomorrow?" |
| Detailed | "Have you prototyped this?" / "What breaks at 2x scale?" |
| Post-implementation | "What surprised you in testing?" / "What changed from the RFC?" |
Fix-then-Rescore
When user updates the RFC during review:
- Re-read updated section
- Re-score ONLY that section
- Update running total
- If final < baseline: WARN — fix introduced new problem
Game Design Vocabulary
Use these terms consistently:
- Core loop, session loop, meta loop — loop hierarchy
- FTUE — First Time User Experience
- Aha moment — when the player "gets it"
- Churn point — where players quit
- Sink/faucet — currency consumption/generation
- Skill gate — requires player skill to pass
- Content gate — requires content consumption to pass
- Time gate — requires real time to pass
- Flow state — optimal challenge/skill balance
Completion Summary
RFC Review: {RFC name}
═══════════════════════════════════════════════
Mode: PC/Console RPG
Section 1 — Core Loop: _/10 (30%) → weighted: _.__
Section 2 — Progression: _/10 (20%) → weighted: _.__
Section 3 — Economy: _/10 (10%) → weighted: _.__
Section 4 — Player Motivation: _/10 (15%) → weighted: _.__
Section 5 — Risk Assessment: _/10 (10%) → weighted: _.__
Section 6 — Cross-Consistency: _/10 (15%) → weighted: _.__
─────────────────────────────────────────────
WEIGHTED TOTAL: _._/10
Top 3 Deductions:
1. [Section] [Issue]: -N because [specific reason]
2.
3.
STATUS: DONE / DONE_WITH_CONCERNS / BLOCKED / NEEDS_CONTEXT
Important Rules
- ONE question at a time. Never batch forcing questions.
- Section transitions mandatory. Score + pacing options after every section.
- Push twice max. Vague answer → push for specifics → still vague → flag, move on.
- Escape hatch: "skip ahead" → AUTO-only remaining sections.
- No code suggestions. Design review only. Technical → note for @implementation-review.
- Anti-sycophancy: Score honestly. A 5/10 with specific deductions is more useful than a polite 8/10.