$npx -y skills add Stanestane/game-design-skills-bundle --skill game-design-peak-end-auditAudit a game, feature, session, level, event, onboarding flow, reward sequence, or return-player experience through the lens of the peak-end rule. Use when evaluating which moments players are most likely to remember, whether the emotional high points are strong enough, whether e
| 1 | # Game Design Peak-End Audit |
| 2 | |
| 3 | Audit a design by asking which moments dominate memory and whether the experience earns the memory it leaves behind. |
| 4 | |
| 5 | Use this skill when the important question is not just how the experience feels moment to moment, but how players will remember it afterward. The goal is to identify the emotional peaks, the ending shape, and the aftertaste that determine whether a session, level, feature, event, or sequence is remembered as exciting, exhausting, disappointing, triumphant, or forgettable. |
| 6 | |
| 7 | Read `references/family-conventions.md` when you want the shared style, prioritization, and diagnosis rules for this game-design skill family. |
| 8 | Read `references/output-patterns.md` when you want the preferred recommendation and minimal-fix structure. |
| 9 | |
| 10 | ## Core principle |
| 11 | |
| 12 | Players do not remember an experience as an average of every second they lived through. |
| 13 | |
| 14 | They disproportionately remember: |
| 15 | - the strongest emotional peaks |
| 16 | - the ending |
| 17 | - the story those moments imply afterward |
| 18 | |
| 19 | That means a design with decent average quality can still be remembered badly if: |
| 20 | - its emotional peaks are weak |
| 21 | - its worst spike dominates memory |
| 22 | - its ending sours the whole experience |
| 23 | |
| 24 | And a rougher experience can still land well if: |
| 25 | - its peak is strong and meaningful |
| 26 | - its ending resolves cleanly |
| 27 | - the memory closes with satisfaction, momentum, relief, or anticipation |
| 28 | |
| 29 | ## What to produce |
| 30 | |
| 31 | Generate: |
| 32 | 1. **Peak-end profile** - the strongest peaks, the ending shape, and the likely remembered summary |
| 33 | 2. **Memory dominance diagnosis** - which moments will overshadow the rest of the experience |
| 34 | 3. **Ending-quality diagnosis** - whether the closing moments strengthen or poison the memory |
| 35 | 4. **Experience-shape risks** - where the design is setting itself up to be remembered for the wrong thing |
| 36 | 5. **Design actions** - what to amplify, soften, reorder, frame, or close differently |
| 37 | |
| 38 | ## Process |
| 39 | |
| 40 | ### 1. Define the audit target |
| 41 | Clarify: |
| 42 | - what exact experience slice is being audited |
| 43 | - where it begins and ends from the player's point of view |
| 44 | - what player segment matters most |
| 45 | |
| 46 | Write: |
| 47 | - **Audit target** |
| 48 | - **Experience window** |
| 49 | - **Primary player segment** |
| 50 | |
| 51 | ### 2. Map the emotional shape over time |
| 52 | Break the experience into phases and ask: |
| 53 | - where does emotional intensity rise? |
| 54 | - where does it sag? |
| 55 | - where does frustration spike? |
| 56 | - where does satisfaction spike? |
| 57 | - what is the final emotional note? |
| 58 | |
| 59 | Do not settle for generic labels like "good pacing." Name the moments. |
| 60 | |
| 61 | ### 3. Identify positive and negative peaks |
| 62 | Look for the moments most likely to become the remembered highlight or remembered wound. |
| 63 | |
| 64 | Examples: |
| 65 | - first big win |
| 66 | - dramatic comeback |
| 67 | - painful difficulty spike |
| 68 | - surprising reveal |
| 69 | - humiliating loss |
| 70 | - beautiful payoff moment |
| 71 | - reward opening high |
| 72 | - last-minute collapse |
| 73 | |
| 74 | For each peak, ask: |
| 75 | - how intense is it? |
| 76 | - how clear is it? |
| 77 | - does it support the intended fantasy? |
| 78 | - is it the thing you actually want remembered? |
| 79 | |
| 80 | ### 4. Audit the ending |
| 81 | The ending may be: |
| 82 | - session close |
| 83 | - boss resolution |
| 84 | - reward-claim sequence |
| 85 | - level completion |
| 86 | - event wrap-up |
| 87 | - return-to-hub state |
| 88 | - defeat screen and immediate aftermath |
| 89 | |
| 90 | Ask: |
| 91 | - what emotion does the player leave with? |
| 92 | - does the ending frame the previous experience positively or negatively? |
| 93 | - does it create closure, momentum, relief, excitement, emptiness, or irritation? |
| 94 | - does it respect the effort that preceded it? |
| 95 | |
| 96 | ### 5. Determine the likely remembered story |
| 97 | Translate the shape into the sentence the player is likely to remember later. |
| 98 | |
| 99 | Examples: |
| 100 | - "That run had one incredible clutch ending" |
| 101 | - "The fight was fine, but the ending reward felt pathetic" |
| 102 | - "That event was mostly chores and then it just sort of stopped" |
| 103 | - "The level was hard, but the payoff made it worth it" |
| 104 | |
| 105 | If the remembered sentence is bad, average quality elsewhere may not save the experience. |
| 106 | |
| 107 | ### 6. Diagnose peak-end failure patterns |
| 108 | Look for: |
| 109 | - strong negative spike dominating weak positives |
| 110 | - flat experience with no memorable high |
| 111 | - rewarding middle but deflating ending |
| 112 | - emotionally strong ending attached to a fantasy mismatch |
| 113 | - punishing final note after meaningful progress |
| 114 | - reward ceremony too weak for the effort invested |
| 115 | - admin or UI sludge contaminating the exit |
| 116 | |
| 117 | ### 7. Check segment differences |
| 118 | Ask whether: |
| 119 | - new players experience the worst peak where veterans experience the best one |
| 120 | - one audience sees relief while another sees boredom |
| 121 | - the ending hits differently depending on skill, investment, or outco |