| name | gamification |
| description | Use when designing a deep engagement or progression system — XP, levels, progression curves, quests, missions, streaks, points and reward economies, badges, achievements, variable rewards, leaderboards, seasons/events, or applying Octalysis to shape a target behavior; "gamify this app", "add XP/levels", "design a progression system", "make it more engaging/sticky", "improve retention with mechanics". NOT for pure UX clarity/friction ("why is this confusing?", "simplify this flow", "users don't know what to do next" → dont-make-me-think skill), NOT for pure visual styling (colors, typography, "make it prettier" → impeccable/frontend-design), NOT for backend-only work with no user-facing engagement surface. |
Gamification
Design a progression/engagement system that reinforces a real target behavior. Three ideas drive every decision:
- Behavior-first — every mechanic exists to reinforce ONE mapped target behavior. A mechanic you can't justify against the Behavior Brief is decoration: cut it.
- Concrete balancing — curves, formulas, and starting numbers, not vibes. An XP system without a curve is a wish.
- Ethics Lens — every mechanic gets an honest White↔Black label with its real trade-off. The skill informs; the user sets the dial. Guide, don't police.
The unit of work is the LOOP (a system of mechanics feeding each other), not an isolated badge.
Announce at start: "Using gamification to design [system] reinforcing [target behavior]."
The Process — 3 phases, 1 gate
UNDERSTAND ──► Behavior Brief ──► DESIGN (+Ethics Lens) ──► 🚦 user approves? ──► BUILD ──► verify the loop
│ no
▼
revise blueprint
Phase 1 — UNDERSTAND → Behavior Brief
Read reference/understand.md and follow it. Map the product's core loop, the user's real win, the target behavior to reinforce, and the dominant player profile (Hexad). If gamification already exists, audit it (Octalysis-label what's there; separate working mechanics from decoration). Produce a Behavior Brief. Do NOT propose mechanics yet.
Phase 2 — DESIGN → approval gate 🚦
Read reference/psychology.md (the why), reference/mechanics.md (the full menu), reference/balancing.md (the numbers), reference/ethics.md (the lens), then reference/blueprint.md (the output format). Select mechanics from the menu — each justified against the target behavior, each labeled by the Ethics Lens, each with starting parameters. Present the blueprint and STOP. Do not touch code until the user explicitly approves.
Phase 3 — BUILD → verified loop
Implement the approved blueprint following the repo's existing patterns. Wire up the success metrics defined in the blueprint. Verify the LOOP works end to end — earn, progress, level, reward, return — not just that it compiles. Delegate pure visuals (palette, typography, celebration art) to impeccable/frontend-design when it makes sense: gamification owns the system, not the pixels. Commit in atomic units per the repo's conventions.
Calibration — declare both dials at the top of every blueprint
Dial 1 — Surface intensity (how much mechanic fits this product):
| Surface | Examples | Mechanics dose |
|---|
| Serious tool | admin panels, tax/finance, internal dashboards | Progress/completion + celebrated real wins only. Deep mechanics feel absurd here. |
| Consumer product | e-commerce, SaaS, content apps | + streaks-lite, badges on real milestones, onboarding-as-level-1 |
| Engagement app / game | habit trackers, education, fitness, games | The full menu is on the table — selected by the Behavior Brief, never all at once |
Dial 2 — White↔Black ethics dial (how aggressive): default is White-Hat (sustainable, trust-building). The user can push toward Black-Hat (urgent, compelling, corrosive) — deliver what they choose with eyes open: label it, state the trade-off once, never silently escalate. See reference/ethics.md.
Red flags — stop and reconsider
| Thought | Reality |
|---|
| "XP would be cool here" | No mapped target behavior = decoration. Justify or cut. |
| "One more mechanic won't hurt" | Every mechanic is cognitive + maintenance cost. The burden of proof is on adding. |
| "I'll skip the Ethics Lens, it's obviously fine" | The user must SEE what they chose. Label every mechanic, even pure White-Hat. |
| "A no-forgiveness streak retains better" | True — and it punishes having a life. Conscious dial choice, never a default. |
| "Variable rewards on everything" | The most addictive and most corrosive mechanic. Careful handling, explicit label. |
| "Reward the thing users already love doing" | Overjustification effect: extrinsic rewards can kill existing intrinsic motivation. Gamify what's hard to start, not what's already fun. |
| "Novelty will carry it" | Every extrinsic mechanic decays. No intrinsic transition + endgame plan = dead system in weeks. |
| "One mechanic fits all users" | Hexad profiles respond differently. Leaderboards delight Achievers and repel Free Spirits. |
| "This is a gamification problem" (but it's UX friction) | Clarity problems ≠ progression problems. Hand off → dont-make-me-think. |
| "Skipping the gate, the design is obvious" | The blueprint gate is mandatory. Present, then wait. |
Sibling skill — dont-make-me-think
dont-make-me-think owns UX clarity and friction (make the journey obvious); this skill owns deep engagement mechanics. Two-way handoff: it sends deep-mechanics work here; you send clarity/friction problems there. If it isn't installed on this machine, say so and continue — this skill is fully self-sufficient.
