AgentKit
DocsKitsCLI ReferenceDesktop App
BetaYou are reading docs for the beta channel (2.13.0-beta.20). Features may change before the next stable release.Switch to stable →

Skills

Design ethical gamification with ak:gamification-marketing

Select bounded mechanics, model reward economics and abuse, plan implementation, and measure behavior without dark patterns.

Use ak:gamification-marketing to design a points, badge, leaderboard, level, streak, challenge, quest, unlockable, reward, or progress mechanic for a marketing outcome. The Skill can produce campaign strategy, sample configuration, implementation guidance, analytics events, and a measurement plan. Mechanics remain proposals until you approve product, campaign, reward, and data changes.

Choose ak:gamification-marketing for a bounded behavior

Use ak:gamification-marketing when

  • You have an acquisition, retention, engagement, conversion, or onboarding behavior to support and evidence that the behavior creates user value.
  • You need to compare mechanics against autonomy, competence, relatedness, ability, timing, abuse risk, and reward economics.
  • You want a phased design, event schema, rules outline, or implementation brief.
  • You can define opt-out, accessibility, fairness, support, and harm guardrails.

Choose another workflow when

  • The underlying offer or journey is unclear. Use ak:funnel first.
  • The issue is specifically post-signup activation. Use ak:onboarding-cro before adding game mechanics.
  • You already selected one experiment. Use ak:ab-test-setup for the test plan.
  • The requested mechanic depends on addiction, deception, coerced sharing, gambling-like rewards, fake scarcity, or loss threats. Redesign the objective rather than optimizing that pattern.

Prepare behavior, reward, and system context

Provide the audience, desired behavior, user benefit, current baseline, segment and eligibility rules, proposed mechanic, reward inventory and monetary value, budget and liability constraints, jurisdictions, age considerations, abuse model, accessibility requirements, data policy, technical stack, and success and harm metrics. Remove credentials and unnecessary user records.

Complete Onboarding, and confirm Marketing Kit is installed for the runtime and scope you are using.

RuntimeInvocationAvailability boundary
Claude Code/ak:gamification-marketing ...Native delivery is the default; explicit plugin delivery is also supported.
Cursor/ak:gamification-marketing ...Slash invocation is user-verified; broader runtime parity is not implied.
Codex$ak:gamification-marketing ...The Skill uses native Codex discovery; Hook projection is partial.

Run the Skill

/ak:gamification-marketing "Design a draft 30-day onboarding challenge using at most two mechanics. Use real progress, voluntary participation, capped non-cash rewards, accessibility and anti-abuse rules. State assumptions and do not implement, notify users, or launch"

Understand the design stages

  1. Identify the goal and user value. The Skill connects the desired behavior to acquisition, retention, engagement, conversion, or onboarding and checks that success benefits participants as well as the business.
  2. Select a small mechanic set. It compares points, badges, leaderboards, levels, streaks, challenges, quests, unlockables, rewards, progress, and social sharing; it should start with the fewest mechanics that can test the hypothesis.
  3. Align motivation and ability. The Skill can use self-determination, behavior, and progression frameworks as design lenses, not proof of effect.
  4. Design rules and campaign. It outlines eligibility, earning, redemption, expiry, tiers, milestones, difficulty, messaging, reset, recovery, and anti-abuse behavior.
  5. Plan implementation. Packaged references include illustrative JSON, PostgreSQL, API, cache, leaderboard, and event-schema patterns. They are starting points, not a required stack or production-ready contract.
  6. Measure and review. The Skill defines engagement, completion, retention, conversion, reward cost, fraud, complaints, opt-outs, fairness, and negative feedback, then recommends a phased experiment.

Protect autonomy, fairness, and reward economics

Do not optimize compulsion

Loss aversion, variable rewards, scarcity, streaks, and leaderboards can create pressure or exclusion. Do not use them to obscure odds, punish rest, threaten earned value, coerce contact sharing, target vulnerable people, or make basic access depend on participation. Legal and policy review may be required.

  • Make participation and sharing voluntary, with understandable rules and an accessible opt-out that does not remove unrelated product value.
  • Show actual progress, eligibility, odds, expiry, reward value, caps, taxes or fees where relevant, and what happens when a campaign ends.
  • Offer non-competitive or private alternatives to public rankings; prevent harassment, shaming, and exposure of personal activity.
  • Model issuance, redemption, breakage, fraud, support, accounting liability, inventory, and worst-case reward cost before launch.
  • Rate-limit and make awards idempotent; define dispute, correction, rollback, and audit paths before connecting rules to purchases or referrals.
  • Obtain consent for analytics and notifications, minimize identifiers, and avoid sensitive behavioral profiling.
  • Approve code, database migrations, provider calls, notifications, rewards, campaign publication, and spend separately from the design.

Review the outputs and evidence

The Skill declares this report path:

assets/reports/performance/{date}-gamification-analysis.md

A complete artifact should include the behavior hypothesis, user benefit, selected mechanics and rejected alternatives, rule table, reward economics, abuse model, consent and accessibility requirements, phased rollout, event definitions, decision metrics, harm guardrails, ownership, and implementation dependencies. Treat all example targets, schedules, case-study figures, prize amounts, schemas, endpoints, and infrastructure values in packaged references as illustrative until verified for your context.

Troubleshoot the workflow

SymptomSafe next step
The design contains many mechanicsReduce it to one behavior hypothesis and one or two mechanics for the first phase.
Engagement rises while complaints or compulsive use risePause expansion, review harm guardrails and participant research, and remove coercive mechanics.
Rewards exceed the modeled budgetStop issuance or rollout through the approved control path, reconcile liability, and revise caps and eligibility.
Leaderboards invite abuse or exposureUse private progress, cohorts, teams, aliases, moderation, and opt-out, or remove the leaderboard.
The runtime cannot find the SkillConfirm target and scope, restart the runtime, then follow Runtime cannot find a Skill or Agent.

Know the limits

  • The Skill does not guarantee engagement, retention, conversion, revenue, ROI, or habit formation.
  • Packaged case-study numbers and psychology claims are not current independent evidence and are intentionally not performance promises here.
  • The implementation references omit many production requirements, including authorization, privacy deletion, financial controls, concurrency, regional rules, observability, and incident response.
  • The documented package releases contain identical ak:gamification-marketing content, references, and invocations.

Continue with ak:ab-test-setup for an approved mechanic experiment, or review the Marketing Kit overview.