agentleFS
ForgeyClapForgeyClap/Forge-Core/.claude/skills/roblox-player-psychology/SKILL.md

roblox-player-psychology

Skill2 starsChanged 2 days agoMITView on GitHub

Use when designing or reviewing player-behavior mechanics: first minute, rewards, streaks, daily rewards, pricing, RNG/pity, community.

1 other project uses a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Roblox Player Psychology

Roblox binding (read first; it wins)

roblox-economy-ethics wins over this skill on every price, luck, reward or pressure question. Nothing bought with Robux gives a random outcome, luck, power or speed. No purchase prompt in the first session or during play, no countdown, fake scarcity, "value" anchor or decoy next to a price, no reward that expires and no streak that resets. The effects below explain how players behave: use them to make playing and earning feel good, never to make stopping or not paying feel bad. About a third of age-checked daily users are under 13.

When to Load

Load for behavior/persuasion questions: what happens in the first 60 seconds, why players return, reward schedules, streaks, pricing psychology, gamepass ladders, crate/RNG odds and pity, or turning players into a community. For structural craft (core loop, FTUE structure, level design, economy sources/sinks, retention phases) load roblox-game-design. Dashboard diagnosis, discovery and the algorithm have no pack skill yet: say so and work from roblox-game-design plus the Roblox analytics docs. For purchase implementation load roblox-economy-ethics (what may be sold) and roblox-security (receipts and purchase exploits).

Quick Reference

First 60 seconds (50% of new players never return)

0–5s spawn in-world (no logo/cutscene) → 5–10s do the core verb → 10–30s first reward lands → 30–45s open a loop (next milestone looks close) → 45–60s show what is ahead (a free preview of the next unlock; no shop prompt in the first session). Six effects: Zeigarnik (hand an incomplete goal in 10s), endowed progress (never start at zero), peak-end (best effect in minute one), cognitive load (one mechanic/one button/one goal), isolation (only the wanted action glows), reciprocity (give before the shop).

Retention benchmarks & levers

D1 20/30/40% · D7 8/15/20% · D30 3/7/10% (good/great/excellent; compounds; 40% D1 ≈ 3× the players of 25% D1 by day 30). Low number → lever: PTR→curiosity gap/social proof; D1→Zeigarnik, endowed progress; session→flow, variable reward; D7→goal gradient, collections; D30→habit loop, social identity; conversion→a fair first purchase, charm pricing (no anchors or decoys); ARPPU→ladders, visible status.

Hooks, pricing, RNG

Four reward schedules (fixed/variable × ratio/interval); layer all four. Hooks: appointments, streaks that pause on a missed day instead of resetting, rotation of free items, shared clan goals, hook model, specific timers ("Back in 4h 32m"). Ladder: 49–99 first purchase · 199–399 convenience and capped time boosts · 499–799 bigger cosmetic sets and private servers · 999+ identity; never luck, power or speed. RNG stays free (rolls earned by play): 5 tiers (55/27/12/5/1%), show the odds, the pool and the pity counter. Paid random items are out (roblox-economy-ethics rule 1).

Full detail, tables, and community loop: references/full.md

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.