agentleFS
sebnadosebnado/worldbuild-bench/skills/game-quality/SKILL.md

game-quality

Skill9 starsChanged 4 months agoMITView on GitHub

The quality bar — what separates a shippable game from a prototype — across lighting, world geometry, VFX, audio, game feel, UI, and performance. Outcomes to hit on every front, with technique families worth exploring; how you hit them is your call.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. wormholeportal/Makone★ 27same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository6 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If you cannot finish everything well, cut whole features cleanly — fewer things, finished — instead of shipping everything at prototype …

    line 110

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

The Quality Bar

You are not building a tech demo that minimally satisfies a checklist. The bar is a game a player would voluntarily keep playing: readable at a glance, juicy to interact with, coherent as a world. Flat-shaded boxes on a green plane do not clear that bar. Everything runs offline from the bundled lib/three/ (the full examples/jsm/ addons tree is available) and the WebAudio API. No external assets exist — geometry, materials, and audio are all made in code, which is a style to embrace (clean, bold, stylized) rather than apologize for.

Each section below sets the bar for one front and names technique families worth exploring — how you hit the bar is yours to decide. Nothing ships as a placeholder: every asset the player sees or hears — geometry, materials, music, sound — is a final product, composed and finished to the best of your ability, not a stand-in. For the arrangement of the world itself, see the world-design skill.

Lighting & atmosphere (the cheapest 10x visual upgrade)

The bar: the scene reads as lit — a definite light direction, shadows that ground objects, depth cues receding toward a designed sky, and a deliberate palette that reads as art direction rather than defaults. Get color management right (output color space, tone mapping) so the colors that ship are the colors you chose.

Worth exploring: physically-based materials and HDR lighting, image-based / environment lighting, global-illumination approximations (ambient occlusion, light probes), screen-space reflections, area lights, shadow mapping, fog matched to the sky, gradient or procedural skies, emissive accents.

Geometry & world building

The bar: shapes with designed silhouettes and materials that read as finished art at gameplay distance — a world that looks authored, not accidental. Repeated elements vary enough that the repetition doesn't read.

Worth exploring: procedural geometry and procedural texturing (generated color / normal / roughness maps — made at runtime or baked to files in assets/ during the build), composition and sculpting of primitives, signed-distance-field modeling, instancing and geometry merging, per-instance variation, level-of-detail.

VFX, shaders, post-processing

The bar: the picture moves — impacts burst, speed leaves traces, the objects that matter draw the eye. Every effect serves a moment the player cares about; nothing runs as decoration for its own sake.

Worth exploring: GPU-instanced or pooled particle systems, trails, small targeted custom shaders where the player actually looks, a restrained post-processing chain.

Animation & motion

The bar: things that move look alive. Locomotion reads as locomotion — parts that should articulate do, secondary motion follows — never a statue sliding across the floor. Motion has weight (acceleration, lean, recoil, follow-through), and births and deaths are animated events, not object removal.

Worth exploring: procedural animation, skeletal or keyframe animation, inverse kinematics, physics-driven secondary motion, squash and stretch.

Audio (WebAudio, all synthesized)

Sound is a design layer, not a checklist. Decide a sonic palette that matches the world (what does this place sound like?), let intensity follow game state (layers and tempo rise with the stakes), keep a mix hierarchy (music ducks under SFX, SFX under stingers), and use silence deliberately. Aim for music and effects that sound produced — composed, layered, mixed; a player should assume they are hearing crafted audio, not raw oscillators.

Environment facts: one AudioContext, unlocked on the first user input (browsers block autoplay before a gesture); route everything through a master gain.

Worth exploring: synthesis for SFX (subtractive, FM, physical modeling), spatial/positional audio, convolution reverb (impulse responses can be generated in code), generative or adaptive layered music, mapping continuous sounds to game state, randomized variation so repeated sounds don't fatigue.

Game feel ("juice") — the difference players actually notice

The bar: every interaction acknowledges the player. Nothing pops into existence unanimated, nothing important happens without feedback the player can feel, and the big moments (wins, losses, near-misses) have anticipation and payoff.

Worth exploring: reactive cameras (smoothed follow, speed response, impact shake), spring-damper motion, tweened transitions, layered hit feedback (visual, camera, and audio landing together).

UI/HUD

The bar: every state — menu, countdown, playing, paused, won, lost — is a designed screen with readable hierarchy, never an alert() or raw unstyled text. Values the player watches visibly react when they change. (A DOM overlay usually out-typesets canvas text for the same effort — your call.)

Performance (60 fps is a feature)

The bar: a steady frame rate on a mid-range machine, verified in a playtest — renderer.info.render.calls tells the truth about draw calls.

Known traps: movement must be delta-time-based, with dt clamped so a backgrounded tab doesn't teleport physics on return; per-frame allocations in hot loops kill frame pacing — pool and reuse; unbatched repeated geometry multiplies draw calls.

Scope discipline

If you cannot finish everything well, cut whole features cleanly — fewer things, finished — instead of shipping everything at prototype quality.

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.