threejs-aaa-graphics-builder
Skill0 starsChanged 5 months agoMITView on GitHub
Upgrade Three.js games from basic/prototype visuals to premium AAA-inspired browser graphics. Combines art-direction critique, procedural model building, mandatory external asset sourcing decisions, threejs-3d-generator assets, threejs-image-generator concept/texture workflows, scene visual polish, material/texture libraries, world prop kits, VFX, render pipeline, and visual scorecard gates. For premium games with characters, vehicles, ships, weapons, buildings, signature props, skies, textures, decals, logos, icons, or GUI art, load the relevant generator skills before deciding procedural assets are enough.
2 other projects use a version of this, 2 of them edited.
Adopted by
Most recently taken up by
- iberi22/skillsedited, first seen Oct 15, 2026
- majidmanzarpour/threejs-game-skillsedited, first seen Sep 28, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 2
- majidmanzarpour/threejs-game-skills★ 2.2k11% same
- iberi22/skills★ 011% same
Wrote their own version · 0
None found.
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 repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- threejs-3d-generatorSkill
- threejs-audio-generatorSkill
- threejs-debug-profilerSkill
- threejs-game-directorSkill
- threejs-gameplay-systemsSkill
- threejs-game-ui-designerSkill
- threejs-image-generatorSkill
- threejs-qa-releaseSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 5 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 34
Run the director credential probe before using key unavailable as a skip reason and paste the SET/MISSING output.
- Distinctive: line 35
Create an asset sourcing ledger for each high-value surface: procedural / threejs-image-generator / threejs-3d-generator / hybrid, plus …
- Distinctive: line 43
Add missing graphics architecture: material library, procedural textures/decals, model factories, world prop kit, VFX system, render …
- Very specific: line 44
Run the credential probe, then fill the external asset sourcing ledger per surface: procedural Three.js factory, threejs-image-generator 2D …
- Very specific: line 46
Add lighting/render/material polish after authored forms exist.
- Distinctive: line 47
Add event-driven VFX tied to gameplay state.
- Distinctive: line 49
Verify renderer diagnostics, desktop/mobile screenshots, console/page errors, canvas pixels, imported asset budgets, and playability.
- Distinctive: line 53
Do not make primitives look AAA by adding glow.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.
Files
Three.js AAA Graphics Builder
Purpose
Own the production graphics pass. Convert basic screenshots into authored, high-density, performance-aware visual experiences.
Use When
Screenshots still look basic, models look primitive, worlds are sparse, UI/world art feels generic, or the user asks for premium, AAA, high-fidelity, showcase, or less-basic graphics.
Required References
These references are required phase-entry gates, not optional reading:
- Load
references/visual-scorecard.mdbefore scoring, judging completion, or making any premium/AAA/showcase claim. - Load
references/implementation-blueprint.mdbefore changing graphics architecture, materials, VFX, rendering, diagnostics, or broad visual systems. - Load
references/model-recipes.mdbefore building or upgrading hero/player, obstacle, enemy, pickup, world-kit, material, or prop models. - Load
references/render-recipes.mdbefore changing lighting, tone mapping, shadows, fog, post-processing, materials, or render composition. - Load
references/checklists/aaa-game-quality-gate.mdandreferences/checklists/aaa-visual-scorecard.mdbefore declaring a game premium, AAA, showcase, complete, release-ready, or less basic. - Load the relevant checklist before focused work:
references/checklists/procedural-model-quality.md,references/checklists/material-lighting-quality.md, orreferences/checklists/performance-safe-visual-detail.md. - Load
references/prompt-templates.mdonly when the user asks for reusable prompts, a graphics-pass prompt, or a task template.
For broad "still looks basic", premium, AAA, high-fidelity, showcase, or less-basic graphics work, load all four references as the first action in the phase. Track them in a reference ledger with yes/no, path, and failure reason. Do not mark the graphics phase complete while any required reference is skipped.
External asset sourcing gate:
- For premium/AAA/showcase/high-fidelity/less-basic graphics with a hero/player, character, creature, boss, vehicle, ship, building, weapon, signature prop, complex pickup, or hero environment piece, load
threejs-3d-generatorbefore deciding procedural geometry is enough. - For premium/AAA/showcase/high-fidelity/less-basic graphics with concept needs, texture/material references, decals, logos, faction marks, icons, GUI art, skies, backgrounds, title/menu art, or image-to-3D inputs, load
threejs-image-generatorbefore deciding 2D external assets are not needed. - Run the director credential probe before using
key unavailableas a skip reason and paste the SET/MISSING output. - Create an asset sourcing ledger for each high-value surface: procedural / threejs-image-generator / threejs-3d-generator / hybrid, plus outputs or skip reason.
not-neededis valid only after the relevant skill was loaded and the ledger explains why external generation would not improve a non-hero support surface, or why the credential probe or attempted generation shows a real blocker.- For premium hero surfaces, procedural-only is not an allowed final answer unless there is real blocker evidence. At least one high-value surface must show a 3D generator task ID, downloaded GLB/GLTF/FBX path, image generator output path, or documented hybrid chain.
Workflow
- Capture or inspect active desktop/mobile screenshots.
- Score visuals across art direction, hero/player, obstacles, rewards, world, materials, render, VFX, UI, and performance evidence.
- Add missing graphics architecture: material library, procedural textures/decals, model factories, world prop kit, VFX system, render pipeline, diagnostics.
- Run the credential probe, then fill the external asset sourcing ledger per surface: procedural Three.js factory,
threejs-image-generator2D reference/texture,threejs-3d-generator3D generation, or a hybrid. - Upgrade every weak visible surface, not only one hero object.
- Add lighting/render/material polish after authored forms exist.
- Add event-driven VFX tied to gameplay state.
- Re-score screenshots. Continue until every premium category is at least 2/3 or report exact blockers.
- Verify renderer diagnostics, desktop/mobile screenshots, console/page errors, canvas pixels, imported asset budgets, and playability.
Core Rule
Do not make primitives look AAA by adding glow. First build authored forms, then materials, then lighting, then effects.
Final Response
Report the reference ledger, credential probe output, external asset sourcing ledger, score before/after, production surfaces upgraded, files changed, screenshots/artifacts, renderer diagnostics, imported asset diagnostics when relevant, and remaining blockers. For premium/AAA/showcase claims, include the filled visual scorecard exactly as defined in references/visual-scorecard.md, including average and automatic failures remaining.
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
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.

