agentleFS
szlayszlay/roblox-base/.claude/skills/roblox-building/SKILL.md

roblox-building

Skill1 starsChanged 4 months agoMITView on GitHub

Use when building Roblox geometry, maps, props, or generated assets with MCP or standalone Luau.

2 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 1 is distinctive, and 4 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.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read back after every asset, script, or geometry batch before continuing.

    line 20
  2. Very specific:

    Use search_asset + insert_asset for Creator Store assets when provenance and licensing are acceptable.

    line 25
  3. Very specific:

    Use generate_procedural_model for configurable primitive-part props and blockouts.

    line 27
  4. Very specific:

    Use generate_mesh for custom textured props and generate_material for surface variants.

    line 28
  5. Very specific:

    Use store_image/upload_image only for permitted images.

    line 29

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

When to Load

Load when building physical geometry, maps, props, or environment assets in Roblox Studio through MCP or standalone scripts. Covers spatial planning, CSG, generated/reused assets, scale, and verification. See references/full.md for complete patterns.

Quick Reference

MCP Build Mode

  1. Discover Studio/context, then inspect Workspace and existing asset conventions.
  2. Establish a named build root, origin, coordinate system, and geometry/asset manifest.
  3. Build in bounded phases: ground → zone shells → landmarks → props → environment.
  4. Read back after every asset, script, or geometry batch before continuing.
  5. Validate structure, capture a deliberate view, and playtest traversal when relevant.

Asset Choice

  • Inspect and reuse a compatible existing asset first.
  • Use search_asset + insert_asset for Creator Store assets when provenance and licensing are acceptable.
  • For cross-owner or paid assets, surface creator/source/price and get explicit consent before insertion.
  • Use generate_procedural_model for configurable primitive-part props and blockouts.
  • Use generate_mesh for custom textured props and generate_material for surface variants.
  • Use store_image/upload_image only for permitted images; pass a returned generationId to wait_job_finished before dependent work.
  • Fall back to native Parts/CSG and report unavailable or unsuitable generation.

Player Scale

Player ~5 studs | Door 4w×7h | Ceiling 10-14 | Counter 3.5-4 | Seat 1.5 | Path 6+

Spatial Rules

  • Named dimensions manifest, no magic numbers.
  • Offset sub-parts from anchor CFrames, not guessed world coordinates.
  • Snap to 0.125/0.25/0.5 studs.
  • Build complex CSG near origin, then PivotTo the destination.
  • Set Anchored, CanCollide, CastShadow, Color, and Material explicitly.

Acceptance

Prop: named model, pivot, scale, bounds, materials, collision, anchoring, no loose parts, asset provenance. Map: root/origin, zones, landmarks, spawns, path widths, traversal, and bounds checks excluding Baseplate/Terrain/SpawnLocation. Evidence: structural readback plus screenshot when supported, console/runtime result when playtested.

Anti-Patterns

Guessing coords | unanchored parts | hardcoded world positions | silent CSG failures | >20-30 parts/call | default gray | claiming generation succeeded without readback

Need detail? Load references/full.md for CSG wrappers, map structure, validation scripts, asset recipes, and evidence workflows.

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.