Skip to content
Agentmart
Submit a tool

ue5-pcg-building

UE5.6/UE5.7 PCG building generation workflow for modular buildings, blockouts, facade rules, and runtime generation.

Skill by IvanYangYangXi

IndexedMIT★ 36
Write a review

Chalk Labs

Need help deploying ue5-pcg-building?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/ivanyangyangxi-ue5-pcg-building

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

UE5.6/UE5.7 PCG building generation workflow for modular buildings, blockouts, facade rules, and runtime generation. Use when requests involve Procedural Content Generation (PCG), Shape Grammar, lot-based building spawn, deterministic random seeds, density/filter pipelines, or converting designer constraints into reusable PCG graphs.

Use cases: writing, productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/ue5-pcg-building

What's inside

Skill name
ue5-pcg-building
Description
UE5.6/UE5.7 PCG building generation workflow for modular buildings, blockouts, facade rules, and runtime generation. Use when requests involve Procedural Content Generation (PCG), Shape Grammar, lot-based building spawn, deterministic random seeds, density/filter pipelines, or converting designer constraints into reusable PCG graphs.
License
MIT
Instructions
139 lines in SKILL.md · view on GitHub

Files (6)

  • SKILL.md
  • SOURCE.md
  • pcg-building-graph-patterns.md · references/
  • project-adapter.md · references/
  • runtime-generation-validation.md · references/
  • shape-grammar-building-pattern.md · references/

SKILL.md (first lines)

# Quick Start
- Define generation target: blockout towers, modular facades, or lot-based building sets.
- Define deterministic inputs: lot splines/points, district tags, style preset, and seed.
- Define runtime mode: static bake, on-demand, or runtime scheduled generation.
- Define output mode: Static Mesh instances first, Spawn Actor only for interactive/stateful parts.

# UE5.7 API Anchors
- Runtime trigger and radii live on `UPCGComponent`:
  - `EPCGComponentGenerationTrigger::GenerateAtRuntime`
  - `bOverrideGenerationRadii`, `GenerationRadii`, `SchedulingPolicyClass`, `SchedulingPolicy`
  - `GenerateLocal(...)`, `Cleanup(...)`
- Runtime scheduler refresh lives on `UPCGSubsystem`:
  - `RefreshRuntimeGenComponent(...)`
  - `RefreshAllRuntimeGenComponents(...)`
  - `CleanupLocalComponentsImmediate(...)`
- Output selection anchor classes:
  - `UPCGStaticMeshSpawnerSettings` for high-count rendering
  - `UPCGSpawnActorSettings` for interactive/stateful outputs
- Shape Grammar anchor classes:
  - `UPCGSubdivisionBaseSettings::GrammarSelection`
  - Do not rely on deprecated grammar fields (`bGrammarAsAttribute_DEPRECATED`, `Grammar_DEPRECATED`)

# Graph Stage Contract
- Every stage must explicitly declare:
  - Input data type (`EPCGDataType` or asset source)
  - Core node/classes (minimum two)
  - Required parameters (seed, tags, ranges, radii, or style keys)
  - Output data type
  - Debug method (node-level checks, debug node, or log/assert path)
- If a stage cannot satisfy these five items, treat the graph design as incomplete.

# Workflow
## 1) Input
- Input data type: actor/spline/point sources.
- Core node/classes: `UPCGDataFromActorSettings`, `UPCGGetActorPropertySettings`, `UPCGCreatePointsSettings`.
- Required parameters: lot tag filters, district/style tags, seed source, source bounds.
- Output: normalized lot point or spline data with stable ordering.
- Debug method: run `UPCGDebugSettings` after input stage and verify point count and bounds.

## 2) Filter
- Input data type: point/spline data from Input stage.
- Core node/classes: `UPCGAttributeFilteringSettings`, `UPCGDensityFilterSettings`, `UPCGFilterByTagSettings`.
- Required parameters: slope range, exclusion tags, min lot area/width, occupancy constraints.
- Output: only buildable lots/candidates.
- Debug method: compare candidate count before/after filter and inspect rejected tag distribution.

## 3) Transform
- Input data type: filtered buildable candidates.
- Core node/classes: `UPCGCopyPointsSettings`, `UPCGCreateSplineSettings`, `UPCGApplyScaleToBoundsSettings`.
- Required parameters: floor height, pivot convention, facade orientation basis, local axes.
- Output: footprint transforms and per-floor transforms.
- Debug method: inspect transform axes and floor index attributes on output points.

## 4) Grammar
- Input data type: segment/spline/point data from Transform stage.
- Core node/classes: `UPCGSubdivideSplineSettings`, `UPCGSubdivideSegmentSettings`, `UPCGSelectGrammarSettings`.
- Required parameters: `GrammarSelection`, module size limits, style-based grammar key mapping.
- Output: grammar-resolved module placements/attributes.
- Debug method: use `UPCGPrintGrammarSettings` for grammar parse and token validation.
- Rule: use `GrammarSelection` only; avoid deprecated grammar fields.

## 5) Output
- Input data type: grammar-resolved placements.
- Core node/classes: `UPCGStaticMeshSpawnerSettings`, `UPCGSpawnActorSettings`, `UPCGCreateTargetActor`.
- Required parameters:
  - Static path: mesh selector, instance packer, ISM/HISM policy.
  - Actor path: actor class, spawn attributes, state/interaction requirements.
- Output: rendered buildings and optional interactive building elements.
- Debug method: split output by layer/tag and validate per-layer counts.
- Default policy: prefer Static Mesh Spawner; use Spawn Actor only when stateful behavior is required.

## 6) Validate
- Input data type: final spawned result and runtime generation state.
- Core node/classes: `UPCGDebugSettings`, `UPCGComponent`, `UPCGSubsystem`.
- Required parameters: expected cell bounds, max per-update spawn budget, nav/collision expectations.
- Output: pass/fail signals and fix actions.
- Debug method: run staged checks for overlap, navigation impact, per-cell generation time, and deterministic replay.

# Constraints
- Keep the main pipeline compatible with both UE5.6 and UE5.7 unless a version-specific note is required.
- Runtime generation must explicitly set:
  - `GenerationTrigger = GenerateAtRuntime`
  - explicit `GenerationRadii` (do not rely on implicit defaults)
  - explicit `SchedulingPolicyClass` for predictable scheduler behavior
- Prefer ISM/HISM style output for large counts; avoid spawning heavyweight actors for each small part.
- Keep runtime generation bounds explicit to avoid uncontrolled world-wide regeneration.
- Avoid hidden dependency on editor-only data when runtime generation is expected.
- Treat World Partition boundaries as hard constraints for runtime scopes.

# Failure Handling
- Symptom: no buildings spawn.
  - Locate: Input stage output count, source bounds, lot tags.
  - Fix: verify source actor/spline ingestion and lot filter tags; confirm non-empty candidate set.
- Symptom: output exists in editor preview but not runtime.
  - Locate: `GenerationTrigger` and runtime radii/scheduling settings.
  - Fix: set `GenerateAtRuntime`, radii override, and valid scheduling policy.
- Symptom: runtime update regenerates too wide an area.
  - Locate: runtime radii and generation source movement.
  - Fix: reduce generation/cleanup radii and tighten source bounds.
- Symptom: stale generated pieces remain after rules shrink.
  - Locate: cleanup path and local component lifecycle.
  - Fix: trigger cleanup with remove-components behavior and force local cleanup when needed.
- Symptom: heavy hitching during runtime generation.
  - Locate: points-per-cell, actor spawn count, per-update workload.
  - Fix: reduce per-cell complexity, cap actor spawns, move non-interactive parts to static mesh instances.
- Symptom: deterministic replay mismatch with same seed.
  - Locate: unstable upstream point ordering or non-seeded random branch.
  - Fix: normalize ordering before random selection and bind every stochastic path to explicit seed inputs.
- Symptom: facade grammar fails or produces empty modules.
  - Locate: grammar parse logs and module token mapping.
  - Fix: validate grammar string, module dictionary, and segment size constraints.
- Symptom: overlap and collision issues.
  - Locate: filter thresholds and final placement constraints.
  - Fix: add clearance/slope filters and occupancy rejection before output stage.
- Symptom: navmesh degradation around generated buildings.
  - Locate: collision profile and nav-affecting flags on spawned outputs.
  - Fix: split nav-affecting vs non-nav-affecting outputs and rebuild nav only where required.
- Symptom: runtime changes do not apply after parameter edits.
  - Locate: scheduler refresh flow.

From IvanYangYangXi/artclaw_bridge, MIT.

More from IvanYangYangXi/artclaw_bridge

Signals

Stars
36
Forks
7
Open issues
1
Last commit (this folder)
Apr 1, 2026 (7 months ago)
Last release
Apr 15, 2026 (6 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from IvanYangYangXi

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get ue5-pcg-building running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying ue5-pcg-building?

CallGet help
Cite this page

ue5-pcg-building skill. Agent Mart. https://agentmart.dev/skills/ivanyangyangxi-ue5-pcg-building (last updated 2026-10-24)