Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add zajalist/hayba --skill hayba-pcg-buildgit clone --depth 1 https://github.com/zajalist/haybaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zajalist/hayba/hayba-pcg-build)<a href="https://agentmods.dev/skills/zajalist/hayba/hayba-pcg-build"><img src="https://agentmods.dev/badge/skills/zajalist/hayba/hayba-pcg-build/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/zajalist/hayba/hayba-pcg-build"><img src="https://agentmods.dev/badge/skills/zajalist/hayba/hayba-pcg-build.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00047 | $0.01089 |
| Opus 5.5 | $0.00019 | $0.00436 |
| Sonnet 5.5 | $0.00009 | $0.00218 |
| Haiku 4.5 | $0.00005 | $0.00109 |
Grade A, and why
hayba-pcg-build scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 22d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hayba PCG build
When to use
Use for a new or existing PCG/PCGEx graph when the result must be repeatable and verified in a real editor. The production method is still a choice to make from the scene requirements.
Decide before authoring
- State the desired scene outcome and its constraints: target platform, frame-time and memory budgets, streaming boundaries, instance and draw-call expectations, texel density, deterministic regeneration, edit frequency, and designer override needs. Record unknown budgets as unknown; do not invent a pass/fail threshold.
- Compare viable methods before choosing a graph. Include manual/Actor Palette placement or reusable composed assets, native PCG, PCGEx, and a hybrid where relevant. Manual composition gives precise art direction but costs placement and maintenance time. Native PCG supports repeatable broad distribution with fewer dependencies but may need bespoke graph logic. PCGEx offers specialized graph, cluster, path, and staging operations but adds a plugin/version dependency and more graph complexity. A hybrid can reserve hero assets and exceptions for manual control while generating the repeatable field, at the cost of a clear ownership boundary. For each candidate, note streaming and regeneration costs, correction workflow, and why it fits or fails this scene.
- Look for a relevant standard production pattern when evidence is available, then propose at least one structural alternative, such as changing route, visibility, or asset grouping rather than only tuning density. Label analogies as design hypotheses until tested in this project.
Tool routing
Ground the graph
- If PCGEx is a candidate, call
hayba_query_pcgex_docsfor the task concepts and candidate classes, thenhayba_get_pcgex_knowledge_detailfor exact IDs. Record the bundle generation date, documented plugin version, source hash/stale flag, and any warnings or conditions. TreatplaceableInPcgGraphas a documentation classification, not proof that the node is installed. The bundle is optional; an unavailable result is an evidence gap, not permission to guess pins or settings. - Use
hayba_get_pattern_templateonly as an annotated starting point. Its preflight is catalog-only and can report absent or unknown pins. In the target scratch editor, usepcg_list_node_classesandpcg_get_node_detailsto confirm loaded settings classes, default pins/types, and properties. Compare the installed PCGEx.upluginversion with the documented version. Resolve a mismatch from live reflection or defer the uncertain connection; never silently convert a documentation pin into a wire. - Sketch the graph as JSON with explicit asset/component ownership. Use
pcg_validate_graphand resolve its errors beforepcg_create_graph. Before wiring a created node, usepcg_list_pinson that graph to check its exact current labels and types, including dynamic pins. A validator pass is only a structural check; it does not establish that the graph produces the intended scene or meets performance budgets.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 22d ago Changed · +22 lines · +15 tokens per session 4c1e1df39c92
- 1mo ago First seen · 17 lines · 32 tokens per session scan A c2911f4a652f
hayba-pcg-build is a skill published in the GitHub repository zajalist/hayba (13 stars, last pushed yesterday), licensed MIT. It adds 47 tokens to every session and 1,089 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
unreal-mcp
Automate Unreal Engine editor scenes, actors, and renders.
asset-management
Import files from disk (images, meshes, any format the editor imports), export textures, query the Content Browser selection, and check if an asset is open (AssetDiscoveryService). Search, load, save, move, rename, duplicate, and delete assets are handled by Unreal's native AssetTools toolset or EditorAssetLibrary.…
level-actors
List, find, place, move, rotate, and modify actors in the current level (ActorService + editor subsystems). Use when the user asks to find/list actors in the level, move or rotate an actor, set an actor's properties/folder/transform, or spawn/delete level actors.
vibeue
Unreal Engine 5 development using the VibeUE Python API. Use when working in Unreal Engine — blueprints, state trees, materials, actors, landscapes, animation, niagara, widgets, sound, foliage, gameplay tags, enhanced input, skeletons, PCG (procedural content generation), and more. VibeUE is an extension of Unreal's…
materials
Create and edit materials and material instances — graph nodes, parameters, functions, custom HLSL, and instance overrides (MaterialService + MaterialNodeService). Use when the user asks to create or edit a material/material instance, wire material nodes, add material parameters, set blend/shading modes, or recreate a…
project-settings
Configure Unreal Engine project settings and editor preferences — UI appearance, toolbar icons, scale, colors, and any UDeveloperSettings subclass (ProjectSettingsService). Use when the user asks to change a project setting or editor preference, configure a UDeveloperSettings category, or read/write project config…