mcpbeat

AI Game Art Pipeline Agent Skill

Provider-neutral open-source skill for planning and producing game-runtime art assets and animation: static props/icons, canonical character sheets, combat sprites, 3D/video motion references, Veo/Seedance-style video-to-frames, cinematic ultimates, backgrounds, FX, SFX, runtime packaging, and QA. Use when building actual game assets for PixiJS/Canvas/Unity/Godot/SpriteKit/etc., not one-off marketing images.

229k tokens
context cost
the whole folder, loaded on every use
27
files
ships runnable scripts
0
copies elsewhere
how many repositories repackaged it
313
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/ybuild-ai/ai-game-art-pipeline-skill --skill ai-game-art-pipeline

The instruction itself

8 sections, as written by the author

AI Game Art Pipeline

Use this skill when a user wants to generate, plan, debug, or package game art that must run in an actual game runtime.

The skill is provider-neutral. It does not assume any specific API key or vendor SDK. If a model call is needed, use the user's available provider through an adapter or ask them to implement scripts/provider_stub.py.

Core Rule

Pick the pipeline by runtime job, not by model hype.

| Runtime job | Pipeline |

|---|---|

| Static props, items, FX icons | Image model + style refs + removable background |

| Existing/brand character | Reuse canonical sprite sheets before regenerating |

| Combat character animation | Keyframe-first + 2D grid + post-processing + curation |

| Walk/run/body mechanics | Prefer video motion reference; use 3D skeleton only when cleanup is acceptable |

| Ambient loops | Video model -> extracted frames |

| Cinematic ultimate / boss intro | Full-screen video-to-frames + code-driven hit logic |

| Runtime feel | Deterministic code: hit pause, shake, particles, trails, SFX |

Default Workflow

  • Identify the runtime job and target engine.
  • Check whether canonical assets already exist.
  • Pick one scenario reference below.
  • Generate the smallest useful vertical slice, not the full asset set.
  • Post-process with deterministic scripts.
  • Preview at target size and target device.
  • Iterate surgically: rerun one animation, reprocess raw output, or repack frames.

Scenario References

Read only the relevant file:

  • references/static-assets.md — props, icons, items, simple monsters, canonical asset reuse.
  • references/battle-sprites.md — keyframe-first combat sprite sheets, 2D grids, anchors, contact sheets.
  • references/motion-video.md — 3D skeleton pitfalls, video motion references, Veo/Seedance-style video-to-frames, cinematic ultimates.
  • references/backgrounds.md — level backgrounds, master-first parallax, outpaint chains, mobile texture limits.
  • references/runtime-shipping.md — formats, compression, CDN layout, audio/SFX, runtime QA.

Bundled Scripts

Scripts are local utilities only. They do not call model APIs.

  • scripts/provider_stub.py — adapter interface the user can implement for their provider.
  • scripts/chroma_key_magenta.py — remove magenta background from generated sheets/icons.
  • scripts/sheet_contact.py — make numbered contact sheets for curation.
  • scripts/extract_video_frames.py — extract/resize video frames for runtime texture sequences.

Provider Adapters

Do not write a full vendor SDK unless the user explicitly asks for one. Prefer a thin adapter around scripts/provider_stub.py so credentials, endpoints, retries, billing, and model choices stay in the user's project.

If the user wants examples, point them to examples/providers/README.md and adapt only the minimal file they need:

  • fake/local provider for testing the pipeline without model calls.
  • generic image HTTP provider for a user's image-generation proxy.
  • generic async video HTTP provider for a user's video-generation proxy.

Output Standard

When using this skill, produce:

  • A concise pipeline choice and why.
  • The exact assets to generate or reuse.
  • Prompt constraints if model generation is needed.
  • Post-processing commands/scripts.
  • Runtime packaging notes: dimensions, format, anchor, frame rate, texture limits.
  • QA checklist for the target device.

Hard Warnings

  • Do not regenerate a known character if a canonical sheet exists.
  • Do not use 1xN horizontal strips for complex combat poses; use 2D grids.
  • Do not let cinematic video own gameplay semantics. Damage, hitboxes, target tracking, cooldown, and SFX belong in code.
  • Do not ship giant backgrounds because they look fine on desktop. Check mobile texture limits.
  • Do not treat AI output as final. The production asset is the generated image plus cleanup, anchors, timing, compression, and runtime test.

Other skills for the same job

different authors, same section of the catalogue
Canvas Design
by anthropics
vendor ×13

Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.

1388k tokens
Algorithmic Art
by anthropics
vendor ×10

Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.

15k tokens scripts
Image Enhancer
by frostant
×6

Improves the quality of images, especially screenshots, by enhancing resolution, sharpness, and clarity. Perfect for preparing images for presentations, documentation, or social media posts.

635 tokens
Video Downloader
by CommandCodeAI
×4

Downloads videos from YouTube and other platforms for offline viewing, editing, or archival. Handles various formats and quality options.

671 tokens
Histolab
by christophacham
×3

Lightweight WSI tile extraction and preprocessing. Use for basic slide processing tissue detection, tile extraction, stain normalization for H&E images. Best for simple pipelines, dataset preparation, quick tile-based analysis. For advanced spatial proteomics, multiplexed imaging, or deep learning pipelines use pathml.

18k tokens
Omero Integration
by christophacham
×3

Microscopy data management platform. Access images via Python, retrieve datasets, analyze pixels, manage ROIs/annotations, batch processing, for high-content screening and microscopy workflows.

32k tokens
Pydicom
by christophacham
×3

Python library for working with DICOM (Digital Imaging and Communications in Medicine) files. Use this skill when reading, writing, or modifying medical imaging data in DICOM format, extracting pixel data from medical images (CT, MRI, X-ray, ultrasound), anonymizing DICOM files, working with DICOM metadata and tags, converting DICOM images to other formats, handling compressed DICOM data, or processing medical imaging datasets. Applies to tasks involving medical image analysis, PACS systems, radiology workflows, and healthcare imaging applications.

13k tokens scripts
Transformers
by christophacham
×3

This skill should be used when working with pre-trained transformer models for natural language processing, computer vision, audio, or multimodal tasks. Use for text generation, classification, question answering, translation, summarization, image classification, object detection, speech recognition, and fine-tuning models on custom datasets.

13k tokens

How to use it

Copy the folder

Take ybuild-ai/ai-game-art-pipeline from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.