# alayarenderer-generative-world

An agent skill that walks you through AlayaRenderer: splitting RGB video into G-buffers and restyling footage from a text prompt.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 564 installs · ★ 88
- Author: reason-machines
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/alayarenderer-generative-world/SKILL.md
- Updated: 2026-10-09
- HTML: https://affpapa.org/en/nevibecode/skills/alayarenderer-generative-world

## What it does

AlayaRenderer works in two stages. The Inverse Renderer takes ordinary RGB footage and pulls out five channels: albedo, normal, depth, roughness and metallic. Game Editing then takes those G-buffers plus a text prompt and produces a restyled clip.

The skill covers the whole setup for your agent. You clone the repo with submodules, create two conda environments on Python 3.10, and pull weights with `huggingface-cli`. The models differ in size — Cosmos-Transfer1-DiffusionRenderer 7B handles inverse rendering, Wan2.1 1.3B does the stylization — and their dependencies clash, which is why the two environments stay apart. Weights land in `inverse_renderer/checkpoints/Diffusion_Renderer_Inverse_Cosmos_7B` and `game_editing/models/train/Wan2.1-T2V-1.3B_gbuffer`.

Game Editing inference runs through `examples/wanvideo/model_inference/inference_gbuffer_caption.py`, driven by flags such as `--style`, `--prompt`, `--gbuffer_dir`, `--num_frames`, `--height` and `--width`. Frame counts must follow the `8n+1` pattern; anything else and the run fails. The G-buffer directory is built by hand: five subfolders of per-frame PNGs.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill alayarenderer-generative-world -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/alayarenderer-generative-world/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/alayarenderer-generative-world/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill alayarenderer-generative-world -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/alayarenderer-generative-world/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill alayarenderer-generative-world -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/alayarenderer-generative-world/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install reason-machines/trending-skills alayarenderer-generative-world
```

2. **Manually: put the files in place**
- File: `.github/skills/alayarenderer-generative-world/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill alayarenderer-generative-world -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/reason-machines/trending-skills --path skills/alayarenderer-generative-world
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/alayarenderer-generative-world/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill alayarenderer-generative-world -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/alayarenderer-generative-world/SKILL.md

## When to use it
- split RGB footage into albedo, normal, depth, roughness and metallic maps
- turn a gameplay recording into a snowy or night scene from text
- set up two conda environments for models with clashing dependencies
- download Cosmos-Transfer1 and Wan2.1 weights into the right paths
- lay out a G-buffer folder of per-frame PNGs for inference

## Example prompts
- Run the inverse renderer on `gameplay_video.mp4` and write the G-buffers to `test_dataset/`
- Generate a game editing clip with style `snowy_winter` and a prompt about a frozen forest
- Help me set up the `game_editing` environment and fetch the `Brian9999/stylerenderer` weights

## Similar
- [skill-creator](https://affpapa.org/en/nevibecode/skills/skill-creator.md) — skill · 414180 installs
- [writing-for-agents](https://affpapa.org/en/nevibecode/skills/writing-for-agents.md) — skill · 466719 installs
- [media-use](https://affpapa.org/en/nevibecode/skills/media-use.md) — skill · 1100000 installs
- [mcp-builder](https://affpapa.org/en/nevibecode/skills/mcp-builder.md) — skill · 130949 installs
- [subagent-driven-development](https://affpapa.org/en/nevibecode/skills/subagent-driven-development.md) — skill · 237746 installs
- [orchestration](https://affpapa.org/en/nevibecode/skills/orchestration.md) — skill · 349800 installs

From the same set: [reason-machines/trending-skills](https://affpapa.org/en/nevibecode/by/reason-machines/trending-skills.md) — 139 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md