codex-pet

Generate Codex-compatible pet spritesheets and manifests from a single reference image.

Updated Jun 11, 2026
One-click install
npx skills add https://github.com/brillianodhiya/VisionScript --skill codex-pet-brillianodhiya
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codex-pet
Source: https://github.com/brillianodhiya/VisionScript/tree/main/.agents/skills/codex-pet
Command: npx skills add https://github.com/brillianodhiya/VisionScript --skill codex-pet-brillianodhiya

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @runcomfy/cli, imagemagick.

What problem does it solve? Creating a custom Codex Pet normally requires Codex Pro and the internal $imagegen skill; this Skill produces the exact same pet.json and spritesheet.webp artifacts using only a RunComfy token, one GPT Image 2 edit call, and ImageMagick. ## Core Features & Use Cases - Single-call canonical pose generation: Calls runcomfy run openai/gpt-image-2/edit once with a chibi-proportion prompt on a magenta chroma-key background. - Deterministic atlas assembly: Uses ImageMagick micro-transforms (shifts, blinks, rotations, mirroring) to build all 9 animation rows into the 1536x1872, 8x9-cell spritesheet Codex expects. - Automatic installation: Writes pet.json and spritesheet.webp into ${CODEX_HOME:-$HOME/.codex}/pets/<name>/ so Codex picks up the custom pet next to the 8 built-ins. - Use Case: Turn a photo of your cat into an animated desktop companion for OpenAI Codex in about two minutes, without a Codex Pro subscription. ## Quick Start Ask the assistant to create a Codex Pet from your source image URL and install it into your Codex pets directory.

Frequently Asked Questions about codex-pet

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a custom Codex Pet without Codex Pro?▼

Use this Skill with the RunComfy CLI and a RUNCOMFY_TOKEN. It makes one GPT Image 2 edit call to create a canonical chibi pose, then assembles the full spritesheet with ImageMagick, producing the same pet.json and spritesheet.webp files Codex expects.

What files does a custom Codex Pet need?▼

Each custom Codex Pet folder contains exactly two files: pet.json (a manifest with id, displayName, description, and spritesheetPath) and spritesheet.webp, a 1536x1872 atlas of 8 columns by 9 rows of 192x208 cells with a transparent background.

How does this compare to OpenAI's official hatch-pet skill?▼

The official hatch-pet skill requires Codex Pro and the internal $imagegen system skill. This Skill produces an identical artifact via the RunComfy Model API, needing only a RUNCOMFY_TOKEN plus the runcomfy and magick binaries.

Why does the Codex Pet generation use a magenta background?▼

GPT Image 2 only outputs RGB with no alpha channel, so the prompt requests a solid #FF00FF chroma-key background. ImageMagick then converts that magenta to transparency with an 18% fuzz before trimming and fitting the sprite into each cell.

Can I use a pet, object, or non-human subject as the source image?▼

Yes, pets, mascots, objects, and foods all work. The base prompt simplifies any source image into the chibi pixel-art house style automatically, though small identity details like accessories may drift.

What are the limitations of the ImageMagick animation approach?▼

Animation comes from micro-transforms of one canonical pose, not multi-frame model generation, so motion is subtle (1-2 px shifts, blinks, tilts) rather than dramatic frame-by-frame movement. This matches the built-in Codex Pets' animation style.