Skip to content

Codex Transparent Asset Generator

Makes transparent PNG game art, UI ornaments and emblems via built-in Codex image gen, then checks alpha.

Repository
MengTo/Skills
Stars
7,187
Licence
MIT
Files
3 files

What it does

When you ask for transparent artwork such as ornaments, sprites, emblems or UI graphics, the agent follows a structured workflow: it writes a detailed image prompt, asks the built-in Codex image tool to generate the picture, and then runs a small Python helper to confirm the PNG really has transparent areas. It also keeps the original file, records the prompt and file hashes, and can copy assets into a project's folders. The skill itself does not create images; it directs the agent to the image tool and checks the results.

Great for

  • Game developers who need transparent PNG sprites, corners or crests for interface layouts
  • Designers producing isolated UI graphics that must stack cleanly over backgrounds
  • Teams that want a documented record of how each generated asset was made and verified

Skip it if

  • Anyone needing a model choice or a specific model version: the built-in tool exposes no model selector, so the skill cannot guarantee which model ran.
  • Environments without the built-in Codex image generation tool: the skill says to stop and explain the blocker rather than switch providers.
  • Users who need vector or hand-drawn SVG output: the skill explicitly avoids substituting vector assets.

Try asking

  • “Generate a transparent PNG corner ornament for my fantasy game menu and verify its alpha channel.”
  • “Create a centered crest emblem with transparent holes for my UI, and save the prompt and hashes alongside it.”
  • “Edit this existing sprite with the built-in image tool, keeping the original as a reference, then check transparency.”
Step-by-step guide: how to use Codex Transparent Asset GeneratorInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 3 original files, the MIT licence and an attribution note.

    codex-gpt-image-2-5-flare.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip codex-gpt-image-2-5-flare.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip codex-gpt-image-2-5-flare.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Describe the silhouette, padding and transparent openings explicitly in your request so the generated image needs fewer edits.
  • Ask for one asset per request when you need separate usable files; the skill avoids sprite sheets unless you ask.
  • Always visually check enclosed gaps on a checkerboard or your real background, since the helper only measures alpha values.

What's inside

  • agents/openai.yaml
  • scripts/inspect_png.py
  • SKILL.md

Redistributed unmodified from MengTo/Skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.