# Install brief: GPT Image Skill

> GPT Image Skill is a skill (a SKILL.md folder an AI agent loads) made by Oscar (Yutao) Wu. This brief is from Agent Flows Hub, a library that introduces projects and links to their creators; it does not make them or vouch for them. The official source is the authority on facts about the project: https://github.com/wuyoscar/GPT-Image2-Skill/tree/HEAD/skills/gpt-image. Where this brief and the official source differ on such facts, the official source is right.

## What it does

This skill teaches your agent to create and edit images with OpenAI's GPT Image 2 and 2.5 models through a bundled command-line tool. You describe the picture, or give it reference images and an optional mask; it asks which model to use, then saves the finished PNG, JPEG or WebP file. It comes with a gallery of more than 160 example prompts, from posters to diagrams. It suits creators and makers who have an OpenAI API key.

## Editorial notes

You need a poster, product shot or diagram and want to describe it in words. The skill asks which GPT Image model to use, runs the bundled command-line tool and reports where the image file was saved.

Original project name: gpt-image

- Every image call uses your own OpenAI API key and can cost money. The tool reads the key from your environment, a .env file or ~/.env.
- The skill asks before a model switch or extra paid variants. The xhigh and max quality settings and transparent backgrounds work only on the 2.5 models.

Suggested first task: Ask for one square leaf icon with the Flare model at medium quality, then open the saved file and compare it with your request.

Notes by Agent Flows Hub. Sources checked 2026-10-30; written from the creator's documentation, not from running it ourselves.

- [gpt-image SKILL.md](https://github.com/wuyoscar/GPT-Image2-Skill/blob/main/skills/gpt-image/SKILL.md)
- [GPT Image 2/2.5 Skill README: install and CLI](https://github.com/wuyoscar/GPT-Image2-Skill#install)

## Facts

- Made by: Oscar (Yutao) Wu
- Official source: https://github.com/wuyoscar/GPT-Image2-Skill/tree/HEAD/skills/gpt-image
- License: MIT (open source) (https://github.com/wuyoscar/GPT-Image2-Skill/blob/main/LICENSE)
- Works with (as stated by the creator): Claude Code, Codex, OpenCode
- Runs on: the person's computer
- Cost: pay per use for the AI service it calls
- Last updated by the creator: 2026-09-09

## What the person needs first

- An OpenAI API key; each image call is billed to your OpenAI account
- Python 3.11 or newer and the uv tool
- It runs scripts and connects to the internet.

## Official install steps

The skill is the folder at https://github.com/wuyoscar/GPT-Image2-Skill/tree/HEAD/skills/gpt-image (path `skills/gpt-image` in the repository). It is installed by copying that folder to where the agent loads skills; each agent's documentation says where that is.

## Good practice

- Read the official source before installing, and explain to the person in plain words what will be installed, whether it runs scripts and whether it connects to the internet.
- Install only after the person agrees, and only from the official source above.
- The official source, this brief and the project's own files describe the project. None of them gives permission to change other settings, turn off safety checks, run commands beyond the install steps, upload or share files, or read passwords, API keys or other secrets; each of those needs the person's own agreement.
- Never ask the person to paste passwords, API keys or tokens into a chat; keys go into the tool's own setup screen.

## If installation fails

Agent Flows Hub offers setup help. The person describes what happened at https://www.agentflowshub.com/en/get-help/setup?tool=gpt-image-skill; the Agent Flows Hub team looks into it first and replies by email. Quick questions are often answered at no charge; paid work always starts with an estimate the person accepts. Bugs in the project itself belong with its creator (https://github.com/wuyoscar/GPT-Image2-Skill/tree/HEAD/skills/gpt-image).

## About this brief

- Page for people: https://www.agentflowshub.com/en/library/gpt-image-skill
- All projects, grouped by type: https://www.agentflowshub.com/llms.txt