How to use GPT Image Generator in your AI agent

What you'll get

This skill lets your AI assistant generate and edit images with OpenAI's GPT Image 2 and 2.5 models, backed by a large gallery of proven prompts. It handles posters, typography-heavy graphics, product shots, diagrams, edits based on reference images, and touch-ups to specific areas of a picture. It helps you pick the right model and quality level for your budget and saves the finished images for you.

Step 1 — Get the skill source URL

The GPT Image Generator skill lives at:

Step 2 — Add it to your AI agent

Copy one of these prompts and paste it into your AI tool. The agent will fetch and install the skill for you.

Add the agent skill from https://github.com/wuyoscar/GPT-Image2-Skill/tree/main/skills/gpt-image to my project.

Install this agent skill: https://github.com/wuyoscar/GPT-Image2-Skill/tree/main/skills/gpt-image

Add this skill to .codex/skills/: https://github.com/wuyoscar/GPT-Image2-Skill/tree/main/skills/gpt-image

Add the skill at https://github.com/wuyoscar/GPT-Image2-Skill/tree/main/skills/gpt-image to my .gemini/skills/ directory.

Install this agent skill from https://github.com/wuyoscar/GPT-Image2-Skill/tree/main/skills/gpt-image

Add this skill: https://github.com/wuyoscar/GPT-Image2-Skill/tree/main/skills/gpt-image

Step 3 — Try these prompts

Once the skill is installed, ask your AI things like:

• You need a poster, social graphic, or product image with accurate text on it
• Edit or restyle an existing photo or design using a reference
• Exploring visual directions for a campaign or presentation

What this skill does

• Generates images from a text description in square, portrait, landscape, or wide formats
• Edits existing images using one or more reference pictures
• Repaints a selected area of an image while leaving the rest untouched
• Draws on a curated gallery of prompts for posters, typography, and visual styles
• Helps you choose the model and quality setting to balance cost and polish