GPT-Image2-Skill
GPT Image 2 prompt gallery, image prompt library, agentic skill, and CLI for OpenAI image generation/editing
GPT-Image2-Skill is an agent skill library that provides OpenAI GPT Image 2 and 2.5 generation and editing capabilities through both a CLI tool and agentic skill interfaces. It includes a curated gallery of 163 prompts across 31 categories (anime, product photography, UI/UX mockups, research figures, etc.) that serve as reference templates for creating high-quality images. The skill supports both text-to-image generation and reference-image-based editing workflows, with model selection between gpt-image-2, gpt-image-2.5-flare (fast, high-quality), and gpt-image-2.5-sunburst (precise edits). Developers can install it as a skill in agent runtimes or use the standalone CLI for direct OpenAI image API access with quality/size controls and multi-image editing.
Key Features
Use Cases
- 01Generating product photography, UI mockups, or marketing visuals directly from agent conversations
- 02Creating research paper figures, data visualizations, and technical diagrams with precise labeling
- 03Building brand identity systems, character design sheets, and multi-panel visual narratives
- 04Editing existing images by providing reference photos and transformation prompts (e.g., winter scene, style transfer)
- 05Extracting reusable prompts from existing images to recreate similar styles or compositions
- 06Rapid prototyping of poster layouts, infographics, or editorial illustrations with quality/budget control
Related Skills
View moresuperpowers
An agentic skills framework & software development methodology that works.
skills
Skills for Real Engineers. Straight from my .agents directory.
skills
Public repository for Agent Skills
ponytail
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
GPT-Image2-Skill — FAQ
What is GPT-Image2-Skill?+
GPT-Image2-Skill is an agent skill and CLI library for OpenAI's GPT Image 2 and 2.5 models. It provides two skills—gpt-image for generation/editing and get-prompt-from-image for prompt extraction—along with a 163-prompt gallery and comprehensive prompting guides.
How do I install it in Claude Code?+
Run '/plugin marketplace add wuyoscar/gpt_image_2_skill' then '/plugin install gpt-image@wuyoscar-skills' in Claude Code. For Codex, invoke the built-in $skill-installer with the GitHub skill-folder URL.
Which AI clients and runtimes does it support?+
It works with Claude Code, Codex, OpenClaw, Hermes Agent, and other skill-capable runtimes. The CLI can also be used standalone with uvx or uv tool install.
Do I need an OpenAI API key?+
Yes, you must set OPENAI_API_KEY in your environment, .env, or ~/.env. The skill reads the key from process env, then .env, then ~/.env without overriding an already-set variable.
Is it free to use?+
The skill library and CLI are free and MIT-licensed. However, you pay OpenAI for image generation/editing based on model, quality, and resolution; use --quality low for cheaper drafts.
Can I edit existing images or only generate new ones?+
Both. Use -i to attach one or more reference images for editing via the /v1/images/edits endpoint. You can also provide a mask with -m for inpainting (opaque pixels are preserved, transparent pixels are regenerated).
How do I install GPT-Image2-Skill?+
Open the source repository on GitHub and follow its README. GPT-Image2-Skill is a skill — MCP Agents Market links you directly to the official repo.
Is GPT-Image2-Skill free?+
GPT-Image2-Skill is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.