All skills
gpt-image-2
Render a structured GPT Image 2 prompt from 80+ templates, then generate or edit the image.
- Part of ConardLi/garden-skills
- 6.5k installs
- Media & Design · AI images
When to use it
You have a rough image idea but need a precise, reusable prompt for a poster, UI mockup, product shot, or diagram.
For example
You say
Use gpt-image-2 to make a live-commerce UI mockup for a skincare brand's evening serum, with host, chat, gift area, and product card.
You get
- Rendered prompt file
- Generated UI mockup image
How it works
- Picks the closest template from 18 categories instead of reading all references.
- Asks targeted questions about missing fields like host identity or product price.
Before installing
End-to-end image generation needs ENABLE_GARDEN_IMAGEGEN enabled and an OPENAI_API_KEY; otherwise it only produces prompts.
This skill includes 4 scripts your assistant may run on your computer.
More AI images skills
See all- higgsfield-generate259kGenerates images, videos, 3D models and audio, and edits or animates existing media.
- higgsfield-product-photoshoot236kGenerates studio, lifestyle and ad-ready product photos, including model shots and creative variations.
- seo-image-gen7.3kGenerates SEO images such as social previews, blog heroes, schema images and infographics with AI.
- generate-image2.2kGenerates or edits photos, illustrations, logos and visual assets with AI image models.
Also in this pack
- web-design-engineer10.3kBuilds or redesigns web pages, dashboards, prototypes, slide decks and data visualizations with HTML, CSS, JavaScript or React.
- web-video-presentation6.7kTurns an article or script into a click-through 16:9 web presentation that feels like a video, with optional narration audio.
- kb-retriever5.8kSearches your local knowledge base folder, reading PDFs and spreadsheets in pieces, and answers questions from it.
- beautiful-article4.5kTurns a URL, PDF, DOCX, Markdown file or pasted text into a single shareable HTML article you can open offline.