gpt-image

Generate photorealistic and concept visuals for CSA presentations via Azure OpenAI GPT-image-2.

2|3|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/huqianghui/csa-ppt-plugin --skill gpt-image
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gpt-image
Source: https://github.com/huqianghui/csa-ppt-plugin/tree/main/skills/gpt-image
Command: npx skills add https://github.com/huqianghui/csa-ppt-plugin --skill gpt-image

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Generates photorealistic and concept visuals for CSA presentations to supplement architecture diagrams and hand-drawn assets, reducing the need for stock photos or manual illustration.

Core Features & Use Cases

  • Photorealistic images and concept art generated by Azure OpenAI GPT-image-2 for title slides, industry scenarios, and hero backgrounds.
  • Multilingual text rendering in generated images (EN/CN/JP/KR) to support multilingual slide decks.
  • Seamless integration with the CSA-PPT workflow and other skills (azure-diagrams, excalidraw-diagram) to provide the "visual expression layer" that programmatic tools cannot produce.
  • Image editing / inpainting support for refining visuals to fit slide themes.

Quick Start

Run the inline generate_image routine with a prompt and an output path to create a photorealistic slide image.

Frequently Asked Questions about gpt-image

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate photorealistic images for presentation slide decks?

You can generate photorealistic presentation images by running the inline generate_image routine with a text prompt and output path, leveraging Azure OpenAI GPT-image-2 to produce PNG, JPEG, or WebP files for slide decks.

Does this image generation workflow support multilingual text rendering?

Yes, multilingual text rendering supports English, Chinese, Japanese, and Korean directly within the generated images, enabling the creation of multilingual infographics and slide decks without manual translation overlays.

What do I need to set up before generating AI images for CSA presentations?

You need Python 3.8 or higher, the requests dependency, and access to an Azure OpenAI GPT-image-2 deployment to specify image size, quality, and output formats before generating visuals for CSA presentations.

Can I use generated AI images alongside architecture diagrams in my slides?

Yes, generated AI images integrate seamlessly with the CSA-PPT workflow and complement other skills like azure-diagrams and excalidraw-diagram, providing the visual expression layer for cover slides and brand backgrounds alongside architecture diagrams.

How do I refine an existing AI-generated visual to match my slide theme?

You can refine existing AI-generated visuals using the image editing and inpainting support features, allowing you to modify specific areas of an image to better fit your presentation's slide theme and visual requirements.

What are the limitations of using AI-generated concept art for slide backgrounds?

AI-generated concept art supplements but does not replace architecture diagrams or hand-drawn assets, and generation requires active Azure OpenAI GPT-image-2 deployment access with predefined size and quality parameters for each visual.