gpt-image2-ppt-skills

A skill that builds decks with gpt-image-2: each slide is drawn as a full layout and packed into .pptx, optionally mimicking a reference template

SkillMedium risk

juneyaooo/gpt-image2-ppt-skills

Install

git clone git@github.com:JuneYaooo/gpt-image2-ppt-skills.git ~/.claude/skills/gpt-image2-ppt-skills

For Codex clone into ~/.codex/skills/gpt-image2-ppt-skills/. Then set OPENAI_API_KEY.

This is third-party code. Review the repository files before installing.

What it does

From a Markdown outline or slides_plan.json the skill renders each slide as a 16:9 image via the OpenAI Images API and packs them into .pptx. You can pick a bundled style or supply a .pptx or image to reuse its layout, palette and visual language with new content. On explicit request an editable mode rebuilds text and basic shapes as PowerPoint objects and moves complex artwork to a separate layer. A multimodal model reviews the finished pages.

Who it is for. For marketers, founders and anyone who needs a visually strong deck fast without a designer.

Good fit when

  • You need a deck with strong design from ready text
  • You have a template you like and need a new deck in its style
  • You need a cover or a few slides for a pitch

Not a fit when

  • Slides need heavy, frequent manual editing
  • You need exact data-driven charts
  • You have no access to the OpenAI API or a compatible gateway

Example request

Make a 6-slide deck on how AI changes content creation and show me the cover for approval first

Limitations

Requires OPENAI_API_KEY or a compatible gateway, and image generation is paid. By default slides are images with non-editable text. Editable mode and template cloning need PowerPoint, Keynote or LibreOffice installed. Requires Python. The OpenAI API is not reachable from Russia without a VPN. Main docs are Chinese with an English translation.

How to disable. Delete the gpt-image2-ppt-skills folder from ~/.claude/skills or ~/.codex/skills.

Security check

  • Runs Python scripts and paid OpenAI API calls
  • Sends slide content and templates to an external service

README in short

The README shows template cloning and editable mode examples and explains when editable mode applies and what it needs. Install by asking the agent or cloning into the Claude Code or Codex skills folder. It documents environment variables for generation and a separate review model. The project grew out of NanoBanana-PPT-Skills and moved to gpt-image-2. Apache-2.0 licensed.

SKILL.md

---
name: gpt-image2-ppt
description: Generate visually striking PPT slides via OpenAI's gpt-image-2 -- use any style in styles/<collection>/STYLE_ID.md or mimic a user-supplied .pptx template; outputs high-res slide PNGs and a 16:9 .pptx. Use when the user asks to make a presentation, slides, deck, pitch deck, investor PPT, magazine-style PPT, or 做一份 PPT / 生成幻灯片 / 用 gpt-image 生成 PPT / 按这个模板生成 PPT.
---

# gpt-image2-ppt -- 用 gpt-image-2 生成 PPT

把一份 markdown 大纲(或 `slides_plan.json`)+ 一种视觉风格,直接喂给 OpenAI 官方 Images API(`gpt-image-2`),逐页出图,最后打包成 16:9 .pptx。

## 可用风格

| 风格 ID | 一句话定位 | 适用场景 |
| --- | --- | --- |
| `gradient-glass` | Apple Vision OS / Spatial Glass | AI 产品发布、技术分享、创意提案 |
| `clean-tech-blue` | Stripe / Linear 级蓝白 | 融资路演、商业计划书、企业战略 |
| `vector-illustration` | 复古矢量插画 + 黑描边 | 教育培训、品牌故事、社区分享 |
| `editorial-mono` | Kinfolk / Monocle 编辑设计 | 品牌发布、文化访谈、读书分享 |
| `dark-aurora` | Linear / Vercel 深色霓虹 | AI 产品、开发者工具、技术分享 |

FAQ

Can I avoid calling OpenAI directly?

Yes, OPENAI_BASE_URL points to any compatible gateway, and page review can use a separate multimodal model via VISION_BASE_URL.

Will the skill read project secrets?

According to the authors, the script reads only process variables, an explicit GPT_IMAGE2_PPT_ENV and the skill folder's .env, never walking up project directories.

Editors’ pick

Microsoft's utility and MCP server that turn PDF, Word, Excel, PowerPoint, HTML and other files into Markdown for models

MCP serverMedium risk184K
Editors’ pick

Anthropic's official skill examples: docx, pdf, pptx and xlsx documents, design, MCP builder and the Agent Skills spec

SkillMedium risk176.3K
Editors’ pick

A skill that builds editable PPTX decks from PDF, DOCX, URLs and Markdown with native shapes, charts and animations

SkillMedium risk54.3K
Editors’ pick

Anthropic's legal plugin suite: contracts, privacy, employment, IP, litigation, AI governance and legal education

PluginMedium risk9.4K