ppt-image-first

ppt-image-first

A conversation-first, image-first PPT workflow skill that progressively advances PPT requirements to a finished product.

Agent SkillDesignOpen source
Type
Agent Skill
Open source
Yes
GitHub Stars
★ 1.1k
Source
skill-github

Overview

ppt-image-first is a PPT workflow skill designed for Codex/Claude Code/Opencode CLI. It uses lightweight conversation to understand requirements, build content foundations, generate real previews, iteratively refine styles, and ultimately produce high-fidelity visual presentation pages. The output is image-first by default: using GPT Image 2 to generate full-page visuals, then packaging them into a PPTX container for delivery. Ideal for presentations, reports, and further image-level retouching, but the text, graphics, and decorative elements within the slides are typically not individually editable like native PPT elements.

Capabilities

  • Lightweight conversation-based requirement understanding
  • Content foundation report generation
  • Multiple style direction previews
  • Style refinement and confirmation
  • High-fidelity visual draft generation

Use cases

Defense presentation preparationResearch reportingProject reportingProduct introduction

Setup

Requires: API Key
Please refer to the installation instructions in the GitHub repository for configuration.

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

What scenarios is this skill suitable for?

Suitable for defense presentations, research reports, project updates, product introductions, and similar use cases.

Is the output PPT editable?

The output is a high-fidelity visual draft; elements within the slides are typically not editable.

Related skills