This project provides a Codex Skill designed to transform textual content into visually distinct, hand-drawn technical explanation-style PNG images, suitable for articles, course materials, and concept visualization. It targets technical writers, educators, and AI content creators seeking a unique, narrative-driven visual output beyond standard templates. The primary benefit is generating a cohesive set of explanatory graphics with AI-assisted content structuring and a consistent visual identity.

How It Works

The ian-handdrawn-ppt skill leverages an AI agent within the Codex environment to process input materials (articles, notes, outlines). It first analyzes the content to establish a narrative structure, then maps each point to appropriate semantic page layouts and a locked "hand-drawn technical explanation" visual DNA. The process prioritizes content understanding and page-level storytelling over simple template application, culminating in the generation of full-page raster PNG images. Recommended models include ChatGPT Image 2.0 for optimal style consistency.

Quick Start & Requirements

1. Clone the repository:

   git clone https://github.com/helloianneo/ian-handdrawn-ppt.git
   cd ian-handdrawn-ppt
   

2. Copy the skill directory to your Codex skills path:

   mkdir -p "${CODEX_HOME:-$HOME/.codex}/skills"
   cp -R ./ian-handdrawn-ppt "${CODEX_HOME:-$HOME/.codex}/skills/"
   

• Prerequisites: A functional Codex environment.
• Recommended Model: ChatGPT Image 2.0.
• Documentation: Examples and usage prompts are available in examples/prompts.md.

Highlighted Details

• Generates 21:9 aspect ratio cover images and 16:9 aspect ratio main content/standard page images.
• Adheres to a specific "Ian's Chinese hand-drawn technical explanation" visual style: off-white background, thin hand-drawn lines, minimal borders, restrained titles, and concise Chinese text.
• Employs a structured workflow involving content intake, narrative planning, selection of page archetypes (e.g., comparison, flow, matrix), and locking visual elements.
• Outputs include individual PNG pages, a contact sheet for multi-page sets, and a slide blueprint summary.

Maintenance & Community

The project is maintained by Ian (伊恩), a Product Designer and AI Builder. Related resources include "Awesome Claude Code Skills" and "Claude Code Handbook." Further engagement and information can be found via his website (ianneo.xyz) or X/Twitter handle (@ianneo_ai). He also runs a "Content Creation | AI Workflow Watch Group."

Licensing & Compatibility

The project is released under the MIT License, which permits broad use, modification, and distribution, including for commercial purposes and integration into closed-source projects.

Limitations & Caveats

This skill exclusively outputs PNG raster images; it does not generate editable PPTX, PDF, Keynote, HTML, SVG, or vector graphics. AI image generation may produce textual errors, hallucinations, or incorrect visual relationships, necessitating multiple generations and careful selection. Rendering complex Chinese text can be unstable, potentially requiring text reduction or post-processing. The tool prioritizes semantic content transformation and narrative visualization over template-based presentation.