Back to directory

Ian Handdrawn PPT

A Codex skill that turns articles, course notes or outlines into a set of Chinese hand-drawn technical explainer images, delivered as full-page PNGs rather than an editable slide deck.

Ian Handdrawn PPT

Ian Handdrawn PPT is a Codex skill that guides an AI agent to turn written material, such as an article, course outline or script, into a set of full-page images in a Chinese hand-drawn technical-explainer style. It is explicitly not a slide template or an editable PPTX generator; its output is meant to be dropped directly into an article, course deck or presentation as finished artwork.

Its workflow first reads the source material and works out a narrative structure, then maps each page to a suitable layout archetype such as a cover metaphor, a side-by-side comparison, a process, a cycle or a summary, and locks a consistent visual style covering the paper background, title treatment, line work, color accents and figure scale before generating images. Each page gets its own written image prompt, and a multi-page set is finished with a contact sheet for quickly checking every page together, plus a short slide-by-slide blueprint summary.

The skill defaults to producing a 21:9 cover image and 16:9 body pages, recommends the ChatGPT Image 2.0 model when available, and installs by copying its files into a Codex skills directory. It is released under the MIT license.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.