A multi-round co-design skill designed to generate publication-ready framework diagrams for research papers. It targets researchers and academics by offering a structured approach to transform paper content (PDFs, abstracts, LaTeX) into visual representations like method overviews, architecture diagrams, and workflows, thereby enhancing scientific communication clarity.

How It Works

This skill operates via a rigorous, multi-stage workflow (S0-P9) executed through LLM interactions, primarily targeting ChatGPT web and Codex environments. It initiates with displaying a startup plan and a built-in reference atlas, followed by material intake (e.g., paper PDFs) and figure-need diagnosis. The process generates textual candidate proposals before moving to visual generation. A key feature is the first round of diverse candidate figures (P5), followed by a mandatory second-round optimization (P6b/P6b-IMAGE/P6c) focused on paper-local details and best practices. Abstract visual decisions are supported by embedded reference/concept images, and visual structure explanations are image-based, not purely textual. Target paper images are strictly isolated to IMAGE_ONLY steps.

Quick Start & Requirements

Download the paper-framework-figure-studio-pro-v2.5.0-skill.zip skill package. For ChatGPT web, place the zip in Sources, enable "Extended thinking," and manually switch to "Create image" mode when prompted for figure generation. For Codex, place the zip in the project directory and utilize $imagegen or an approved API. The target paper's PDF is required. Official documentation links beyond the README are not provided.

Highlighted Details

• Structured multi-round co-design process (S0-P9) enabling iterative refinement.
• Mandatory second-round optimization (P6b/P6b-IMAGE/P6c) for paper-local details before final prompt construction.
• Strict separation of target paper images (IMAGE_ONLY steps) from explanatory text.
• Emphasis on visual explanations for abstract concepts and structure using embedded reference/concept images.

Maintenance & Community

The provided README does not specify notable contributors, sponsorships, or community channels such as Discord or Slack.

Licensing & Compatibility

The license type and any compatibility notes for commercial use or closed-source linking are not specified in the README.

Limitations & Caveats

The skill's effectiveness is contingent on the underlying LLM's image generation capabilities. Its complex, step-by-step workflow necessitates strict user adherence. The absence of a specified software license is a potential adoption blocker. No performance benchmarks or explicit comparisons against alternative tools are presented.