research-paper-figure-skill-factory

What problem does it solve?

This skill eliminates the ad-hoc, one-off approach to creating figures for research papers. Instead of manually designing each diagram from scratch, it builds a reusable, evidence-backed specialized skill for a specific figure class so you can consistently produce publication-ready figures across multiple papers.

Core Features & Use Cases

• Evidence-Backed Taxonomy Builder: Automatically extracts figure patterns, captions, and visual evidence from a local corpus of research papers to build a class-specific taxonomy.
• Reusable Skill Generation: Packages the taxonomy, style atlas, prompts, and workflow into a locked, portable skill usable in ChatGPT, Codex, or OpenClaw.
• Two-Round Visual Selection: Enforces a diverse first-round candidate board followed by paper-local best-practice optimization before final prompt construction, ensuring high-quality reviewer-friendly outputs.

Quick Start

Use the research-paper-figure-skill-factory skill to generate a specialized figure-making skill for method framework diagrams by providing your corpus of recent CS top-conference oral papers and confirming the target figure class.