Do I need to provide templates for the diagrams?

No, you simply describe the diagram in natural language. The AI handles the layout, composition, labels, and styling automatically based on scientific best practices.

What types of diagrams can this skill generate?

It supports a wide range of visuals, including neural network architectures (Transformers, CNNs), biological pathways, methodology flowcharts (CONSORT, PRISMA), circuit diagrams, and system architectures.

How does the quality review system work?

The skill uses Gemini 3 Pro to score diagrams against a threshold based on your document type (e.g., 8.5/10 for journals). If the score is below the threshold, it automatically triggers a refined regeneration.

Where are the generated diagrams stored?

All diagrams are automatically saved in the figures/ subfolder of your project directory, accompanied by a JSON review log containing quality scores and critiques.

Scientific Schematics & Diagrams

Name: Scientific Schematics & Diagrams
Author: x-cmd

byx-cmd

•

데이터 과학 및 ML

Generates publication-quality scientific diagrams and architectural schematics using AI-driven iterative refinement.

Scientific Schematics is a specialized Claude Code skill designed to bridge the gap between complex technical concepts and professional visual representations. By leveraging Nano Banana Pro for generation and Gemini 3 Pro for quality assurance, it creates high-fidelity visuals ranging from neural network architectures to biological pathways. The skill employs a unique smart iteration workflow that automatically evaluates and improves diagrams against specific document standards—such as Nature-grade journals or academic posters—ensuring output meets professional publication thresholds without the need for manual drawing or complex design software.

주요 기능

01Smart iterative refinement that stops once document-specific quality thresholds are met

02Specialized support for neural networks, biological pathways, and system architectures

03Gemini 3 Pro-powered automated quality review and multi-point scoring

04Automated generation of publication-ready scientific visuals via natural language

05Versioned image output with detailed review logs and improvement critiques

069 GitHub stars

사용 사례

01Visualizing complex biological signaling cascades and molecular interactions for publication

02Generating methodology flowcharts like CONSORT or PRISMA for clinical and technical reports

03Creating high-fidelity architecture diagrams for machine learning and AI research papers

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add x-cmd/skill scientific-schematics

For use in Claude.ai and ChatGPT

Download Skill