What types of diagrams can this skill create?

It specializes in neural network architectures (Transformers, CNNs), CONSORT/PRISMA flowcharts, biological pathways, system architectures, circuit diagrams, and general scientific block diagrams.

Where are the generated images and logs stored?

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

Does this require manual coding or drawing templates?

No, you simply describe the diagram in natural language. The AI handles the layout, composition, labeling, and professional styling automatically.

How does the quality threshold system work?

The skill uses Gemini 3 Pro to score diagrams against standards specific to your output type. For example, a journal paper requires an 8.5/10 score, while a presentation only requires 6.5/10, triggering iterations only when necessary.

Scientific Schematics AI

Name: Scientific Schematics AI
Author: unerue

byunerue

0•

Data Science & ML

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

Scientific Schematics is a specialized Claude Code skill designed to transform complex technical concepts into professional, publication-ready visual representations. By leveraging Nano Banana Pro AI for generation and Gemini 3 Pro for automated quality audits, it ensures diagrams meet specific academic standards—ranging from high-threshold journal papers to presentation slides. The skill automates the entire workflow of layout, labeling, and styling, allowing researchers and engineers to create sophisticated neural network architectures, biological pathways, and system flowcharts through simple natural language descriptions without the need for manual drawing or complex design templates.

Key Features

01Versioned output with detailed review logs and quality scores

02Document-specific quality standards for journals, conferences, theses, and posters

03Automated generation of neural network architectures, flowcharts, and biological pathways

040 GitHub stars

05Smart iterative refinement using Gemini 3 Pro for quality threshold verification

06Natural language to publication-quality image conversion with automatic labeling

Use Cases

01Creating CONSORT or PRISMA flowcharts for clinical trials and systematic reviews

02Visualizing complex deep learning architectures for machine learning research papers

03Designing biological signaling cascades and molecular interaction maps for life sciences

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add unerue/skills scientific-schematics

For use in Claude.ai and ChatGPT

Download Skill