Create publication-quality scientific diagrams using Nano Banana Pro AI with smart iterative refinement. Uses Gemini 3 Pro for quality review. Only regenerates if quality is below threshold for your document type. Specialized in neural network architectures, system diagrams, flowcharts, biological pathways, and complex scientific visualizations.
Detected risks:
The scientific-schematics skill is designed to create publication-quality scientific diagrams efficiently. It addresses the challenge of visually representing complex scientific concepts in a way that is clear and easily understood by a diverse audience. By utilizing advanced AI technology, this skill automates the process of generating diagrams, ensuring a high level of quality while significantly reducing the time and effort traditionally required for such tasks.
This skill incorporates Nano Banana Pro AI for the automatic generation of diagrams and employs Gemini 3 Pro for a quality review process. Key features include a smart iterative refinement system that only regenerates diagrams when the output quality is below specified thresholds for different document types. This allows users to obtain high-quality visuals quickly, without the need for coding or manual drawing. The skill supports a variety of scientific visualizations, including neural network architectures, system diagrams, biological pathways, and more, making it a versatile tool for researchers and academics alike.
Target users for this skill include researchers, scientists, and students who need to produce high-quality visual content for publications, presentations, posters, and grant proposals. The automated nature of this skill not only enhances productivity but also ensures that users can focus more on their research rather than the intricacies of diagram design. With the ability to refine outputs based on quality assessments, this skill is particularly beneficial in academic settings where visual clarity and adherence to publication standards are crucial.
You can generate a diagram by describing it in natural language and using the provided command in the terminal.
You can create various diagrams including neural network architectures, system diagrams, flowcharts, and biological pathways.
Yes, you can specify a maximum number of iterations for the generation process, with a default limit of 2.
Quality thresholds vary by document type, with journal papers requiring the highest threshold of 8.5/10.
You will need to set your OpenRouter API key and have access to the Nano Banana Pro AI.
Quick Setup:
.claude/skills/Repository
davila7/claude-code-templates