scientific-slides

Generate scientific presentation slides as images and combine them into PDF decks.

Updated Jun 18, 2026
One-click install
npx skills add https://github.com/svpfahad/RES200 --skill scientific-slides-svpfahad
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: scientific-slides
Source: https://github.com/svpfahad/RES200/tree/main/claude-scientific-writer-main/.claude/skills/scientific-slides
Command: npx skills add https://github.com/svpfahad/RES200 --skill scientific-slides-svpfahad

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, pymupdf, python-dotenv, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Creating visually engaging, research-backed scientific presentations is time-consuming, and default slide templates produce dry, text-heavy decks. This Skill automates slide image generation with AI and assembles them into polished PDF presentations for conferences, seminars, and thesis defenses. ## Core Features & Use Cases - AI Slide Generation: Generate complete slide images via Nano Banana Pro with iterative quality review by Gemini 3 Pro, supporting reference image attachments for style consistency and data figure integration. - PDF Assembly and Review: Combine slide images into a single PDF and convert PDFs back to images for systematic visual inspection of overflow, contrast, and readability issues. - Structured Guidance: Reference guides cover narrative structure, slide design principles, data visualization for slides, talk-type-specific formats, and timing rules. - Use Case: Preparing a 15-minute conference talk, you plan each slide, generate styled slide images with citations and attached result figures, combine them into a PDF, then visually validate every slide before presenting. ## Quick Start Generate a 15-minute conference presentation about my machine learning results as a PDF, using the figures in my results folder and a consistent dark blue visual style.

Frequently Asked Questions about scientific-slides

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate presentation slides with AI?

Run the generate_slide_image.py script with a natural language description of the slide, including title, key points, and a formatting goal for colors and style. The script calls Nano Banana Pro to create the slide image and Gemini 3 Pro to review quality, iterating up to two times.

How do I keep slide formatting consistent across a deck?

Define a formatting goal with colors, fonts, and layout style, and include it in every generation prompt. Attach the previous slide with the --attach flag so the model matches the existing visual style throughout the deck.

Can I include my own data figures in generated slides?

Yes, attach figure files such as charts or diagrams using the --attach flag, which accepts multiple images. Describe in the prompt how the figure should be incorporated, for example building a results slide around an attached accuracy chart.

What API key is required for AI slide generation?

An OpenRouter API key is required, set via the OPENROUTER_API_KEY environment variable or the --api-key flag. Keys are available at openrouter.ai, and the scripts exit with an error if no key is found.

How do I check slides for text overflow or readability problems?

Convert the finished PDF to images with pdf_to_images.py, then inspect each slide for overflow, element overlap, small fonts, and low contrast. Regenerate problem slides with improved prompts and recombine until no critical issues remain.