How to install slide-generation
npx skills add https://github.com/lingzhi227/agent-research-skills --skill slide-generationFull instructions (SKILL.md)
Source of truth, from lingzhi227/agent-research-skills.
name: slide-generation description: Convert a completed paper into presentation slides (Beamer LaTeX) or poster. Extract key figures, tables, equations, and create a narrative flow for oral presentation. Identified gap in existing tools — designed from best practices. argument-hint: [paper-tex]
Slide Generation
Convert a completed paper into presentation slides or poster.
Input
$0— Paper LaTeX file (main.tex) or paper directory
References
- Slide templates and layout patterns:
~/.claude/skills/slide-generation/references/slide-templates.md
Scripts
Extract paper elements for slides
python ~/.claude/skills/slide-generation/scripts/extract_paper_elements.py --tex main.tex --output slides_skeleton.tex
python ~/.claude/skills/slide-generation/scripts/extract_paper_elements.py --tex main.tex --format json --output elements.json
python ~/.claude/skills/slide-generation/scripts/extract_paper_elements.py --tex main.tex --output slides.tex --theme metropolis
Parses .tex, extracts title/authors/sections/equations/figures/tables, generates Beamer skeleton.
Workflow
Step 1: Extract Key Content
From the paper, extract:
- Title, authors, affiliations
- Core contribution (1-3 bullet points from abstract)
- Key figures (all \includegraphics paths)
- Key tables (simplified versions)
- Key equations (numbered equations from Methods)
- Main results (best numbers from Results section)
Step 2: Design Slide Structure
Standard oral presentation flow (~15-20 slides):
| Slide # | Content | Source Section |
|---|---|---|
| 1 | Title slide | Title/Authors |
| 2 | Motivation / Problem | Introduction |
| 3 | Why existing solutions fail | Related Work |
| 4-5 | Our approach (high-level) | Methods |
| 6-8 | Technical details + equations | Methods |
| 9 | Experimental setup | Experiments |
| 10-13 | Results (figures + tables) | Results |
| 14 | Ablation study | Results |
| 15 | Limitations & Future work | Discussion |
| 16 | Conclusion | Conclusion |
| 17 | Thank you + Q&A | — |
Step 3: Generate Beamer LaTeX
\documentclass[aspectratio=169]{beamer}
\usetheme{metropolis}
\title{Paper Title}
\author{Authors}
\date{Venue Year}
\begin{document}
\maketitle
\begin{frame}{Motivation}
\begin{itemize}
\item Problem statement
\item Why it matters
\end{itemize}
\end{frame}
% ... more frames
\end{document}
Step 4: Simplify for Presentation
- Tables: reduce to essential rows/columns
- Equations: show only the key insight, not full derivation
- Figures: use largest versions, add annotations
- Text: bullet points only, no paragraphs
Step 5: Generate Poster Layout (Optional)
For poster sessions, use a multi-column layout:
- Column 1: Introduction + Motivation
- Column 2: Methods + Key Equations
- Column 3: Results + Figures
- Column 4: Conclusions + References
Rules
- Maximum 1 key message per slide
- Figures should be large and readable
- No more than 6 bullet points per slide
- Equations should be simplified versions
- Include slide numbers
- Use consistent color scheme matching the paper's figures
- Presentation should be self-contained (understandable without reading the paper)
Related Skills
- Upstream: paper-compilation, figure-generation
- See also: self-review, paper-assembly
Related skills
More from lingzhi227/agent-research-skills and the wider catalog.

survey-generation
Generate complete academic survey papers using multi-LLM parallel outline generation, RAG-based subsection writing, citation validation, and local coherence enhancement. Based on AutoSurvey pipeline. Use for writing comprehensive literature surveys.

symbolic-equation
Discover scientific equations from data using LLM-guided evolutionary search (LLM-SR). Multi-island algorithm with softmax-based cluster sampling, island reset, and LLM-proposed equation mutations. Use for symbolic regression and equation discovery.

table-generation
Generate publication-quality LaTeX tables from experimental results. Convert JSON/CSV data to booktabs-styled tables with bold best results, multi-row layouts, and proper captions. Use when creating result tables, comparison tables, or ablation tables for papers.

algorithm-design
Design algorithms with LaTeX pseudocode and UML diagrams. Generate algorithmic environments, Mermaid class/sequence diagrams, and ensure consistency between pseudocode and implementation. Use when formalizing methods for a paper.

linkfox-multimodal-recognize-image
基于多模态AI的图片识别与分析。当用户想分析、描述、从图片URL中提取信息、image recognition, image analysis, image description, image content understanding, OCR text recognition, visual Q&A时触发此技能。当用户提到图片识别、图片分析、图片描述、识别图片内容、分析产品图、从图片中读取文字、描述图片、提取视觉内容或理解照片内容时触发。当用户提供图片URL并就其视觉内容提问时,即使未明确说"图片识别",也应触发此技能。

anti-render
智能识别图像内容并生成"理想承诺 vs 残酷现实"风格的视觉对比。触发词:"anti-render"、"理想vs现实"、"对比图"、"渲染vs真实"。适用于任何领域:建筑、人像、产品、食物、旅游、游戏、健身、家居、科技等。