beamer-skill  by Noi1r

AI assistant for academic Beamer LaTeX presentations

Created 4 months ago
297 stars

Top 89.2% on SourcePulse

GitHubView on GitHub
Project Summary

A beamer-skill project offers an AI coding assistant for the complete lifecycle of academic Beamer LaTeX presentations. It targets users of AI coding assistants (Claude Code, Codex CLI, VS Code extensions) by automating creation, compilation, review, and polishing, providing features like automated quality scoring, pedagogical review, and TikZ auditing to significantly enhance presentation quality and workflow efficiency.

How It Works

The skill orchestrates a full lifecycle: create for iterative lecture generation with phase gates, compile using a robust 3-pass XeLaTeX + bibtex process with diagnostics, and review/audit for proofreading and visual layout checks. It employs a novel quality scoring system, penalizing issues against an automated rubric. Unique approaches include enforcing "no overlays" via multiple slides and color emphasis, implementing content density guards (e.g., max 7 bullets/slide), and ensuring "motivation before formalism" by requiring a "Why?" before a "What?".

Quick Start & Requirements

  • Prerequisites: A full TeX distribution with XeLaTeX is mandatory (commands for macOS, Ubuntu/Debian, Arch provided). pdf-mcp is recommended for figure extraction.
  • Installation: Clone the repo. Integration varies by AI platform: copy skill directory for Claude Code/Antigravity, specific files (AGENTS.md, .github/copilot-instructions.md, etc.) for Codex CLI and VS Code extensions.
  • Links: Examples are available in the example/ directory.

Highlighted Details

  • Automated quality scoring system with thresholds (Excellent/Good/Needs Work/Poor).
  • Enforces "no overlays" (\pause, \onslide) in favor of multiple slides and color emphasis.
  • Content density guards (e.g., max 7 bullets, 2 equations per slide) and lower bounds.
  • Box overflow detection, TikZ precision enforcement, and semantic, colorblind-safe color system.
  • Figure extraction from paper PDFs using pdf-mcp for direct inclusion.
  • Built-in slide-count heuristics and automatic appendix section for Q&A.

Maintenance & Community

The README does not explicitly mention notable contributors, sponsorships, partnerships, or community links.

Licensing & Compatibility

  • License: MIT License.
  • Compatibility: Permissive MIT license allows for commercial use and integration with closed-source projects.

Limitations & Caveats

Strictly requires XeLaTeX, limiting compatibility with older TeX engines. The benchmark indicates moderate overhead in token usage (~30%) and wall time (~73%) compared to a baseline, attributed to enhanced quality checks.

Health Check
Last Commit

3 weeks ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
37 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.