Research
paper-compile avatar

paper-compile

Automated LaTeX compilation, error diagnosis, and PDF verification pipeline for academic paper submissions.

Introduction

The paper-compile skill provides a robust, iterative workflow for transforming LaTeX source files into submission-ready PDFs. It is specifically designed for researchers and academic writers who need an automated way to manage the build process of complex documents, including handling bibliographies, multi-pass compilation, and common typesetting errors. The system utilizes industry-standard tools like latexmk and pdflatex to ensure consistent output while providing built-in diagnostic routines to parse and resolve log-level errors such as missing packages, undefined references, or figure path issues.

  • Automated full-cycle compilation using latexmk to handle dependencies, BibTeX, and multi-pass builds efficiently.

  • Integrated error diagnostic loop that parses compile.log to automatically identify and flag missing citations, figure file mismatches, or LaTeX syntax warnings.

  • Verification layer that performs post-compilation checks, including PDF existence validation, page count auditing based on specific venue constraints, and visual presentation sanity checks.

  • Support for multiple LaTeX engines such as pdflatex, xelatex for CJK character support, and lualatex for advanced font requirements.

  • Configurable page limit monitoring for specific ML conferences like ICLR, NeurIPS, ICML, and IEEE venue-specific standards where references and appendices are handled differently.

  • The skill expects a standardized repository structure with a main.tex file and dedicated directories for sections, figures, and references.

  • Users can trigger the compilation with natural language prompts like compile paper, 编译论文, or build PDF.

  • It acts as a safety-first automation layer; if automated fixes fail after a specified number of attempts, it recommends manual intervention or further expert diagnostics via secondary tools like /codex:rescue.

  • It is highly optimized for CI/CD workflows and local IDE integrations, allowing researchers to ensure their paper remains submittable throughout the writing process without manual build overhead.

Repository Stats

Stars
7,776
Forks
728
Open Issues
52
Language
Python
Default Branch
main
Sync Status
Idle
Last Synced
Apr 29, 2026, 01:55 PM
View on GitHub