skills/43-wentorai-research-plugins/skills/writing/templates/scientific-article-pdf/SKILL.md
Generate publication-ready scientific article PDFs from templates
npx skillsauth add brycewang-stanford/Awesome-Agent-Skills-for-Empirical-Research scientific-article-pdfInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Producing a publication-ready PDF that meets the exact formatting requirements of a target journal or conference is one of the final and most frustrating steps of the research publication process. Each venue has its own LaTeX class file, formatting rules, margin specifications, font requirements, and reference style. Getting these details wrong can lead to desk rejection before your paper even reaches a reviewer.
This skill provides a systematic approach to generating correctly formatted scientific article PDFs. It covers obtaining and using official LaTeX templates, configuring common elements (title blocks, author affiliations, abstracts, section numbering, bibliography), and troubleshooting formatting issues that arise during compilation. The goal is to produce a PDF that compiles cleanly and passes the publisher's automated format checks on the first try.
The skill focuses on LaTeX-based workflows, as LaTeX remains the standard for most scientific venues, particularly in STEM fields. It includes guidance for major publishers and conference families.
Always start from the official template provided by your target venue. Using a custom or outdated template is a common cause of formatting problems.
IEEE:
# IEEE conference and journal templates
# Download from: https://www.ieee.org/conferences/publishing/templates.html
# Key files: IEEEtran.cls, IEEEtran.bst
wget https://www.ieee.org/content/dam/ieee-org/ieee/web/org/pubs/conference-latex-template_10-17-19.zip
ACM (Association for Computing Machinery):
# ACM Primary Article Template
# Download from: https://www.acm.org/publications/proceedings-template
# Key file: acmart.cls (supports all ACM formats)
# Use: \documentclass[sigconf]{acmart} % for conference
# \documentclass[acmlarge]{acmart} % for journal
Springer (LNCS - Lecture Notes in Computer Science):
# Download from: https://www.springer.com/gp/computer-science/lncs/conference-proceedings-guidelines
# Key file: llncs.cls
# Typical usage: \documentclass[runningheads]{llncs}
Elsevier:
# Elsevier article template
# Download from: https://www.elsevier.com/authors/policies-and-guidelines/latex-instructions
# Key file: elsarticle.cls
# Usage: \documentclass[preprint,12pt]{elsarticle} % for submission
# \documentclass[final,3p,times]{elsarticle} % for camera-ready
NeurIPS / ICML / ICLR:
# NeurIPS: https://neurips.cc/Conferences/2026/PaperInformation/StyleFiles
# ICML: https://icml.cc/Conferences/2026/StyleAuthorInstructions
# ICLR: Uses OpenReview; template available from submission portal
After downloading a template, verify it compiles correctly before writing:
# Compile the sample document
pdflatex sample.tex
bibtex sample # or biber sample
pdflatex sample.tex
pdflatex sample.tex # Two passes to resolve references
# Check for warnings
grep -i "warning" sample.log | head -20
Here is a general-purpose structure that works across most scientific article templates:
\documentclass[options]{template-class}
% ========== Packages ==========
\usepackage[utf8]{inputenc} % UTF-8 encoding
\usepackage{amsmath,amssymb} % Math symbols
\usepackage{graphicx} % Figures
\usepackage{booktabs} % Professional tables
\usepackage{hyperref} % Clickable links
\usepackage{algorithm2e} % Algorithm pseudocode
\usepackage{microtype} % Typographic refinement
% ========== Metadata ==========
\title{Your Paper Title}
\author{%
Author One\thanks{Corresponding author: [email protected]} \\
University of Example \\
\and
Author Two \\
Institute of Research
}
\date{}
\begin{document}
\maketitle
\begin{abstract}
Your abstract text here (typically 150-300 words).
\end{abstract}
\section{Introduction}
\label{sec:intro}
\section{Related Work}
\label{sec:related}
\section{Method}
\label{sec:method}
\section{Experiments}
\label{sec:experiments}
\section{Conclusion}
\label{sec:conclusion}
\section*{Acknowledgments}
This work was supported by...
\bibliographystyle{template-bst}
\bibliography{references}
\appendix
\section{Additional Results}
\label{app:results}
\end{document}
\begin{table}[t]
\centering
\caption{Comparison of methods on the benchmark dataset.}
\label{tab:results}
\begin{tabular}{lccc}
\toprule
Method & Accuracy & F1 Score & Params (M) \\
\midrule
Baseline A & 82.3 & 79.1 & 110 \\
Baseline B & 84.7 & 81.4 & 145 \\
\textbf{Ours} & \textbf{87.6} & \textbf{84.9} & 95 \\
\bottomrule
\end{tabular}
\end{table}
\begin{figure}[t] % [t]=top, [b]=bottom, [h]=here, [p]=page of floats
\centering
\includegraphics[width=0.9\columnwidth]{figures/architecture.pdf}
\caption{Overview of the proposed architecture. The encoder (left)
processes input features while the decoder (right) generates
predictions autoregressively.}
\label{fig:architecture}
\end{figure}
\begin{algorithm}[t]
\caption{Training Procedure}
\label{alg:training}
\KwIn{Dataset $\mathcal{D}$, learning rate $\eta$, epochs $T$}
\KwOut{Trained model parameters $\theta^*$}
Initialize $\theta$ randomly\;
\For{$t = 1$ \KwTo $T$}{
\ForEach{batch $(x, y) \in \mathcal{D}$}{
$\hat{y} \gets f_\theta(x)$\;
$\mathcal{L} \gets \text{Loss}(\hat{y}, y)$\;
$\theta \gets \theta - \eta \nabla_\theta \mathcal{L}$\;
}
}
\Return $\theta$\;
\end{algorithm}
MAIN = paper
LATEX = pdflatex
BIB = bibtex
all: $(MAIN).pdf
$(MAIN).pdf: $(MAIN).tex references.bib
$(LATEX) $(MAIN)
$(BIB) $(MAIN)
$(LATEX) $(MAIN)
$(LATEX) $(MAIN)
clean:
rm -f *.aux *.bbl *.blg *.log *.out *.toc *.fls *.fdb_latexmk
check:
@echo "=== Overfull boxes ==="
@grep -c "Overfull" $(MAIN).log || echo "None"
@echo "=== Undefined references ==="
@grep -c "undefined" $(MAIN).log || echo "None"
@echo "=== Missing citations ==="
@grep -c "Citation.*undefined" $(MAIN).log || echo "None"
.PHONY: all clean check
# latexmk automatically determines how many passes are needed
latexmk -pdf paper.tex
# Continuous mode: recompiles on file changes
latexmk -pdf -pvc paper.tex
# Clean up auxiliary files
latexmk -c paper.tex
| Problem | Solution |
|---------|----------|
| "Overfull hbox" warnings | Rephrase text, use \sloppy locally, or adjust figure widths |
| Missing references | Run BibTeX/Biber and re-run LaTeX twice |
| Figures not found | Use paths relative to the main .tex file; check file extensions |
| Font not found | Install the required font package: tlmgr install <font> |
| Page limit exceeded | Move content to appendix; tighten prose; reduce figure sizes |
| "Too many unprocessed floats" | Add \clearpage or use the placeins package with \FloatBarrier |
tools
Recommend AND run open-source AI tools, agents, Claude Code / Codex skills, and MCP servers for any stage of a literature review — searching, reading, extracting, synthesizing, screening, citation-checking, and paper writing. Use when the user asks "what tool should I use to..." OR "install/run/use <tool> to ..." for research/lit-review work: automating a survey or related-work section, PDF→Markdown extraction for LLMs (MinerU/marker/docling), PRISMA / systematic review (ASReview), citation-backed Q&A over PDFs (PaperQA2), wiring papers into Claude/Cursor via MCP (arxiv/paper-search/zotero servers), or chatting with a Zotero library. Ships a launcher (scripts/litrun.py) that installs each tool in an isolated venv and runs it. Curated catalog of 70+ vetted projects. 支持中英文(用于「文献综述工具选型」与「一键安装/运行」)。
development
Route empirical-research requests through the Auto-Empirical Research Skills catalog when this whole repository is installed as one skill in Codex, CodeBuddy, Claude Code, or another IDE. Use to choose and load the right vendored AERS skill for causal inference, econometrics, replication, data acquisition, manuscript writing, peer review and referee responses, citation checking, de-AIGC editing, or full empirical-paper workflows without reading the entire repository at once.
documentation
Use when the project collects primary data or runs a field, lab, or survey experiment, before the intervention begins — write the pre-analysis plan, size the sample from a power calculation, and register with the AEA RCT Registry. Apply after the design is chosen in aer-identification and before any outcome data are seen.
tools
Guide economists to authoritative data sources with explicit, confirmed data specifications before retrieval; interfaces with Playwright MCP to navigate portals and extract real data, not articles about data.