skills/43-wentorai-research-plugins/skills/writing/latex/tikz-diagrams-guide/SKILL.md
Create publication-quality scientific diagrams with TikZ in LaTeX
npx skillsauth add brycewang-stanford/Awesome-Agent-Skills-for-Empirical-Research tikz-diagrams-guideInstall 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.
A skill for creating publication-quality scientific diagrams directly in LaTeX using the TikZ package. Covers basic drawing commands, flowcharts, neural network architectures, data flow diagrams, and integration with PGFplots for camera-ready figures.
\documentclass{article}
\usepackage{tikz}
\usetikzlibrary{arrows.meta, positioning, shapes.geometric, calc, fit}
\begin{document}
\begin{tikzpicture}
% Your drawing commands here
\end{tikzpicture}
\end{document}
% Lines and shapes
\draw (0,0) -- (3,0) -- (3,2) -- cycle; % Triangle
\draw[thick, ->] (0,0) -- (4,0); % Arrow
\draw[dashed, blue] (0,0) circle (1.5); % Dashed circle
\filldraw[fill=gray!20, draw=black] (2,1) ellipse (1 and 0.5);
% Nodes (text labels with optional shapes)
\node[draw, rectangle, minimum width=2cm] (A) at (0,0) {Input};
\node[draw, circle] (B) at (3,0) {Process};
\draw[->] (A) -- (B);
% Relative positioning (requires positioning library)
\node[draw, rectangle] (C) [right=2cm of B] {Output};
\draw[->] (B) -- (C);
\begin{tikzpicture}[
block/.style={rectangle, draw, fill=blue!10, text width=5em,
text centered, rounded corners, minimum height=3em},
decision/.style={diamond, draw, fill=green!10, text width=4em,
text centered, inner sep=0pt, aspect=2},
line/.style={draw, -Stealth}
]
\node[block] (data) {Collect Data};
\node[block, below=1cm of data] (clean) {Clean \& Preprocess};
\node[decision, below=1cm of clean] (valid) {Valid?};
\node[block, below=1cm of valid] (analyze) {Analyze};
\node[block, right=2cm of valid] (fix) {Fix Issues};
\path[line] (data) -- (clean);
\path[line] (clean) -- (valid);
\path[line] (valid) -- node[right] {Yes} (analyze);
\path[line] (valid) -- node[above] {No} (fix);
\path[line] (fix) |- (clean);
\end{tikzpicture}
\begin{tikzpicture}[
neuron/.style={circle, draw, minimum size=0.8cm, fill=orange!20},
layer/.style={rectangle, draw, dashed, inner sep=0.3cm}
]
% Input layer
\foreach \i in {1,2,3,4} {
\node[neuron] (I\i) at (0, -\i*1.2) {};
}
% Hidden layer
\foreach \j in {1,2,3} {
\node[neuron, fill=blue!20] (H\j) at (3, -\j*1.2 - 0.6) {};
}
% Output layer
\foreach \k in {1,2} {
\node[neuron, fill=green!20] (O\k) at (6, -\k*1.2 - 1.2) {};
}
% Connections
\foreach \i in {1,2,3,4} {
\foreach \j in {1,2,3} {
\draw[->] (I\i) -- (H\j);
}
}
\foreach \j in {1,2,3} {
\foreach \k in {1,2} {
\draw[->] (H\j) -- (O\k);
}
}
% Labels
\node[above=0.5cm of I1] {Input};
\node[above=0.5cm of H1] {Hidden};
\node[above=0.5cm of O1] {Output};
\end{tikzpicture}
\usepackage{pgfplots}
\pgfplotsset{compat=1.18}
\begin{tikzpicture}
\begin{axis}[
xlabel={Epoch},
ylabel={Loss},
legend pos=north east,
grid=major,
width=8cm, height=6cm
]
\addplot[blue, thick, mark=none] table {
1 0.95
5 0.72
10 0.45
20 0.22
30 0.15
50 0.08
};
\addlegendentry{Training}
\addplot[red, thick, dashed, mark=none] table {
1 0.98
5 0.75
10 0.52
20 0.35
30 0.30
50 0.28
};
\addlegendentry{Validation}
\end{axis}
\end{tikzpicture}
1. Font consistency:
- Use the same font family as your document body
- Minimum 8pt for axis labels and annotations
- Match font size to caption text
2. Color considerations:
- Use colorblind-friendly palettes (avoid red-green only)
- Ensure figures are readable in grayscale
- Use patterns or line styles as secondary differentiators
3. Size and resolution:
- TikZ produces vector output (PDF) -- always sharp
- Set figure width to match column width (single or double)
- Use consistent sizing across all figures in the paper
4. Labeling:
- Label all axes with units
- Use (a), (b), (c) for sub-figures
- Place legends inside the plot area when possible
% standalone.tex -- compile separately, include as PDF
\documentclass[tikz, border=2mm]{standalone}
\usetikzlibrary{arrows.meta, positioning}
\begin{document}
\begin{tikzpicture}
% ... your diagram ...
\end{tikzpicture}
\end{document}
% In your main document:
% \includegraphics{standalone.pdf}
| Library | Purpose |
|---------|---------|
| positioning | Relative node placement (right=of, below=of) |
| arrows.meta | Modern arrow tip styles |
| shapes.geometric | Diamond, trapezium, ellipse nodes |
| calc | Coordinate calculations |
| fit | Fit a node around a set of other nodes |
| decorations.pathreplacing | Braces, snakes, zigzag lines |
| backgrounds | Draw behind other elements |
| matrix | Grid-based node layouts |
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.