Skip to content
Agentmart
Submit a tool

latex_mastery

Comprehensive guide to writing publication-grade LaTeX (.tex) documents, equations, TikZ diagrams, bibliographies, and autonomous compilation workflows.

Skill by ParisNeo

IndexedApache-2.0★ 13
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/parisneo-latex-mastery

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Comprehensive guide to writing publication-grade LaTeX (.tex) documents, equations, TikZ diagrams, bibliographies, and autonomous compilation workflows.

Use cases: documents, writing

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/latex-mastery

What's inside

Skill name
latex_mastery
Description
Comprehensive guide to writing publication-grade LaTeX (.tex) documents, equations, TikZ diagrams, bibliographies, and autonomous compilation workflows.
License
Apache-2.0
Instructions
215 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • name: missing

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Professional LaTeX Engineering & Scientific Typesetting

This skill teaches how to formulate clean, modular, publication-quality LaTeX documents (`.tex`), mathematical equations, IEEE/ACM format papers, TikZ architecture diagrams, and automate zero-error compilation with Python.

---

## 🎨 Part 1: High-Level Typographic Standards & Package Arsenal

### Essential Package Arsenal (Standard Preamble)
```latex
\documentclass[11pt,a4paper]{article}

% --- Core Geometry & Fonts ---
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{lmodern}
\usepackage[margin=1in]{geometry}
\usepackage{microtype} % Superior kerning and character protrusion

% --- Mathematics & Science ---
\usepackage{amsmath,amssymb,amsfonts,amsthm}
\usepackage{mathtools}
\usepackage{siunitx} % Proper unit formatting: \SI{100}{\kilo\meter\per\hour}

% --- Tables & Arrays ---
\usepackage{booktabs} % Publication tables: \toprule, \midrule, \bottomrule
\usepackage{tabularx} % Dynamic auto-width columns (X)
\usepackage{multirow}

% --- Figures & Graphics ---
\usepackage{graphicx}
\usepackage{subcaption}
\usepackage{tikz}
\usetikzlibrary{shapes,arrows.meta,positioning,calc}

% --- Code Listings & Algorithms ---
\usepackage{listings}
\usepackage{xcolor}

% --- Hyperlinks & References ---
\usepackage{hyperref}
\hypersetup{
    colorlinks=true,
    linkcolor=blue!70!black,
    citecolor=green!50!black,
    urlcolor=blue!80!black
}
```

### Table Best Practices (Booktabs Rule)
- ❌ **Never** use vertical rules (`|`).
- ❌ **Never** use double horizontal rules.
- ✅ **Always** use `\toprule`, `\midrule`, and `\bottomrule`.
- ✅ Right-align numeric columns; left-align text descriptions.

---

## 💻 Part 2: Complete Publication-Grade Article Template

```latex
\documentclass[11pt,a4paper]{article}

\usepackage[margin=1in]{geometry}
\usepackage{amsmath,amssymb}
\usepackage{booktabs}
\usepackage{tabularx}
\usepackage{tikz}
\usetikzlibrary{positioning,shapes.geometric,arrows.meta}
\usepackage{hyperref}

\title{\textbf{Autonomous Agent Orchestration in Heterogeneous Workspaces: A Two-Tier Architectural Evaluation}}
\author{
    \textbf{ParisNeo} \quad \textbf{Lollms Research Team}\\
    \small Sovereign AI Systems Laboratory\\
    \small \texttt{contact@lollms.com}
}
\date{\today}

\begin{document}

\maketitle

\begin{abstract}
We introduce a decoupled, two-tier execution doctrine separating cognitive planning from atomic tool dispatch in large language model (LLM) agents. By enforcing sandbox confinement, dynamic context budgeting, and formal state verification, our architecture achieves robust verification loops while maintaining zero-amnesia state transitions. Experimental results demonstrate a 34\% reduction in token consumption and complete elimination of phantom tool invocation loops.
\end{abstract}

\section{Introduction}
Modern autonomous agents frequently suffer from context pollution when long-running tool execution outputs overwhelm working memory budgets \cite{vaswani2017attention}. In this paper, we formalize the \textit{Decoupled Execution Protocol} (DEP).

\section{Mathematical Formulation}
Let $\mathcal{S}$ denote the state space of the discussion workspace, and let $\mathcal{A}$ represent the bounded set of executable tool operations. The state transition probability given observation $o_t$ is defined as:

\begin{equation}
P(s_{t+1} \mid s_t, a_t) = \sigma\left( \frac{\mathbf{W}_s s_t + \mathbf{W}_a a_t}{\sqrt{d_k}} \right)
\end{equation}

where $\mathbf{W}_s$ and $\mathbf{W}_a$ denote the state and action projection matrices respectively, and $\sigma(\cdot)$ is the softmax activation function.

\section{System Architecture}
Figure~\ref{fig:architecture} illustrates the high-level orchestration workflow between the Orchestrator tier and the Worker sandboxes.

\begin{figure}[htbp]
\centering
\begin{tikzpicture}[
    node distance=1.8cm,
    block/.style={rectangle, draw=blue!80, fill=blue!10, thick, rounded corners, minimum width=3.5cm, minimum height=1.0cm, align=center},
    worker/.style={rectangle, draw=green!80!black, fill=green!10, thick, rounded corners, minimum width=3.5cm, minimum height=1.0cm, align=center},
    arrow/.style={-Latex, thick}
]
    \node[block] (orch) {\textbf{Orchestrator Tier}\\(Persistent Context)};
    \node[worker, below=of orch] (worker) {\textbf{Worker Sandbox}\\(Disposable Context)};
    \node[block, right=2.5cm of worker] (tools) {\textbf{Tool Engine}\\(Filesystem \& Shell)};

    \draw[arrow] (orch) -- node[right]{\small Task + Files} (worker);
    \draw[arrow] (worker) -- node[above]{\small Tool Call} (tools);
    \draw[arrow] (tools) |- node[below right]{\small Result} (worker);
    \draw[arrow] (worker.west) to[out=180,in=180] node[left]{\small Plain Report} (orch.west);
\end{tikzpicture}
\caption{Decoupled Two-Tier Execution Flow Diagram}

From ParisNeo/lollms_client, Apache-2.0.

More from ParisNeo/lollms_client

Signals

Stars
13
Forks
5
Open issues
0
Last commit (this folder)
Sep 20, 2026 (1 month ago)
Last release
Mar 5, 2025 (1 year ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from ParisNeo

Cite this page

latex_mastery skill. Agent Mart. https://agentmart.dev/skills/parisneo-latex-mastery (last updated 2026-10-24)