My Claude Code Setup

Pedro H. C. Sant'Anna · Emory University

Work in progress. This summarizes how I use Claude Code for academic work — lecture slides, R scripts, Beamer-to-Quarto pipelines. I keep learning and updating these files. Sharing what I've figured out so far with friends and colleagues.

A starter kit for academics using Claude Code with LaTeX/Beamer, R, and Quarto. You describe what you want; Claude plans the approach, runs specialized agents, fixes issues, verifies quality, and presents results — like a contractor who handles the entire job. Extracted from a PhD course I developed (Econ 730 at Emory, 6 lectures, 800+ slides).

Read the full guide

What's in the template

Contractor mode orchestrator — you describe the task; Claude autonomously plans, implements, reviews with agents, fixes issues, and re-verifies until quality gates pass
Beamer → Quarto pipeline — 11-phase translation with TikZ-to-SVG and ggplot-to-plotly conversion
10 specialized agents — proofreader, slide auditor, pedagogy reviewer, R reviewer, TikZ critic, domain reviewer, adversarial QA pair, translator, verifier
Adversarial critic-fixer loop — two agents that check each other's work across up to 5 rounds — the critic can't fix, the fixer can't approve
Quality scoring with mandatory verification — every file scored 0–100; nothing ships below 80; the orchestrator verifies every output before reporting done
19 slash commands + 17 context-aware rules — /compile-latex, /proofread, /deploy, /commit, /qa-quarto, /lit-review, /review-paper, /data-analysis, and more — plus quality gates, notation consistency, and R conventions
Research workflow skills — /lit-review for literature synthesis, /research-ideation for hypothesis generation, /interview-me to formalize ideas, /review-paper for manuscript review, /data-analysis for end-to-end R analysis
Smart hooks — macOS desktop notifications, file protection for bibliography and settings, pre-compaction context snapshots to session logs
Plan-first workflow with continuous learning — non-trivial tasks start in plan mode; [LEARN:tag] corrections persist in MEMORY.md across sessions; plans survive context compression
Research exploration workflow — structured explorations/ sandbox with fast-track prototyping (60/100 threshold), simplified orchestrator for R/research, and merge-only quality reporting
Automated session log enforcement — a lightweight Stop hook reminds Claude to keep session logs current; Beamer-Quarto sync enforced via auto-loaded rules

Getting started

# Fork this repo on GitHub, then:
git clone https://github.com/YOUR_USERNAME/claude-code-my-workflow.git my-project
cd my-project

Then start Claude Code and paste the starter prompt from the guide. Describe your project in 2–3 sentences and Claude will read all the configuration files, fill in the placeholders, and adapt the workflow to your use case.

The full guide walks through everything: setup, agents, quality gates, workflow patterns, and customization.

Read the full guide Copy the starter prompt

Source

Everything is on GitHub: pedrohcgs/claude-code-my-workflow