AI guides for academic economists
A short collection of practical guides for working with agentic AI on empirical research — Claude Code, Codex, and the surfaces around them — plus the slides from my workshop on using these tools day to day.
Where to start
If you're not sure which tool or surface fits your workflow, start with the overview. It is a five-minute tour of where these agents can run (VS Code, terminal, desktop, web, mobile) and which surface suits which task.
-
Where to run agentic AI
Six surfaces, one decision — a short tour of VS Code, terminal, desktop apps, web, and mobile, with a cheat sheet for which one to pick for which task.
Hands-on guides
Walk-throughs for getting Claude Code or Codex set up in VS Code for empirical research — installation, recommended extensions, file-format handling, persistent project context, reusable workflows, and managing the context window — and for checking what these tools hand back before it reaches a paper.
-
Claude Code in VS Code
Setting up and using Claude Code inside VS Code for R, Stata, Python, and LaTeX work. Covers
CLAUDE.md, skills, and the everyday research workflow. -
Codex in VS Code
The same ground for OpenAI's Codex — installation,
AGENTS.md, skills, approval modes, and cloud delegation. -
Codex vs Claude Code
A side-by-side look at how the two agents differ in everyday research use — project instructions, slash commands, permissions, and local-versus-cloud workflows.
-
Verifying LLM output
Practical checks for AI-generated code and analysis — reviewing diffs with a fresh agent, second opinions from a different model, comprehension quizzes, and independent reimplementation in another language.
-
Linking VS Code and Overleaf
Sync your LaTeX papers through GitHub or Dropbox so you can edit locally with Claude Code or Codex while coauthors keep working in Overleaf.
Skills for research
A skill is a markdown file that Claude Code loads when you type its / command. Most skills used in the workshop is in one public repository, github.com/claesbackman/AI-research-feedback
(MIT licence). Fork them, edit them, and change them however you like.
Download all skills from my Github repo (zip).
| Command | What it does | When to use it |
|---|---|---|
/review-paper |
Pre-submission referee report. Eight subagents run in parallel: spelling and style, internal consistency, unsupported claims and identification, mathematics, tables and figures, a referee assessment, a contribution advocate, and a contribution skeptic. Add a journal: /review-paper JF. |
A draft you are about to circulate or submit. Ten to fifteen minutes. |
/review-paper-code |
Maps the paper's empirical claims to the code. Checks reproducibility, sample restrictions, clustering, fixed effects. | Before a replication package leaves your hands. |
/explain-diff |
Explains a code change as an offline HTML page: what the code did before, what changed, what it does to the results, and a five-question quiz. | When you need to understand a change well enough to defend it. |
/explorable-explanations |
Builds an interactive explanation of a topic as a tree of short HTML pages the reader moves through, in the style of Nicky Case's The Evolution of Trust. Written by Nityesh Agarwal. You can download it here. | Explaining a mechanism or a model to students or a non-specialist reader. |
/explorable-deck |
Builds an interactive presentation based on the same format as /explorable-explanations. The presentation is done in Quarto-Markdown and is best done by Fable. . | Creating an interactive presentation with a clear narrative. |
Installation
Without a terminal
Windows Explorer or macOS Finder.
- Download the zip below and unzip it.
- Open the
Skillsfolder inside. Copy each skill folder (review-paper,review-paper-light, and so on) into~/.claude/skills/. - That is a hidden folder in your user folder:
C:\Users\you\.claude\skillson Windows,/Users/you/.claude/skillson a Mac. To see hidden folders in Explorer: View → Show → Hidden items. In Finder: Cmd-Shift-. in your home folder. - Restart VS Code. Type
/in Claude Code and the skills appear.
With a terminal
macOS, Linux, or WSL. Paste this one line. Run it again later to update.
git clone --depth 1 https://github.com/claesbackman/AI-research-feedback.git /tmp/airf && mkdir -p ~/.claude/skills && cp -R /tmp/airf/Skills/. ~/.claude/skills/ && rm -rf /tmp/airfA skill in ~/.claude/skills/ works in every project. One in
project/.claude/skills/ works only in that project. The
README has one-line installers
for individual skills.
Related
The resources page collects writing from other economists on AI-assisted research, plus the broader hidden curriculum on writing, presenting, and coding. I also write about AI, coding, research, and housing on Substack.