aer-replication — Claude Code skill: what it does, how to install
aer-replication builds a replication package for AEA journals under the February 2026 Data and Code Availability Policy. Checks the README and folder layout before the AEA Data Editor review.
no license ★ 240 updated 2026-09-18
npx skills add franklee16/academic-research-skills --skill aer-replication -a claude-codeWhat aer-replication does
The skill targets AER, AER:Insights and AEJ submissions: it covers the data deposit, the code deposit, and a README that lets an outside researcher rerun every table and figure in the paper. It lays out a folder structure (data/raw, data/intermediate, code, output, docs) and the README sections the Data Editor's office expects, down to a table that maps each Table/Figure to its script and line number.
It also enforces reproducibility habits: a single master script for the whole pipeline, relative paths only, an explicit random seed, and a record of package versions for Stata, R or Python. When data can't be shared because of proprietary terms, access restrictions or IRB rules, the skill asks the author to state in the README a commitment to keep data and code for at least five years and to help with reasonable replication requests.
The final README has to ship as a PDF in the top-level folder, and the skill points authors toward the AEA Data and Code Repository on openICPSR, which gives the Data Editor access to draft deposits before publication.
What it helps with
- lay out a replication-package folder structure that matches AEA requirements
- draft a README with a Data Availability Statement in the AEA format
- build the table mapping each Table/Figure to its script and line number
- write up Computational Requirements with exact Stata, R or Python package versions
- prepare an openICPSR deposit ahead of the Data Editor review
Example prompts for the agent
- Set up the replication package structure for a paper accepted at AER
- Check this README against the AEA Data and Code Availability Policy
- Draft the List of Tables and Programs table for my replication package
How to install aer-replication
1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill aer-replication -a claude-code2Manually: put the files in place
~/.claude/skills/aer-replication/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/aer-replication/
1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill aer-replication -a codex2Manually: put the files in place
.agents/skills/aer-replication/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill aer-replication -a cursor2Manually: put the files in place
.cursor/skills/aer-replication/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install franklee16/academic-research-skills Awesome-Journal-Skills-main/AER-Skills-main/skills/aer-replication2Manually: put the files in place
.github/skills/aer-replication/SKILL.md1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill aer-replication -a gemini-cli2Platform CLI
gemini skills install https://github.com/franklee16/academic-research-skills --path Awesome-Journal-Skills-main/AER-Skills-main/skills/aer-replicationManually: put the files in place
.devin/skills/aer-replication/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about aer-replication
How do I install aer-replication in Claude Code?
Run npx skills add franklee16/academic-research-skills --skill aer-replication -a claude-code; the skill lands at ~/.claude/skills/aer-replication/SKILL.md.
What does aer-replication do?
It assembles and audits a data, code and README deposit for AER, AER:Insights and AEJ submissions under the February 2026 AEA Data and Code Availability Policy.
What does the README need to cover under this skill?
It requires an Overview, a Data Availability and Provenance Statement, a Dataset List, Computational Requirements, a Description of Programs, Instructions to Replicators, and a List of Tables and Programs, with the final version delivered as a PDF.