Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/aer-paper-body
Use when drafting or revising the body sections of an AER, AER:Insights, or AEJ manuscript — institutional background, data, empirical strategy, results, mechanisms, and conclusion. Covers equation conventions, results-paragraph narration, magnitude interpretation, and back-of-envelope policy calculations. Apply after the empirics are stable and before or alongside aer-introduction.
Passed 13 of 25 test cases with the skill, and 13 of 25 without it.
2 fixed18 passed either way4 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
Write the concluding sentence for our Mechanisms section after showing that cash transfers increase savings primarily among unbanked households.
| Checked | Without skill | With skill |
|---|---|---|
| Uses calibrated consistency language such as 'most consistent with' rather than asserting definitive causal proof for the mechanism. | Missed | Passed |
The instructions your model reads when the skill loads.
The introduction decides whether the editor sends the paper out; the body sections decide what the referees write. Referees live in Data, Empirical Strategy, and Results, checking that the estimand is defined, the assumption stated, the magnitudes interpreted, and the prose matched to the tables. Draft the body before the introduction — the introduction summarizes a paper that already exists, and writing it first produces promises the body fails to keep.
aer-identification and aer-robustness)
and the manuscript needs full section draftsA full-length empirical AER paper, after the unlabeled introduction:
I. Background (or: Institutional Setting; Policy Context)
II. Data (sources, sample construction, measurement, summary stats)
III. Empirical Strategy (estimand, equation, identifying assumption, inference)
IV. Results (main estimates, dynamics, robustness pointers)
V. Mechanisms (or: Heterogeneity and Mechanisms; Interpretation)
VI. Conclusion
Variants: a conceptual framework goes between Background and Data (or replaces Background for theory-led papers); AER: Insights compresses to Data and Design → Results → Discussion; structural papers add Model and Estimation, where the rules below bind with more force, not less. One rule binds everywhere: **every term, dataset, and design feature is defined before first
test cases passed, vs 13 / 25 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.