Use when building or trimming the exhibits of an American Economic Review: Insights (AER: Insights) short-format manuscript to the hard five-exhibit budget, one page each, no significance asterisks. Designs the minimum exhibit set; it does not run the analysis behind them.
SkillScore96Maintained
Automatic
◆Writing & Editing pack
Last verified 8/15/2026
What it changes
Passed 22 of 23 test cases with the skill, and 16 of 23 without it.
Without skill70%
With skill96%
Checks it passed that the model alone missed
States the main-text word limit is 7,000 words.
Calculates the main-text word allowance as 6,400 words.
Confirms that JFE allows significance asterisks in table footnotes.
and 5 more in the full report
7 fixed15 passed either way0 still failing1 broke
Pass-rate lift
+26 pp
Tokens
+25%
Agent turns
0%
Response time
-37%
Model gemini-3.6-flashMeasured 8/15/2026v122 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 7
Prompt
If an author submits an AER: Insights manuscript containing zero figures and zero tables, what is the maximum main-text word limit?
What was checked, without and with the skill
Checked
Without skill
With skill
States the main-text word limit is 7,000 words.
Missed
Passed
case-18 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
An exhibit spills past one page, or carries significance asterisks/boldface
Each exhibit costs 200 words of your text budget and you must choose
The central result is buried in a dense multi-panel table
The AER: Insights exhibit constraints (检索于 2026-06；以官网为准)
Maximum five exhibits (figures or tables combined). Manuscripts over the limit are returned unreviewed.
Each exhibit is limited to one page.
Each exhibit subtracts 200 words from the main-text allowance (7,000 with zero exhibits → 6,000 with five). So every exhibit must out-earn 200 words of prose.
AEA house style: report standard errors / confidence sets; no asterisks or boldface to denote significance (an AEA-wide convention). Notes state the estimator, sample, clustering, and units.
Budgeting the five slots
Choose the exhibit set that, together, tells the single-insight story. A common allocation for an empirical short paper:
Slot
Typical content
1
The headline result — the one figure or table a reader remembers (event-study, RD plot, treatment-effect estimate)
2
The central identification diagnostic (pre-trend leads, density, first-stage, balance)
3
A mechanism or key heterogeneity exhibit, if it is part of the single insight
4–5
Held in reserve — a robustness summary or a descriptive table only if load-bearing
If two exhibits can merge into one well-designed multi-panel page, merge them and reclaim a slot (and 200 words). Everything else goes to the Supplemental Appendix.
Design rules
The headline exhibit must be self-reading — a reader who sees only it should grasp the insight.
Figures over tables where a figure communicates the magnitude faster (AER: Insights papers lean visual).
One page each, legible at print size; no microscopic multi-panel grids spilling over.
Notes carry the inference since there are no asterisks: report point estimates with SEs / 95% confidence sets in the cell or note.
Units and signs explicit; the magnitude, not just the sign, is the point.
Execution bridge (StatsPAI / Stata MCP)
Generate exhibits from the fitted result, not by retyping numbers (the usual source of
Use when building or revising exhibits for an AEJ: Economic Policy manuscript so they meet AEA house style and carry the policy message — no significance asterisks, a self-contained headline exhibit, and figures that show the policy effect with uncertainty. Designs exhibits; it does not run the estimation or write the surrounding prose.