Use when deciding whether a project is the ONE crisp, important, well-identified insight that an American Economic Review: Insights (AER: Insights) short-format manuscript requires, rather than a multi-result paper for AER or a field paper for the AEJ family. Tests single-insight fit; it does not run the analysis.
SkillScore77Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 15 of 20 test cases with the skill, and 5 of 20 without it.
Without skill23%
With skill77%
Checks it passed that the model alone missed
Uses the header tag 【Importance】.
Uses the header tag 【Venue check】.
Uses the header tag 【Prior overturned】.
and 9 more in the full report
12 fixed5 passed either way5 still failing0 broke
Pass-rate lift
+55 pp
Tokens
+21%
Agent turns
0%
Response time
-41%
Model gemini-3.6-flashMeasured 8/15/2026v117 of 22 cases passed with skill · 20 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 11 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 11
Prompt
I have a working paper examining carbon offset markets. The draft has one strong main causal finding and two secondary mechanism extensions, and my coauthors and I are debating whether to target AER: Insights, the main AER, or AEJ: Policy. Can you help us figure out the target venue and how to refine the paper's core thesis? Please structure your output to include:
1. A one-sentence summary of the main insight
2. Broad importance to non-specialist economists
3. The existing prior belief being challenged
4. Whether the core argument fits a short manuscript format
5. A venue recommendation comparing AER: Insights, AER, and AEJ
6. The immediate next step to take.
What was checked, without and with the skill
Checked
Without skill
With skill
Uses the header tag 【One-sentence insight】.
Missed
Passed
Advises moving secondary mechanism extensions to the appendix or a separate paper rather than keeping them in the main text.
Missed
Passed
Specifies aeri-literature-positioning under the header 【Next step】.
Missed
Passed
case-01 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md84 linessingle file
Topic Selection — Single-Insight Fit (aeri-topic-selection)
When to trigger
A project has a strong main result but also two or three secondary results
Unsure whether to send to AER: Insights, AER, or an AEJ field journal
The draft is creeping past the word cap because there is more than one idea
A result feels surprising and clean but you are unsure it is "important enough"
The AER: Insights bar: one idea, AER-level importance, short
AER: Insights publishes short, self-contained papers built around a single important, well-executed idea. The decision is not "is this good?" but "is there exactly ONE crisp, surprising, well-identified insight that can stand alone in a short paper at AER-level importance?" Three filters:
Single-insight test. State the paper in one sentence with one headline object (one estimate, one theorem, one fact, one method-changes-the-answer result). If you need "and" to describe the contribution, you likely have an AER paper or two AER: Insights papers, not one.
Importance test. The single insight must clear AER-level general interest — a non-specialist economist should care. AER: Insights is short, not minor; a small or incremental result is a reject regardless of length.
Self-containment test. The whole argument — setup, identification, headline result, the one or two checks that matter — must fit in ≤7,000 words (minus 200 per exhibit, ≤5 exhibits) with everything else in the online appendix. If the core argument cannot be made short, it is not an AER: Insights paper.
Where it goes if it is NOT AER: Insights
Symptom
Better venue
Multiple inter-dependent results that need each other
AER (long, multi-result flagship)
One result but field-specific, not general interest
AEJ: Applied / Micro / Macro / Policy
Incremental / a robustness note on existing work
a field journal or a comment
Pure methods with no substantive payoff in a short paper
a methods/econometrics journal
Sharpening a candidate into a single insight
Pick the headline. Of the results you have, which ONE is most surprising and most cleanly identified? Make that the paper; demote the rest.
Cut, do not summarize. Secondary results become appendix material or a future paper — not a compressed in-text paragraph.
Name the surprise. AER: Insights rewards a result that overturns or sharply refines a prior belief; write the prior you are correcting.
Confirm identification early (aeri-identification) — a clean insight with a shaky design is not publishable short, because there is no room to defend a weak design.
Common AER: Insights archetypes (and what makes each "one insight")
Archetype
The single insight is…
Failure mode
One clean causal estimate
a single headline effect from a credible design
reporting every heterogeneity split in-text
A striking new fact / measurement
one important fact a non-specialist did not know
a descriptive paper with no single takeaway
A sharp theory result
one surprising theorem / impossibility
a general framework with the result as a corollary