Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/aeri-workflow
Use when deciding which aeri-* sub-skill to invoke next, or when sequencing work on a short-format manuscript from topic selection through the conditional-accept response for an American Economic Review: Insights (AER: Insights) submission. Routes — it does not replace — the specialized skills.
Passed 22 of 22 test cases with the skill, and 4 of 22 without it.
and 17 more in the full report
18 fixed4 passed either way0 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 18 test cases the skill fixed. Same model, same prompt, with and without the skill.
An author is submitting an AER: Insights manuscript that contains 3 main-text exhibits. What is the exact maximum word count limit for the main text under the journal cap rules?
| Checked | Without skill | With skill |
|---|---|---|
| States that the maximum main text word count with 3 exhibits is 6,400 words. | Missed | Passed |
The instructions your model reads when the skill loads.
This is the router. It tells you which aeri- skill to use at the current stage* of a manuscript aimed at American Economic Review: Insights (AER: Insights) — the AEA's short-format, general-interest journal (founded 2019) for papers built around ONE clear, important, well-executed idea, empirical, theoretical, or methodological. The bar is AER-level importance compressed into a short paper. AER: Insights is the short sibling of the American Economic Review (long, multi-result, flagship) and is general-interest, unlike the field-oriented full papers of the AEJ family.
Default assumption: unless the user says otherwise, treat the target as AER: Insights. The two operational tells that you are at AER: Insights and not AER or an AEJ: (1) a hard length cap — main text ≤7,000 words with no exhibits, minus 200 words per exhibit, maximum five exhibits, abstract ≤100 words — papers over the cap are returned unreviewed; (2) a single insight must carry the whole paper, with everything secondary pushed to the online Supplemental Appendix. Process tells: AEA single-blind review that is fast and decisive — no traditional revise-and-resubmit; first decisions are conditional accept or reject, and conditional accepts (usually an 8-week window) are normally not sent back to referees. Editor as of 2026: Matthew Gentzkow (lead editor beginning January 2024; founding editor Amy Finkelstein). Live-check volatile specifics on the official AEA pages.
| Current symptom | Next skill |
|---|---|
| Paper has several results / no single crisp insight; fit to short format unclear | aeri-topic-selection |
| Contribution vs. the closest published work is fuzzy or oversold | aeri-literature-positioning |
| The causal or parameter identification is shaky | aeri-identification |
| A model is present but bloated, or unclear what it must minimally show | aeri-theory-model |
| Robustness is sprawling; unclear what stays in-text vs. appendix | aeri-robustness |
| Over the five-exhibit budget; exhibits are dense or carry significance asterisks | aeri-tables-figures |
| Prose buries the result; abstract/intro do not lead with it; over word cap | aeri-writing-style |
| Data/code deposit, README, or AEA Data Editor prep | aeri-replication-package |
| Want to understand the fast review / conditional-accept-vs-reject odds | aeri-referee-strategy |
| Ready to submit via the AEA portal; need a preflight | aeri-submission |
| Received a conditional accept; need a short, decisive response | aeri-rebuttal |
aeri-topic-selection — confirm ONE crisp, important, well-identified insight that stands aloneaeri-literature-positioning — stake the contribution in one or two paragraphsaeri-identification — make the design or parameter identification clean and self-containedaeri-theory-model — keep only the model the single insight requiresaeri-robustness — the few in-text checks that matter; rest to the appendixaeri-tables-figures — at most five exhibits, no significance asterisksaeri-writing-style — lead with the result; enforce the word budget (abstract + intro last)test cases passed, vs 4 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.