aeja-referee-strategy — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited aeja-referee-strategy (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
AEJ: Applied uses single-blind review through the AEA ScholarOne system: author identities are visible to referees, while referee identities remain anonymous to authors. An editor assigns the paper to a coeditor or handles it directly, and before a first R&R the handling coeditor consults a second coeditor. Referees here are identification-literate applied microeconomists; the modal report attacks the credibility of the design first, then magnitude/mechanism, then external validity. The editor weighs whether the contribution is novel and the design clears the bar. Plan to pre-empt the predictable objections in the paper itself, leaving the response letter for genuinely new asks.
| Referee objection (by design) | Pre-emption to build in now |
|---|---|
| "TWFE is biased under staggered timing" | heterogeneity-robust estimator + Bacon decomposition already in the paper |
| "Parallel trends is not credible" | clean pre-trend leads + honest-DID sensitivity bound |
| "Instrument is not excludable" | institutional + theoretical exclusion argument + falsification tests |
| "RD jump is manipulation / bandwidth-driven" | density test + bandwidth sensitivity + donut |
| "Effect is just selection / confounding" | Oster δ / coefficient-stability bounds |
| "Local/ITT effect ≠ the policy-relevant effect" | explicit estimand + calibrated external-validity discussion |
| "Inference too narrow (few clusters)" | wild-cluster bootstrap / randomization inference |
| "Mechanism is a black box" | a test that distinguishes the proposed channel from rivals |
| "Not reproducible" | the openICPSR package built and referenced (aeja-replication-package) |
The shared, venue-neutral objection catalog is in ../../resources/README.md (links to the empirical-methods hub) — walk it before drafting.【Front matter】single-blind title/byline and disclosures ready? [Y/N]
【Objection map】per assumption: [objection → pre-emption in paper]
【Biggest weakness】addressed in text? [Y/N] — how: ___
【Robustness placement】main vs appendix split: ___
【External validity】estimand + scope stated? [Y/N]
【Reproducibility visible】package referenced? [Y/N]
【Next step】aeja-submission~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.