Use when the question is what makes the result tight or what the data identify for an American Economic Journal: Microeconomics (AEJ: Micro) manuscript — covering both (a) structural/empirical-IO and experimental identification and (b) for pure theory, which assumptions drive the result and how robust the mechanism is. Stress-tests credibility; it does not build the model (see aejmic-theory-model).
SkillScore82Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 18 of 22 test cases with the skill, and 15 of 22 without it.
Without skill68%
With skill82%
Checks it passed that the model alone missed
The output uses the exact bracketed tag 【Branch】.
The output uses the exact bracketed tag 【Next step】.
The Next step field explicitly references aejmic-robustness.
and 6 more in the full report
5 fixed13 passed either way2 still failing2 broke
Pass-rate lift
+14 pp
Tokens
+24%
Agent turns
0%
Response time
-33%
Model gemini-3.6-flashMeasured 8/15/2026v118 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 5
Prompt
I am revising a theoretical microeconomic paper for AEJ Micro. A referee suspects our main existence theorem relies on a knife-edge condition and wants us to pinpoint which assumption is actually driving the result. Could you evaluate our theoretical mechanism and output a structured diagnostic? Please include the relevant branch, a single sentence stating what makes the mechanism tight, the core load-bearing assumptions, tightness evidence (such as counterexamples when an assumption fails), limits on what the proof fails to establish, and the recommended next step for further robustness analysis.
What was checked, without and with the skill
Checked
Without skill
With skill
The output uses the exact bracketed tag 【Branch】.
Missed
Passed
The output uses the exact bracketed tag 【Next step】.
Missed
Passed
The Next step field explicitly references aejmic-robustness.
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.md83 linessingle file
Identification & What Makes the Result Tight (aejmic-identification)
AEJ: Micro is theory-first, so "identification" here is two things. For pure theory it means: which assumptions are doing the work, and how tight/robust the mechanism is. For structural and experimental work it means the standard data-to-object mapping. Pick the branch.
When to trigger
(Theory) A referee asks whether the result is a knife-edge artifact of one assumption
(Theory) You cannot say cleanly which primitive drives the comparative static
(Structural) Parameters are estimated but it is unclear what in the data identifies them
(Experimental) The estimand or the assumptions behind the treatment effect are not pinned down
Branch A: Pure theory — what makes the result tight
The AEJ: Micro bar is that the reader sees exactly which assumption is load-bearing and how far the mechanism extends.
Decompose the assumptions. For each substantive assumption, ask: is the result false without it, weaker without it, or unchanged (then it was WLOG — say so)? The result is "tight" when you can name the assumption that breaks it.
Comparative statics as identification. Show the sign/magnitude of the key comparative static and what primitive drives it (single-crossing? a supermodularity? a curvature condition?). Monotone-comparative-statics tools (Topkis, Milgrom–Shannon) make the driver explicit.
Necessity, not just sufficiency. Where you can, show the assumption is necessary (a counterexample when it fails), not merely sufficient — this is what makes a characterization tight.
Robustness of the mechanism (then hand to aejmic-robustness for full extensions): does the result survive a small perturbation of the information structure, the timing, or the type distribution?
Branch B: Structural / empirical IO
Name what identifies each parameter. Tie parameters to specific data features / moments; argue identification from the model's structure, not "the estimator converged."
Targeted vs. untargeted moments; report a sensitivity/informativeness measure so readers see which data move which parameters.
Estimation regularity: objective (MLE/GMM/MSM), starting values, tolerances, multi-start; Monte Carlo recovery of known parameters.
Counterfactual validity: argue the estimated parameters are policy-invariant enough for the counterfactual (Lucas critique).
For reduced-form companions, use design-appropriate diagnostics (pre-trends, first-stage strength, density tests) and report SEs, not asterisks.
Branch C: Experimental (theory-grounded)
Design maps to the model: each treatment isolates a model primitive or prediction; state the estimand.
Pre-registration in a recognized registry where applicable; report deviations; include instructions/transcripts.
Estimate and audit the identification claim, don't only argue it. Full map:
execution-with-mcp. AEJ: Micro spans applied and structural micro; the chain below is for the reduced-form / causal lane — structural estimation uses the field's own solvers.
detect_design → recommend → fit with as_handle=true → audit_result to list
Use when the empirical identification of a macro shock or dynamic causal effect is the bottleneck for an American Economic Journal: Macroeconomics (AEJ: Macro) manuscript — SVAR, local projections, narrative, high-frequency/proxy-VAR, or micro-data macro designs. Stress-tests the identification to the AEJ: Macro broad-interest quantitative bar; for model-parameter identification see aejmac-theory-model.