Use when the model setup, equilibrium concept, or proof architecture is the bottleneck for an American Economic Journal: Microeconomics (AEJ: Micro) manuscript — the central skill for a theory-first paper. Builds the model to the AEJ: Micro bar (clean, general, well-motivated result); it does not check whether your theorem is correct.
SkillScore73Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 16 of 22 test cases with the skill, and 7 of 22 without it.
Without skill32%
With skill73%
Checks it passed that the model alone missed
Recommends aejmic-robustness under 【Next step】.
Uses the bracketed header 【Model】 for primitives.
Recommends aejmic-identification under 【Next step】.
and 19 more in the full report
9 fixed7 passed either way6 still failing0 broke
Pass-rate lift
+41 pp
Tokens
+15%
Agent turns
0%
Response time
-26%
Model gemini-3.6-flashMeasured 8/15/2026v116 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 9
Prompt
I am working on an economic theory paper about dynamic platform pricing under asymmetric information. I need to formulate my theoretical setup for an AEJ Micro style submission. Please organize my setup into the core primitives (agents, choices, information structure, sequence of play, and payoffs, distinguishing substantive from simplifying assumptions), specify the formal solution concept and existence conditions, articulate the primary proposition in clear prose, detail the mathematical proof strategy, explain the trade-off between generality and tractability, indicate which proofs should appear in the main paper versus the appendix, and recommend the appropriate next step.
What was checked, without and with the skill
Checked
Without skill
With skill
Uses the bracketed header 【Model】 for primitives.
Missed
Passed
Uses the bracketed header 【Solution concept】 for equilibrium choices.
Missed
Passed
Uses the bracketed header 【Main result】 for the primary proposition.
Missed
Passed
Uses the bracketed header 【Proof architecture】 for the strategy.
Missed
Passed
Uses the bracketed header 【Generality call】 for tractability trade-offs.
Missed
Passed
Uses the bracketed header 【Proofs live】 for main text vs appendix location.
Missed
Passed
Uses the bracketed header 【Next step】 for recommending subsequent skills.
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.md73 linessingle file
Theory Model & Proof Architecture (aejmic-theory-model)
This is the central skill for AEJ: Micro. The journal is theory-first: the deliverable is a clean, general, well-motivated result — a theorem, characterization, or impossibility — that a broad microeconomics audience finds illuminating. Get the model and the proof right before anything else.
When to trigger
The primitives (players, actions, information, timing, payoffs) are not yet pinned down
The equilibrium / solution concept is unstated or mismatched to the question
A result is conjectured but the proof strategy is unclear or fragile
You are trading generality against tractability and cannot decide where to cut
A referee questions which assumption the result actually depends on (then pair with aejmic-identification)
The AEJ: Micro modeling bar
AEJ: Micro wants the economics to be the point: the model should be the simplest environment in which the mechanism is visible, and the result should travel beyond the example. It sits between JET/GEB (specialist, often maximal generality) and AEJ: Applied (empirical). Aim for a result that a non-specialist micro theorist can state in one sentence and that a specialist finds non-obvious.
1. Model setup — make primitives minimal and motivated
State players, action sets, information structure, timing, and payoffs explicitly; flag every assumption that is substantive vs. WLOG.
Prefer the minimal environment that exhibits the mechanism; add structure only when it earns its keep in the result.
Motivate each non-standard assumption economically (an institution, a friction), not just for tractability.
2. Equilibrium / solution concept — name it and justify it
Pick the concept that matches the question: Nash / subgame-perfect / sequential / PBE; dominant-strategy or Bayesian incentive compatibility (mechanism design); stability / core (matching, market design); rationalizability or level-k (behavioral).
State existence and, where it matters, uniqueness (or characterize the equilibrium set). Refinements should be justified, not invoked to rescue a result.
3. Proof strategy — choose the architecture before writing it
Common architectures: construction (exhibit the object), revelation principle then optimize over direct mechanisms, fixed-point / monotone-comparative-statics (Topkis, single-crossing), duality / LP (information design, Bayesian persuasion), induction on type/stage, variational / first-order arguments, impossibility by contradiction.
For a characterization, prove both directions; for an optimum, verify the candidate is global (not just a local/first-order point).
4. Generality vs. tractability — decide deliberately
A finite/binary type space that exposes the mechanism cleanly often beats a continuum that buries it. State the cost: what does the simplification rule out?
If the general case is the contribution, prove it generally; if the insight is the contribution, prove it where it is cleanest and discuss extension in aejmic-robustness.
5. Where proofs live
Short, illuminating proofs (or proof sketches) in the main text; long or routine proofs in the appendix. The main paper should be self-contained for the key results — do not hide a load-bearing proof in supplementary material.
Checklist
Primitives stated explicitly; each assumption marked substantive vs. WLOG, with economic motivation
Solution concept named, matched to the question, with existence (and uniqueness/characterization where relevant)