Aer Topic Selection logo

Aer Topic Selection

Community
brycewang-stanford
aer-topic-selection

Use when evaluating whether a research idea clears the AER top-5 bar, when routing between AER, AER:Insights, and the AEJ family, or when sharpening a fuzzy contribution sentence into one publishable claim. Apply before any writing begins.

Overview

Publisherbrycewang-stanford
RepositoryAER-Skills
Skill nameaer-topic-selection
Stars
62
Forks
11
Bundled files
2
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by brycewang-stanford on GitHub. Read the source before you install it.

Installation

Install the Aer Topic Selection AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/brycewang-stanford/AER-Skills/tree/main/skills/aer-topic-selection
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/brycewang-stanford/AER-Skills.git /tmp/AER-Skills
mkdir -p .claude/skills
cp -r /tmp/AER-Skills/skills/aer-topic-selection .claude/skills/aer-topic-selection
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Aer Topic Selection in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Aer Topic Selection on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Aer Topic Selection is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

AER Topic Selection

Overview

The single most expensive mistake in top-5 economics is writing a polished manuscript around a contribution that was always going to be desk-rejected. This skill is the pre-mortem: stress-test the idea, the audience, and the venue before the introduction is drafted.

AER accepts only a small share of submissions — on the order of 6–8% historically (Card and DellaVigna 2013, Nine Facts about Top Journals), and recent Reports of the Editor put the rate lower still. A large fraction of submissions are desk-rejected before ever reaching a referee; for AER: Insights, founding editor Amy Finkelstein has reported desk-rejecting "roughly 45%" of submissions. Either way, much of a paper's survival probability is determined before the first paragraph is written.

When to Use

  • A new project just started and the contribution sentence cannot be written in one line
  • The user is undecided between submitting to AER, AER: Insights, or an AEJ
  • A prior submission was desk-rejected and the user wants to diagnose whether the topic itself was the problem
  • The paper "feels solid" but no senior colleague has agreed to send it to AER

The Top-5 Bar

A paper clears the top-5 bar if and only if all four are true:

  1. Cross-subfield interest. A labor economist's paper must matter to public, macro, IO, and development economists. Not the median person — the editor. If you cannot name three subfields that would cite it, it is an AEJ paper.
  2. Substantive contribution. Methodological extension or substantive new fact or new identification or genuinely new data. Competent application of an existing toolkit is desk-rejected.
  3. Self-contained in the first 3 pages. The editor decides in ten minutes. If understanding the contribution requires the appendix, the paper fails.
  4. Defensible identification or model discipline. Empirical: design-based, not selection-on-observables. Theory: tractable result with an empirical or policy hook.

If any single test fails, the realistic target is an AEJ or a top field journal, not AER.

Venue Routing

SignalTarget
Cross-subfield interest + ≥ 40 pages of substanceAER
One sharp result, fits the AER: Insights word/exhibit formula, can be told without a long lit reviewAER: Insights
Strong but subfield-bounded contribution; methodologically soundAEJ: Applied / Policy / Macro / Micro
Conventional method, modest extension, useful for one literatureTop field journal
Field experiment with policy hookAER, AEJ:Applied, or QJE — register in AEA RCT Registry before submission

For AER: Insights specifically, the editor (Amy Finkelstein) has stated that a great short paper "makes one point and makes it clearly, concisely, and effectively." If you have a second point, write a second paper.

The Contribution Sentence Test

Force the contribution into one sentence with this template:

We show that X causes Y, identified by Z, using data D, with magnitude M,
which changes the way economists think about Q.

If any blank cannot be filled in:

  • X / Y unclear → the research question is not yet a question
  • Z weak → return to aer-identification before doing anything else
  • D thin → small-sample, unreplicable, or off-the-shelf data is desk-reject bait
  • M unknown → results don't exist yet; this is not a topic-selection problem
  • Q absent → this is an AEJ paper, not an AER paper

Novelty Audit

Run these five questions (the search protocol and antecedents map live in aer-literature — use it to answer them with evidence, not recall):

  1. Has someone in NBER, IZA, CEPR, or SSRN already done this? (Search before writing.)
  2. Is the closest published paper in a top-5 within the last 5 years? If yes, what does yours add — new method, new setting, new mechanism, opposite sign?
  3. Does the result change a policy debate, flip a stylized fact, or unify two strands of literature?
  4. Could a senior referee compress your contribution into "well, we already knew that"?
  5. Could a senior referee compress your contribution into "interesting but it's not economics"?

A "yes" to (4) or (5) is fatal at AER.

Desk Rejection Red Flags

The editor scans for these in the first ten minutes:

  • Correlational language masquerading as causal ("our results suggest a relationship between ...")
  • An introduction that runs more than 3 pages without naming the data or the identification strategy
  • A 14-column table 1 with no headline takeaway
  • Spelling, formatting, or LaTeX-bibliography errors visible on the first page
  • Abstract over 100 words
  • Excessive reliance on the online appendix ("see appendix B for the main result")
  • Methodology mismatch — claims causal identification but reports OLS with controls

If any of these are present, fix them before submission.

Pre-Submission Seminar Sequence

Peer feedback measurably improves journal placement: one SD more comments → 47% higher journal quality. Before sending to AER, target:

  1. Internal brown-bag — 2 rounds, separated by ≥ 6 weeks
  2. ≥ 2 external seminars at peer or top-tier departments
  3. ≥ 1 NBER / CEPR / IZA conference presentation
  4. Working paper posted ≥ 3 months before submission

Skip none of these for an AER target. AER: Insights tolerates a shorter cycle because the contribution is sharper and the venue is younger.

Repository Resources

Bundled with the installed skill, no repository checkout needed --- read it before the repo resources below:

  • references/venue-router.md --- AER vs Insights vs AEJ routing tables, novelty audit, and kill criteria

When working from the AER-skills repository or plugin bundle, load only the relevant resource:

  • Recent AER/AEJ exemplars by subfield: examples/modern-aer-exemplars.md
  • Classic contribution archetypes: examples/aer-exemplars.md
  • Desk-rejection no-go audit: docs/desk-rejection-audit.md
  • Workflow routing after the topic decision: docs/workflow-map.md

Go / No-Go Gate

Advance to aer-identification / aer-introduction only if all hold; otherwise route down (AEJ / field journal) or stop:

  • Contribution sentence written in one line with every blank (X, Y, Z, D, M, Q) filled
  • All four Top-5 Bar tests evaluated; AER chosen as target only if all four pass
  • Target venue named explicitly, with a one-sentence reason it is not one tier lower
  • No kill switch live — no "and we also explore", no "we already knew that", no "interesting but not economics"
  • At least three subfields that would plausibly cite the paper can be named

Handoff

When this skill finishes, emit:

text
CONTRIBUTION SENTENCE: <one line>
TARGET VENUE: <AER | AER:Insights | AEJ:Applied | AEJ:Policy | AEJ:Macro | AEJ:Micro>
TOP-5 BAR TESTS PASSED: <count>/4
KILL SWITCHES TRIGGERED: <list, or "none">
NEXT SKILL: <aer-literature | aer-identification>

Anti-Patterns

  • Choosing the venue based on prestige rather than fit. An AEJ acceptance beats two AER rejections.
  • Letting the contribution sentence include "and we also explore ..." — that and is where the paper dies.
  • Submitting to AER because "the bar is the same as AEJ anyway." It is not. Cross-subfield interest is the explicit filter.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Aer Topic Selection AI skill do?

Use when evaluating whether a research idea clears the AER top-5 bar, when routing between AER, AER:Insights, and the AEJ family, or when sharpening a fuzzy contribution sentence into one publishable claim. Apply before any writing begins.

Why use Aer Topic Selection on TypingMind?

Because you install it once and use it with any model. Aer Topic Selection is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Aer Topic Selection in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/brycewang-stanford/AER-Skills/tree/main/skills/aer-topic-selection. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Aer Topic Selection?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Aer Topic Selection?

As many as you like. As long as a model supports skills, you can use Aer Topic Selection with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Aer Topic Selection AI skill free?

Yes. It is published on GitHub by brycewang-stanford under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇