agentleFS
stepaniniwisstepaniniwis/claude-skills/skills/audit/a-journal-architecture/SKILL.md

a-journal-architecture

Skill0 starsChanged 7 months agoMITView on GitHub

Enforces the structural DNA of award-winning A-journal papers (JMS, AMJ, ASQ, OS, SMJ, Strategy Science). Not about writing style — about intellectual architecture. Ensures: big puzzle, named mechanism, portable device, contribution hierarchy, reviewer-memorable sentence, boundary conditions. Derived from analysis of JMS award papers. Applies to any paper targeting A-journals. Trigger on: 'is this JMS-ready', 'A-journal check', 'award paper structure', 'paper architecture', 'contribution check', 'reviewer-proof', 'is this publishable', 'Zieljournal-Check', or when a paper draft is being prepared for A-journal submission.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 8 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run these sequentially on any paper targeting an A-journal.

    line 16
  2. Distinctive:

    Diagnostic question: If a researcher in an adjacent field read only the first paragraph, would they recognize the problem as important to …

    line 25
  3. Distinctive:

    Pass: After reading, a reviewer can say "This paper introduces [device] which explains [mechanism]."

    line 42
  4. Distinctive:

    Test: Read the Introduction. Can you find the one sentence?

    line 58
  5. Distinctive:

    Diagnostic question: If a reviewer forced you to keep only ONE contribution, which would it be?

    line 67
  6. Distinctive:

    After running the seven tests, verify:

    line 92
  7. Distinctive:

    All three should be named, defined, and linked.

    line 119
  8. Distinctive:

    If the paper fails multiple tests, prioritize fixes in this order:

    line 125
  9. Very specific:

    Before this skill: Use paper-revision-diagnostic for pre-revision assessment of framing, data, cases

    line 149
  10. Very specific:

    After this skill: Use draft-review-loop for execution quality checks during writing

    line 150
  11. Very specific:

    Parallel: Use argument-architecture (CARS + Toulmin) for micro-level claim structure

    line 151
  12. Very specific:

    Parallel: Use paper-arc-architect for narrative tension per section

    line 152

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

A-Journal Paper Architecture

What award-winning A-journal papers have in common — not style, but intellectual structure. Use this skill to diagnose whether a paper has the architecture that survives review at JMS, AMJ, ASQ, OS, SMJ, or Strategy Science.

Origin

Derived from analysis of six JMS award/exemplar papers: Building Character (longitudinal hybrid identity), Causal Triangulation (causal inference methodology), Serendipity (systematic review + conceptualization), Stakeholder Governance (responsible innovation), How Are Skills Changing with Digital Technologies (systematic review), The Acceleration of AI (conceptual). Pattern holds across empirical, conceptual, and review papers.

The Seven Tests

Run these sequentially on any paper targeting an A-journal. Each test produces a verdict: PASS, WEAK, or FAIL.

Test 1: Big Puzzle, Not Small Gap

The paper must open with a problem that many researchers recognize as consequential — not a narrow niche, not "nobody has studied X in context Y."

Pass: The opening problem touches multiple research conversations and makes readers think "yes, that matters." Fail: The opening reads as gap-spotting ("surprisingly little research has examined...").

Diagnostic question: If a researcher in an adjacent field read only the first paragraph, would they recognize the problem as important to THEIR work?

Award papers open with: hybrid identity under organizational pressure, weak causal claims across management research, conceptual confusion around serendipity, governance failures in responsible innovation, skill transformation under digitalization, theoretical drift caused by treating AI as monolith.

Test 2: Intellectual Deficiency, Not Missing Context

The gap must be framed as a deficiency in how the field THINKS, not as a missing empirical context.

Pass: "Existing accounts assume X, but this assumption obscures Y" or "The field conflates A and B, which prevents understanding of C." Fail: "No study has examined this phenomenon in [country/industry/population]."

Diagnostic question: Does the gap statement identify a flaw in theory, concepts, or analytical approach — or just a missing data point?

Test 3: Named Theoretical Device

The paper must produce a transportable intellectual artifact: a named mechanism, a model, a typology, a set of necessary conditions, a process framework. Something a reviewer can paraphrase in one sentence and that other researchers can apply elsewhere.

Pass: After reading, a reviewer can say "This paper introduces [device] which explains [mechanism]." Fail: The paper produces findings but no portable conceptual contribution.

Examples from award papers: re-keying (mechanism), causal triangulation (model), three necessary conditions of serendipity (boundary definition), three-stage stakeholder governance model, four skill framings, four AI types.

Diagnostic question: What is the one thing a doctoral student would cite this paper FOR five years from now? Can you name it in three words?

Test 4: Paper-in-One-Sentence

There must be a single sentence — ideally in the first third of the Introduction — that a reviewer would paraphrase in their review. This sentence captures the entire argument: puzzle + mechanism + outcome.

Pass: The sentence exists and is memorable. Fail: The argument is spread across paragraphs but never crystallized.

Template: "This paper explains why [puzzle] by showing that [mechanism], which produces [outcome]."

Test: Read the Introduction. Can you find the one sentence? If not, write it. If you cannot write it, the paper does not yet have a clear enough argument.

Test 5: Contribution Hierarchy

Multiple contributions must be ranked, not listed as equals. The strongest contribution leads. Weaker contributions (especially methodological) are subordinated or dropped.

Pass: The primary contribution is unmistakably identified and given the most space. Secondary contributions are flagged as such. Fail: Three contributions listed on equal footing, giving the impression of trying to win on too many fronts.

Diagnostic question: If a reviewer forced you to keep only ONE contribution, which would it be? That one leads. The others serve it.

Test 6: Concept Boundary Defense

The central concept must be defended against adjacent concepts. Reviewers will always ask: "Is this not just [existing concept] under a new label?"

Pass: The paper contains explicit contrast sentences: "X is not Y because..." for at least three neighboring concepts. Fail: The new concept is defined but never differentiated from what already exists.

Template for boundary defense: "[New concept] is not simply [adjacent concept A], because [specific distinction]. It is not equivalent to [adjacent concept B], because [specific distinction]. And it is not reducible to [adjacent concept C], because [specific distinction]."

These three sentences must appear immediately after the first definition of the new concept.

Test 7: Portability Beyond Setting

The paper must show that its theoretical device travels beyond the specific empirical setting. The setting is the revealing case, not the scope limit.

Pass: The Discussion contains a concrete paragraph showing how the mechanism applies to at least two other contexts. Fail: The paper reads as a case study of one setting with a theoretical label attached.

Diagnostic formula: "[Setting] is the revealing case, not the terminal scope condition. The mechanism should travel to any [general class of fields/organizations] in which [abstract version of the conditions]."

Structural Checklist (Post-Test)

After running the seven tests, verify:

ElementLocationStatus
Big puzzlePara 1-2 of Introduction
Intellectual deficiencyPara 3-4 of Introduction
NO FINDINGS in IntroductionIntroduction must not contain numbers, breakdowns, D-scores, or interview quotes. It opens the puzzle, not the results.
Paper-in-one-sentenceEnd of first third of Introduction
Named mechanismTheory Section, recapped in Abstract
Named structural conceptTheory Section
Named outcomeTheory Section + Findings
Mechanism-Structure-Outcome chainAbstract + Introduction + Theory
Concept boundary defenseImmediately after first definition
Contribution hierarchyIntroduction, contributions section
Portability paragraphDiscussion, first half
Boundary conditionsDiscussion, second half
Call-fit languageIntroduction + Discussion (1-2 sentences)
Research agendaConclusion or end of Discussion

Mechanism-Structure-Outcome Naming

The strongest papers have three named elements in a chain:

  • Mechanism (the process): what happens and why. E.g., convergent neglect, re-keying, causal triangulation
  • Structural location (where it happens): the institutional/organizational space. E.g., governance interstitium, hybrid identity space, knowledge boundary
  • Outcome (what results): the observable consequence. E.g., selective adaptation, identity coherence, robust inference

All three should be named, defined, and linked. The Abstract should contain all three. The Theory Section should derive them. The Findings should demonstrate them.

Diagnostic question: Can you fill in: "[Mechanism] produces [Outcome] because [Structural Location] exists"? If not, the chain is incomplete.

Revision Priorities (When Diagnosing an Existing Draft)

If the paper fails multiple tests, prioritize fixes in this order:

  1. Paper-in-one-sentence (Test 4) — without this, nothing else holds
  2. Named theoretical device (Test 3) — without this, no contribution
  3. Concept boundary defense (Test 6) — without this, reviewers reject as relabeling
  4. Contribution hierarchy (Test 5) — without this, the paper feels scattered
  5. Big puzzle (Test 1) — can usually be fixed by reframing the opening
  6. Portability (Test 7) — usually needs only one Discussion paragraph
  7. Intellectual deficiency (Test 2) — usually fixable by rewriting the gap statement

Anti-Patterns

Anti-patternWhat it looks likeFix
Gap-spotting"No study has examined X in Y"Reframe as intellectual deficiency
Contribution inflationThree equally weighted contributionsHierarchize: one leads, others serve
Setting trapPaper reads as defense/healthcare/fintech case studyAdd portability paragraph + abstract version of conditions
Label-without-mechanismNew term introduced but no process explainedDefine mechanism + structural location + outcome
Defensive breadthPaper covers everything to preempt all criticismCut: better to be deep on one front than shallow on three
Missing contrast sentencesNew concept defined but not differentiatedAdd three "X is not Y because..." sentences
Buried insightThe best sentence is on page 14 of DiscussionMove it to Introduction, first third

Integration with Other Skills

  • Before this skill: Use paper-revision-diagnostic for pre-revision assessment of framing, data, cases
  • After this skill: Use draft-review-loop for execution quality checks during writing
  • Parallel: Use argument-architecture (CARS + Toulmin) for micro-level claim structure
  • Parallel: Use paper-arc-architect for narrative tension per section

This skill operates at the level of intellectual architecture — the skeleton that holds everything. The other skills handle positioning (CARS), argumentation (Toulmin), prose (academic-writing), and quality assurance (draft-review-loop).

Sprache folgt dem Input (Deutsch / English).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.