Paper Architecture Overview Generator

Version: 1.0 Created: 2026-02-07 Purpose: Generate concise strategic architecture overviews for academic papers

Overview

This skill generates a high-level paper architecture overview document (150-300 lines) that distills:

•Overall contribution in one paragraph
•Key data points and metrics
•Three main contributions with emphasis percentages
•Theoretical and policy implications
•5-act storytelling arc
•Strategic framing guidance (what to emphasize/downplay)
•Scope and boundaries (what's in/out)
•Venue positioning

Target Output: Ready-to-use strategic blueprint for writing the paper.

Invocation

Simple Usage:

code

/paper-architecture @paper-structure.md

With Context:

code

/paper-architecture @paper-structure.md @slides.pdf

Note: Downplay causality claims, emphasize measurement innovation.
Target venue: MISQ.

From Scratch:

code

/paper-architecture

[Paste abstract or provide paper description]

Workflow

Step 1: Auto-Detect (Silent)

Detect Paper Type:

•Look for: "design principles", "artifact", "instantiation" → Design Science
•Look for: "theoretical framework", "propositions", "mechanisms" → Theory
•Look for: "regression", "coefficient", "p<0.05", "N=" → Empirical
•If multiple found → Hybrid (rank by text distribution)

Detect Venue:

•Search for: "MISQ", "Management Information Systems Quarterly"
•Check mentions of: "MIS Quarterly", "ISR", "JMIS"
•If not found → Default: MISQ

Detect Contributions:

•Search for: "contribution", "we contribute", "our work advances"
•Extract: List of claimed contributions (typically 2-4)
•Rank by: Frequency of mention, text devoted to each
•Auto-assign emphasis %: Based on paragraph distribution

Detect Strategic Constraints:

•If sees: "suggestive", "exploratory", "preliminary", "not definitive" → Add: acknowledge_incomplete_evidence
•If sees: "IV" + "not significant" OR "2/4 outcomes" OR "NS" → Add: downplay_causality
•If sees: "artifact", "first to", "novel", "unprecedented" → Add: emphasize_innovation
•If sees: "single case", "one context", "limited generalizability" → Add: acknowledge_scope_limitations

Extract Key Data Points:

•Sample sizes: N=, "226,146 physicians", "24.7M prescriptions"
•Costs: "$X per unit", "~$0.10 per physician"
•Validation metrics: "kappa", "correlation", "MAE", "RMSE"
•Effect sizes: "β=", "OR=", "coefficient"
•Significance: "", "", "", "p<0.05"
•Consistency patterns: "8/8 positive", "all outcomes"

Step 2: Generate Complete Overview

Output format:

markdown

Session 1: Paper Architecture Overview (YYYY-MM-DD HH:MM)
==========================================================

Location: Generated from [input files]
Status: ✅ Auto-detected configuration

Auto-Detected Configuration:
┌─────────────────────────────────────────────────────────┐
│ Paper Type: [detected type]                             │
│ Target Venue: [detected venue]                          │
│ Contributions Detected: [N]                             │
│   - C1: [title] (Primary/Secondary/Tertiary, X%)        │
│   - C2: [title] (Primary/Secondary/Tertiary, Y%)        │
│   - C3: [title] (Primary/Secondary/Tertiary, Z%)        │
│ Strategic Constraints:                                  │
│   - [constraint_1] (detected from "[evidence]")         │
│   - [constraint_2] (detected from "[evidence]")         │
│ Key Metrics Extracted:                                  │
│   - [metric 1]                                          │
│   - [metric 2]                                          │
└─────────────────────────────────────────────────────────┘


OVERALL CONTRIBUTION: One-Paragraph Summary
============================================

[3-5 sentences combining: scale + innovation + findings + implications]


KEY DATA POINTS
===============

**Scale:**
- [N samples, N datapoints, timeframe, cost]

**Validation:**
- [Reliability metrics, correlations, benchmarks]

**Core Findings:**
- [Effect sizes with significance, consistency patterns]


THREE MAIN CONTRIBUTIONS
=========================

**C1: [Title] (Primary/Secondary/Tertiary, X%)**
- [Key point 1]
- [Key point 2]
- [Key point 3]
[Max 5 bullets per contribution]

**C2: [Title] (Primary/Secondary/Tertiary, Y%)**
- [Key point 1]
- [Key point 2]

**C3: [Title] (Primary/Secondary/Tertiary, Z%)**
- [Key point 1]
- [Key point 2]


THEORETICAL IMPLICATIONS
=========================

**T1: [Implication Title]**
- [Why this matters for theory]
- [Connection to literature]
[Max 3 bullets per implication, 3-5 implications total]

**T2: [Implication Title]**
- [Why this matters]

**T3: [Implication Title]**
- [Why this matters]


POLICY IMPLICATIONS
===================

**P1: [Policy Title]**
- [Actionable recommendation 1]
- [Actionable recommendation 2]
[Max 3 bullets per implication, 3-5 implications total]

**P2: [Policy Title]**
- [Actionable recommendation]

**P3: [Policy Title]**
- [Actionable recommendation]


STORYTELLING ARC
================

**Act 1: Problem & Gap**
┌─────────────────────────────────────────────────────────┐
│ [Problem statement, 1-2 lines]                          │
│ [Gap in current approaches, 1-2 lines]                  │
└─────────────────────────────────────────────────────────┘

**Act 2: Innovation**
┌─────────────────────────────────────────────────────────┐
│ [Method/artifact innovation, 2-3 lines]                 │
└─────────────────────────────────────────────────────────┘

**Act 3: Discovery**
┌─────────────────────────────────────────────────────────┐
│ [Key findings, 2-3 lines with numbers]                  │
└─────────────────────────────────────────────────────────┘

**Act 4: Mechanism**
┌─────────────────────────────────────────────────────────┐
│ [Why/how findings occur, 2-3 lines]                     │
└─────────────────────────────────────────────────────────┘

**Act 5: Impact**
┌─────────────────────────────────────────────────────────┐
│ [Theory + policy + future research, 2-3 lines]          │
└─────────────────────────────────────────────────────────┘


NARRATIVE STRATEGY
==================

**Primary Story (X% emphasis):**
"[Headline capturing primary contribution]"

**Supporting Evidence (Y% emphasis):**
"[Headline capturing secondary contribution]"

**Language to Use:**
- ✅ "[Example phrase matching strategic constraints]"
- ✅ "[Example phrase 2]"
- ✅ "[Example phrase 3]"

**Language to Avoid:**
- ❌ "[Over-claiming phrase to avoid]"
- ❌ "[Over-claiming phrase 2]"
- ❌ "[Over-claiming phrase 3]"


SCOPE & BOUNDARIES
==================

What We Deliver
---------------
- [Deliverable 1 with concrete metrics]
- [Deliverable 2 with concrete metrics]
- [Deliverable 3 with concrete metrics]
[Max 5 items, each with numbers/evidence]

What Is Out of Scope
---------------------
- [Specific analysis/claim not made]: [why out of scope]
- [Specific population/context]: [why excluded]
- [Specific methodology]: [why not conducted]
[Max 5 items, neutral framing]

Limitations vs Design Choices
------------------------------

**Acknowledged Limitations:**
- [Limitation 1]: [what's missing and impact]
- [Limitation 2]: [what's missing]
[Items that weaken claims]

**Intentional Design Choices:**
- [Choice 1]: [why this scope is appropriate]
- [Choice 2]: [justification]
[Items that reflect strategic decisions]

**Rationale:** [1-2 sentences connecting to contribution emphasis %]

Future Work Prioritization
---------------------------

**High Priority** (strengthen current claims):
- [Item 1]: [what it would add to current paper]
- [Item 2]: [what it would address]

**Medium Priority** (extend scope):
- [Item 1]: [what new contexts it would enable]
- [Item 2]: [what new questions it would answer]

**Low Priority** (aspirational):
- [Item 1]: [long-term research agenda]


PAPER POSITIONING FOR [VENUE]
==============================

**Primary Contribution:** [Type] (X%)
- [Why this fits venue priorities]
- [Unique value proposition]

**Secondary Contribution:** [Type] (Y%)
- [Why this fits venue]

**Tertiary Contribution:** [Type] (Z%)
- [Why this fits venue]
- [Caveat if needed]


===================================================================

End of Architecture Overview

> JL: [Space for user refinement comments]

Step 3: User Refinement (Conversational)

Wait for user to add refinement comments using > JL: format.

User can request:

•Change emphasis percentages: > JL: Make artifact 50%, reduce causal to 10%
•Add boundaries: > JL: Add boundary about not testing other Big Five traits
•Adjust language: > JL: Stronger emphasis on "supplementary evidence"
•Modify sections: > JL: Add policy implication about medical education

Respond with >> CC::

•Start with action: DONE, UPDATED, ADDED
•Be specific: state what changed (1-2 sentences)
•Update relevant sections accordingly

Step 4: Finalize

After refinements complete, confirm:

code

>> CC: Architecture overview finalized. Saved to:
       code-logging/0-paper-architecture-overview.md

Output Requirements

Structure:

•✅ All 9 sections present
•✅ Auto-detected config shown at top
•✅ Total length: 150-300 lines
•✅ Each contribution: ≤5 bullets
•✅ Each implication: ≤3 bullets
•✅ Emphasis % sum to 100%

Content:

•✅ Concrete numbers (not "large" or "significant")
•✅ Specific analyses in scope/boundaries (not vague)
•✅ Copy-pasteable language examples (≥3 DO, ≥3 AVOID)
•✅ Story arc matches contribution emphasis
•✅ Neutral tone in "Out of Scope" section

Strategic Value:

•✅ Emphasis % reflect evidence strength
•✅ Language guidance addresses actual over-claiming risks
•✅ Venue positioning explains fit with journal priorities
•✅ Honest about limitations vs design choices

Auto-Detection Rules

Paper Type Classification:

python

# Pseudo-code for detection logic

if contains("design principles", "artifact", "instantiation"):
    if contains("regression", "p<0.05", "coefficient"):
        type = "empirical_with_artifact"  # Hybrid
    else:
        type = "design_science"  # Pure DS

elif contains("theoretical framework", "propositions", "mechanisms"):
    if contains("hypothesis", "regression", "empirical"):
        type = "theory_with_empirical"  # Hybrid
    else:
        type = "pure_theory"

elif contains("regression", "effect size", "N=", "sample"):
    type = "pure_empirical"

elif contains("meta-analysis", "systematic review", "synthesis"):
    type = "review"

else:
    type = "conceptual"  # Default

Contribution Ranking:

python

# Extract contributions
contributions = find_sections_with_keywords(
    "contribution", "we contribute", "advances", "novelty"
)

# Calculate emphasis
for contrib in contributions:
    paragraphs_devoted = count_paragraphs(contrib)
    emphasis_pct = (paragraphs_devoted / total_paragraphs) * 100

    if emphasis_pct >= 35:
        label = "Primary"
    elif emphasis_pct >= 20:
        label = "Secondary"
    else:
        label = "Tertiary"

Strategic Constraints:

Pattern Detected	Constraint Added
"suggestive", "exploratory", "preliminary"	`acknowledge_incomplete_evidence`
"IV" + "not significant" OR "NS"	`downplay_causality`
"artifact", "first to", "novel"	`emphasize_innovation`
"single case", "one context"	`acknowledge_scope_limitations`
"measurement error", "reliability"	`acknowledge_measurement_limitations`

Venue-Specific Adaptations

MISQ:

•Emphasize: Design science artifact OR theory contribution (40%+)
•Secondary: Empirical evidence (30-40%)
•Language: "rigorous", "validated", "generalizable principles"
•Positioning: Innovation + rigor + relevance balance

MIS Quarterly:

•Emphasize: Theoretical framework (50%+)
•Secondary: Empirical/method (30%)
•Language: "theoretical mechanisms", "conceptual development"
•Positioning: Conceptual depth and novel theorizing

ISR:

•Emphasize: Rigor + relevance balance
•Strong methodology required
•Language: "robust evidence", "practical implications"
•Positioning: Applied context with theoretical grounding

Conference (ICIS/AMCIS):

•More concise scope acceptable
•Emphasize: Novelty and timeliness
•Language: "emerging", "preliminary evidence"
•Positioning: Work-in-progress framing acceptable

Examples

See examples/ directory for:

•example-empirical-artifact.md - Hybrid empirical + design science paper
•example-pure-theory.md - Pure theoretical contribution
•example-design-science.md - Artifact-focused paper

Quality Checklist

Before finalizing, verify:

Completeness:

• All 9 sections present
• Auto-detected config box at top
• Key data points have actual numbers (not TBD)
• Contributions ranked and labeled with emphasis %
• Emphasis % sum to 100%
• Story arc has all 5 acts in boxes
• Language guidance has ≥3 DO and ≥3 AVOID

Clarity:

• Overall contribution readable in <30 seconds
• Each contribution ≤5 bullets
• Each implication ≤3 bullets
• No jargon without definition
• Scope boundaries specific (not vague)

Strategic Value:

• Emphasis % match evidence strength
• Language guidance addresses actual over-claiming risks
• Venue positioning explains fit
• Story arc coherent (acts flow logically)
• Honest about limitations vs design choices

Brevity:

• Total length: 150-300 lines
• No section exceeds 60 lines
• Visual elements ≤5 boxes (story arc only)
• No redundant content across sections

Common Pitfalls to Avoid

❌ Too Vague: "Significant results" → ✅ "β=0.041***, 8/8 positive associations" ❌ Too Long: >300 lines defeats "concise overview" purpose ❌ Over-Claiming: Language doesn't reflect strategic constraints ❌ Mismatch: Story arc emphasizes different contribution than rankings ❌ Incomplete: Missing sections or placeholder text like "TBD" ❌ Fabricated: Including numbers not in source documents

Success Criteria

Document succeeds if user can:

•✅ Pitch paper in 30 seconds (Overall Contribution)
•✅ Decide abstract emphasis (Contribution % and ranking)
•✅ Write introduction (5-Act Arc)
•✅ Avoid over-claiming (Language Guidance)
•✅ Position for venue (Paper Positioning section)
•✅ Structure discussion (Implications sections)
•✅ Make strategic decisions (Primary vs Secondary emphasis)

If user still asks "What should I emphasize?" → Skill failed If user can immediately start writing → Skill succeeded