/positioning-engine

name: positioning-engine
description: >
  Activate when the user needs to analyze how their paper positions itself
  relative to the closest existing work. Builds a differentiation matrix,
  identifies unique positioning, and generates a positioning statement that
  strengthens the contribution argumentation. Complements the theory-engine
  by focusing on the competitive landscape of related papers.

> **Orchestration Log**: When this skill is activated, append a log entry to `outputs/orchestration_log.md`: > ``` > ### Skill Activation: Positioning Engine > **Timestamp:** [current date/time] > **Actor:** AI Agent (positioning-engine) > **Input:** Paper draft + [N] comparison papers identified > **Output:** Differentiation matrix with [N] dimensions, positioning_analysis.md saved > ```

Positioning Engine

Core Principle

"How is your paper different from X?" is the question every reviewer asks. This engine produces a systematic answer. It identifies the 5-10 most similar existing papers, builds a structured differentiation matrix, and generates a positioning statement that makes the unique contribution explicit and defensible.

The output directly strengthens the Introduction (gap + contribution paragraphs) and the Discussion (theoretical implications).

When to Activate

• User says "position my paper", "how is this different from X?", "differentiation"
• User says "compare to related work", "positioning analysis", "unique contribution"
• During Phase 2 (Framing) to sharpen the gap and contribution
• When a reviewer challenges the novelty or contribution
• User runs `/analyze-positioning`

Prerequisites

• `draft.md` or `paper.tex` exists (to understand the paper's claims)
• `references.bib` and/or `literature_base.csv` exist (comparison candidates)
• Research questions and contribution are at least tentatively defined

---

Step 1: IDENTIFY Closest Competitors

Finding the Most Similar Papers

Sources for comparison papers:

1. **From the paper itself:** Papers cited in the Introduction and Related Work that address the same or very similar research questions 2. **From `literature_base.csv`:** Papers with the highest topical overlap 3. **From snowballing:** Forward citations of the paper's key references that appeared after those references were published 4. **Direct search:** Search for papers with very similar titles or identical keywords

Selection Criteria

Select 5-10 papers that are the **closest competitors** — papers that a reviewer might say "but Author X already did this." These are papers that:

• Address the same or very similar research questions
• Study the same phenomenon in a similar context
• Use a similar methodology
• Apply the same theoretical lens
• Were published recently (high overlap risk)

Output

For each selected paper, extract:

{
  "bib_key": "[key]",
  "title": "[title]",
  "authors": "[authors]",
  "year": [year],
  "venue": "[venue]",
  "rq": "[their research question or objective]",
  "method": "[their method]",
  "theory": "[their theoretical lens]",
  "context": "[their empirical context]",
  "key_finding": "[their main finding]",
  "similarity_to_ours": "high/medium",
  "citation_count": [N]
}

---

Step 2: BUILD Differentiation Matrix

Matrix Structure

Build a table comparing the competitor papers against YOUR paper across key dimensions:

## Differentiation Matrix

| Dimension | Our Paper | [Author1 (Year)] | [Author2 (Year)] | [Author3 (Year)] | ... |
|-----------|-----------|-------------------|-------------------|-------------------|-----|
| **Research Question** | [our RQ] | [their RQ] | [their RQ] | [their RQ] | |
| **Theoretical Lens** | [our theory] | [their theory] | [their theory] | [their theory] | |
| **Method** | [our method] | [their method] | [their method] | [their method] | |
| **Context/Sample** | [our context] | [their context] | [their context] | [their context] | |
| **Time Period** | [our period] | [their period] | [their period] | [their period] | |
| **Technology Focus** | [our tech] | [their tech] | [their tech] | [their tech] | |
| **Unit of Analysis** | [our UoA] | [their UoA] | [their UoA] | [their UoA] | |
| **Key Contribution** | [our contribution] | [their contribution] | [their contribution] | [their contribution] | |
| **Scope** | [our scope] | [their scope] | [their scope] | [their scope] | |

Dimension Selection

Choose 6-10 dimensions that are most relevant for differentiation. Common dimensions:

| Dimension | When to Use | |-----------|------------| | Research Question / Objective | Always | | Theoretical Lens | When theory is a differentiator | | Method | When method is a differentiator | | Context / Industry / Sample | When context is new | | Time Period / Technology Generation | When recency matters | | Level of Analysis (individual, team, org, industry) | When level differs | | Unit of Analysis | When what's being studied differs | | Geographic / Cultural Context | For cross-cultural studies | | Dependent Variables / Outcomes | When outcomes differ | | Stage of Adoption / Maturity | For implementation studies | | Practitioner vs. Academic Focus | For applied research |

---

Step 3: ANALYZE Positioning

Gap Identification

From the differentiation matrix, identify:

1. **Unique dimensions:** Where does our paper differ from ALL competitors?

• These are the strongest positioning arguments

2. **Underrepresented combinations:** What combination of dimensions is new?

• Even if individual dimensions are not new, the combination may be

3. **Temporal advantage:** What has changed since the competitors published?

• New technology, new phenomena, new policy context

4. **Methodological gap:** What method has NOT been applied to this topic?

• E.g., many surveys, no case studies; many qualitative, no quantitative

5. **Contextual gap:** What context has NOT been studied?

• E.g., many US studies, no European context; many