Ideation Engine
Skill Metadata
- Skill Name: Ideation Engine
- Version: 2.0.0
- Purpose: Systematically generate, evaluate, and refine novel research ideas with quantified novelty assessment and contribution mapping
- Trigger: User wants to develop a new research idea, find a research gap, formulate hypotheses, or assess novelty of a proposed study
- Integration: Feeds into Literature Command (02) for gap validation, Methodology Architect (03) for design, and Writing Engine (06) for articulation
ACTIVATE: IDEATION
When the user invokes the Ideation Engine, execute the following five-phase protocol. Do not skip phases. Do not proceed to the next phase until the current one produces output that meets the quality criteria specified.
PHASE 1: FIELD CARTOGRAPHY
Objective: Map the intellectual terrain of the target domain to locate frontiers, adjacencies, and contrarian opportunities.
1.1 Frontier Scan
- Identify the 5 most actively cited research streams in the domain over the past 3 years
- For each stream, catalog: leading authors, core journals, dominant methodologies, typical sample sizes, prevailing effect sizes
- Tag each stream as: Advancing (growing citation velocity), Plateauing (stable citation velocity), or Retreating (declining citation velocity)
- Flag streams where methodological sophistication exceeds theoretical depth — these represent opportunities for theoretical contribution
1.2 Adjacency Scan
- Identify 3 neighboring fields whose methods, theories, or data could cross-pollinate into the target domain
- For each adjacency, specify:
- Transferable construct: What concept or method could be imported?
- Adaptation requirement: What modifications are needed for the target domain?
- Novelty potential: Low / Medium / High — with justification
- Risk assessment: Likelihood that the transfer produces trivial or forced results
1.3 Contrarian Probe
- Identify 3 widely accepted claims in the field that have never been rigorously challenged
- For each claim, assess:
- Evidential basis: How strong is the supporting evidence actually? (Rate: Iron-clad / Strong / Moderate / Weak / Folklore)
- Challenge feasibility: Could a well-designed study plausibly overturn this? (Rate: High / Medium / Low)
- Impact if overturned: How much would the field change? (Rate: Paradigm shift / Major revision / Moderate correction / Minor nuance)
- Generate one contrarian research question per challengeable claim
Quality Gate: Phase 1 output must include at least 5 frontier streams, 3 adjacencies, and 3 contrarian probes. If the domain is too narrow to support this, expand scope to the parent discipline.
PHASE 2: GAP TYPOLOGY
Objective: Classify identified gaps using the seven-type taxonomy to ensure comprehensive coverage and avoid over-concentration on a single gap type.
Gap Types
| # | Gap Type | Definition | Diagnostic Question |
|---|---|---|---|
| 1 | Knowledge Gap | What is not yet known about a phenomenon | "What do we still not understand about X?" |
| 2 | Method Gap | Existing methods are inadequate or absent | "How could we study X more rigorously?" |
| 3 | Theory Gap | No adequate theoretical explanation exists | "Why does X occur — and can we explain it?" |
| 4 | Application Gap | Knowledge exists but is not applied | "How can we use what we know about X to solve Y?" |
| 5 | Replication Gap | Findings lack independent replication | "Has X been replicated in different contexts?" |
| 6 | Population Gap | Findings limited to specific populations | "Does X hold for underrepresented group Z?" |
| 7 | Mechanistic Gap | Correlation known but mechanism unknown | "Through what process does X lead to Y?" |
Gap Assessment Matrix
For each identified gap, score on:
- Severity (1-5): How significant is the gap's absence for the field?
- Feasibility (1-5): How tractable is addressing this gap with available resources?
- Impact (1-5): How much would closing this gap advance knowledge?
- Novelty (1-5): How many existing studies already target this gap?
Compute the Gap Priority Index (GPI) = (Severity × 0.3) + (Feasibility × 0.2) + (Impact × 0.3) + (Novelty × 0.2)
Rank gaps by GPI descending. Only gaps scoring ≥ 3.0 should advance to Phase 3.
Quality Gate: Must produce at least 3 gaps spanning at least 2 different gap types. No more than 40% of gaps should be of a single type.
PHASE 3: QUESTION FORGE
Objective: Transform validated gaps into precise, answerable, and novel research questions.
Research Question Template
[UNDER WHAT CONDITIONS / HOW / WHY / TO WHAT EXTENT]
does [INDEPENDENT VARIABLE / INTERVENTION / EXPOSURE]
[affect / influence / moderate / mediate / predict]
[DEPENDENT VARIABLE / OUTCOME]
in [POPULATION / CONTEXT]
given [THEORETICAL LENS / MECHANISM]?
Five-Dimensional Scoring
| Dimension | Weight | Scoring Criteria |
|---|---|---|
| Specificity | 0.20 | 1=Vague topic area → 5=Precise variables, direction, and population specified |
| Answerability | 0.20 | 1=Unanswerable with any method → 5=Clearly answerable with standard methods |
| Novelty | 0.25 | 1=Already answered in literature → 5=No prior work addresses this |
| Impact | 0.20 | 1=Niche interest only → 5=Field-altering implications |
| Groundedness | 0.15 | 1=No theoretical basis → 5=Directly derived from established theory |
Research Question Quality Index (RQI) = Σ(Dimension Score × Weight)
RQI Interpretation:
- 4.0–5.0: Excellent — proceed directly to Phase 4
- 3.0–3.9: Good — refine wording and re-score
- 2.0–2.9: Marginal — return to Phase 2 and identify alternative gap
- Below 2.0: Unviable — discard and restart
Question Refinement Checklist
- Variables are operationally definable
- The question implies a falsifiable hypothesis
- The population is specified and accessible
- The scope is neither too broad nor too narrow
- The question is not a disguised "does X exist?" (trivial)
- The question does not presuppose its answer
- Alternative explanations are acknowledged
Quality Gate: Must produce at least 2 research questions per gap, select the highest RQI question, and verify it scores ≥ 3.5.
PHASE 4: CONTRIBUTION MAPPING
Objective: Forecast and articulate the specific contributions the proposed research will make across four domains.
Contribution Domains
4.1 Theoretical Contribution
- New theory construction: Does this create a new theoretical framework?
- Theory extension: Does this extend an existing theory to a new domain?
- Theory refinement: Does this refine or qualify an existing theory?
- Theory integration: Does this bridge two or more theoretical traditions?
- Theory falsification: Does this test a theory's predictions under new conditions?
4.2 Methodological Contribution
- New measure/instrument development
- Novel analytical approach or statistical technique
- Improved research design for existing questions
- New data collection methodology
- Validation of existing methods in new contexts
4.3 Practical Contribution
- Direct intervention or policy implication
- Clinical or organizational application
- Stakeholder decision-making support
- Tool or resource creation
- Training or educational application
4.4 Field Impact
- Opens new research stream
- Resolves standing debate
- Shifts methodological norms
- Bridges sub-disciplines
- Influences funding priorities
Contribution Statement Template
This study contributes to [FIELD] by [CONTRIBUTION TYPE] [WHAT IS CONTRIBUTED],
which [HOW IT ADVANCES] [SPECIFIC AREA], thereby [BROADER IMPLICATION].
Unlike prior work that [WHAT OTHERS DID], this study [WHAT IS DIFFERENT AND BETTER].
Quality Gate: The research must articulate at least one contribution in at least two domains. At least one contribution must be rated "substantial" (not merely incremental).
PHASE 5: NOVELTY CERTIFICATE
Objective: Quantify the novelty of the proposed research using the four-dimension scoring system and issue a formal novelty assessment.
Scoring Dimensions (0–25 each)
| Dimension | 21–25 | 16–20 | 11–15 | 6–10 | 0–5 |
|---|---|---|---|---|---|
| First-of-its-kind | No prior work in this space | One preliminary study exists | A few studies exist but are narrow | Several studies exist | Well-trodden territory |
| Methodological Innovation | Entirely new method or paradigm | Significant adaptation of existing method | Moderate methodological refinement | Minor methodological tweak | Standard method, no innovation |
| Theoretical Advancement | New theory or paradigm shift | Major theory extension | Moderate theory refinement | Minor theoretical nuance | No theoretical contribution |
| Empirical Impact | Findings would reshape the field | Findings would influence multiple sub-fields | Findings would influence one sub-field | Findings would be cited but not transformative | Findings would have limited reach |
Total Score Bands
| Score Range | Classification | Description |
|---|---|---|
| 90–100 | Paradigm Shift | Field-altering work; will be cited for decades |
| 75–89 | Major Contribution | Substantially advances understanding; high-impact publication likely |
| 60–74 | Solid Contribution | Meaningful advance; suitable for strong specialty journal |
| 45–59 | Incremental | Adds to literature but does not shift understanding |
| 30–44 | Marginal | Limited novelty; significant revision needed |
| 0–29 | Redundant | Insufficient novelty; do not pursue in current form |
Novelty Certificate Format
╔══════════════════════════════════════════════╗
║ NOVELTY CERTIFICATE ║
╠══════════════════════════════════════════════╣
║ Research Topic: [TOPIC] ║
║ Date: [DATE] ║
║ ║
║ First-of-its-kind: [XX]/25 ║
║ Methodological Innovation: [XX]/25 ║
║ Theoretical Advancement: [XX]/25 ║
║ Empirical Impact: [XX]/25 ║
║ ───────────────────────────────── ║
║ TOTAL NOVELTY SCORE: [XX]/100 ║
║ Classification: [CLASSIFICATION] ║
║ ║
║ Key Strength: [STRONGEST DIMENSION] ║
║ Key Weakness: [WEAKEST DIMENSION] ║
║ Recommended Action: [PROCEED/REVISE/PIVOT] ║
╚══════════════════════════════════════════════╝
Quality Gate: If the total novelty score is below 45, return to Phase 1 with expanded scope. If between 45–59, identify specific strategies from the novelty_calculator.md to strengthen the weakest dimensions before proceeding.
OUTPUT FORMAT
The Ideation Engine produces the following structured output:
# Ideation Engine Output
## 1. Field Cartography
### Frontier Streams
[Numbered list with tags]
### Adjacency Opportunities
[Numbered list with novelty potential]
### Contrarian Probes
[Numbered list with challengeability assessment]
## 2. Gap Analysis
[Gap Priority Matrix with GPI scores and rankings]
## 3. Research Questions
[Top questions with RQI scores and refinement notes]
## 4. Contribution Map
[Contributions by domain with substantiveness rating]
## 5. Novelty Certificate
[Full certificate with scores and classification]
## 6. Recommended Next Steps
[Suggested integration with Literature Command and Methodology Architect]
INTEGRATION NOTES
With Literature Command (02_literature_command)
- Feed identified gaps and questions into the Literature Command for systematic validation
- Use the Deep Synthesis Engine to confirm gaps are genuine (not already filled)
- Use Citation Strategy to identify which works to position against
With Methodology Architect (03_methodology_architect)
- Feed the top-ranked research question and contribution map to the Methodology Architect
- The methodology must align with the gap type identified (e.g., Method Gaps require methodological innovation)
- Pre-registration should reference the novelty certificate's dimensions
With Writing Engine (06_writing_engine)
- The Contribution Statement Template feeds directly into the Introduction and Discussion sections
- The Novelty Certificate informs the "contributions" subsection
- Gap typology terms should appear in the literature review framing
With Review Simulator (07_review_simulator)
- Feed the Novelty Certificate to the Review Simulator to predict reviewer objections about novelty claims
- Use contrarian probes to anticipate reviewer skepticism
USAGE EXAMPLES
Example 1: Exploring a new domain
User: "I want to study AI-assisted medical diagnosis but I'm not sure where the real gaps are."
→ Execute all 5 phases with "AI-assisted medical diagnosis" as the target domain.
Example 2: Refining an existing idea
User: "I think there's a gap in how mindfulness interventions affect decision-making under uncertainty, but I'm not sure if it's novel enough."
→ Skip Phase 1, validate the gap in Phase 2, forge questions in Phase 3, and assess novelty in Phase 5.
Example 3: Assessing novelty of a drafted proposal
User: "Here's my research proposal [PASTED TEXT]. How novel is it?"
→ Execute Phase 5 only, extracting contributions from the proposal text.
VERSION HISTORY
- v2.0.0 — Added five-dimensional RQI scoring, Gap Priority Index, Novelty Certificate format, integration protocols
- v1.0.0 — Initial release with basic gap identification and question generation