@wentorai/research-plugins 1.2.2 → 1.3.0

package/skills/research/deep-research/research-cog/SKILL.md

@@ -1,153 +0,0 @@
----
-name: research-cog
-description: "Deep research powered by CellCog for comprehensive investigations"
-metadata:
-  openclaw:
-    emoji: "🔍"
-    category: "research"
-    subcategory: "deep-research"
-    keywords: ["deep research", "multi-source investigation", "AI literature review", "automated literature synthesis"]
-    source: "https://github.com/cellcog/research-cog"
----
-
-# Research Cog - Deep Research Powered by CellCog
-
-**#1 on DeepResearch Bench (Feb 2026).** Your AI research analyst for comprehensive, citation-backed research on any topic.
-
-Leaderboard: https://huggingface.co/spaces/muset-ai/DeepResearch-Bench-Leaderboard
-
-## Prerequisites
-
-This skill requires the `cellcog` skill for SDK setup and API calls.
-
-```bash
-clawhub install cellcog
-```
-
-**Read the cellcog skill first** for SDK setup. This skill shows you what's possible.
-
-**Quick pattern (v1.0+):**
-```python
-# Fire-and-forget - returns immediately
-result = client.create_chat(
-  prompt="[your research query]",
-  notify_session_key="agent:main:main",
-  task_label="research-task",
-  chat_mode="agent team"  # Deep research
-)
-
-# Daemon notifies you when complete - do NOT poll
-```
-
-## What You Can Research
-
-### Competitive Analysis
-Analyze companies against their competitors with structured insights:
-- **Company vs. Competitors**: "Compare Stripe vs Square vs Adyen - market positioning, pricing, features, strengths/weaknesses"
-- **SWOT Analysis**: "Create a SWOT analysis for Shopify in the e-commerce platform market"
-- **Market Positioning**: "How does Notion position itself against Confluence, Coda, and Obsidian?"
-- **Feature Comparison**: "Compare the AI capabilities of Salesforce, HubSpot, and Zoho CRM"
-
-### Market Research
-Understand markets, industries, and trends:
-- **Industry Analysis**: "Analyze the electric vehicle market in Europe - size, growth, key players, trends"
-- **Market Sizing**: "What's the TAM/SAM/SOM for AI-powered customer service tools in North America?"
-- **Trend Analysis**: "What are the emerging trends in sustainable packaging for 2026?"
-- **Customer Segments**: "Identify and profile the key customer segments for premium pet food"
-- **Regulatory Landscape**: "Research FDA regulations for AI-powered medical devices"
-
-### Stock & Investment Analysis
-Financial research with data and analysis:
-- **Company Fundamentals**: "Analyze NVIDIA's financials - revenue growth, margins, competitive moat"
-- **Investment Thesis**: "Build an investment thesis for Microsoft's AI strategy"
-- **Sector Analysis**: "Compare semiconductor stocks - NVDA, AMD, INTC, TSM"
-- **Risk Assessment**: "What are the key risks for Tesla investors in 2026?"
-- **Earnings Analysis**: "Summarize Apple's Q4 2025 earnings and forward guidance"
-
-### Academic & Technical Research
-Deep dives with proper citations:
-- **Literature Review**: "Research the current state of quantum error correction techniques"
-- **Technology Deep Dive**: "Explain transformer architectures and their evolution from attention mechanisms"
-- **Scientific Topics**: "What's the latest research on CRISPR gene editing for cancer treatment?"
-- **Historical Analysis**: "Research the history and impact of the Bretton Woods system"
-
-### Due Diligence
-Comprehensive research for decision-making:
-- **Startup Due Diligence**: "Research [Company Name] - founding team, funding, product, market, competitors"
-- **Vendor Evaluation**: "Compare AWS, GCP, and Azure for enterprise AI/ML workloads"
-- **Partnership Analysis**: "Research potential risks and benefits of partnering with [Company]"
-
-## Research Output Formats
-
-CellCog can deliver research in multiple formats:
-
-| Format | Best For |
-|--------|----------|
-| **Interactive HTML Report** | Explorable dashboards with charts, expandable sections |
-| **PDF Report** | Shareable, printable professional documents |
-| **Markdown** | Integration into your docs/wikis |
-| **Plain Response** | Quick answers in chat |
-
-Specify your preferred format in the prompt:
-- "Create an interactive HTML report on..."
-- "Generate a PDF research report analyzing..."
-- "Give me a markdown summary of..."
-
-## When to Use Agent Team Mode
-
-For research, **always use** `chat_mode="agent team"` (the default).
-
-Agent team mode enables:
-- Multi-source research and cross-referencing
-- Citation verification
-- Deeper analysis with multiple reasoning passes
-- Higher quality, more comprehensive outputs
-
-Use `chat_mode="agent"` only for trivial lookups like "What's Apple's stock ticker?"
-
-## Research Quality Features
-
-### Citations (On Request)
-**Citations are NOT automatic.** CellCog focuses on delivering accurate, well-researched content by default.
-
-If you need citations:
-- **Explicitly request them**: "Include citations for all factual claims with source URLs"
-- **Specify format**: "Provide citations as footnotes" or "Include a references section at the end"
-- **Indicate placement**: "Citations inline" vs "Citations in appendix"
-
-Without explicit citation requests, CellCog prioritizes delivering accurate information efficiently.
-
-### Data Accuracy
-CellCog cross-references multiple sources for financial and statistical data, ensuring accuracy even without explicit citations.
-
-### Structured Analysis
-Complex research is organized with clear sections, executive summaries, and actionable insights.
-
-### Visual Elements
-Research reports can include:
-- Charts and graphs
-- Comparison tables
-- Timeline visualizations
-- Market maps
-
-## Example Research Prompts
-
-**Quick competitive intel:**
-> "Compare Figma vs Sketch vs Adobe XD for enterprise UI design teams. Focus on collaboration features, pricing, and Figma's position after the Adobe acquisition failed."
-
-**Deep market research:**
-> "Create a comprehensive market research report on the AI coding assistant market. Include market size, growth projections, key players (GitHub Copilot, Cursor, Codeium, etc.), pricing models, and enterprise adoption trends. Deliver as an interactive HTML report."
-
-**Investment analysis:**
-> "Build an investment analysis for Palantir (PLTR). Cover business model, government vs commercial revenue mix, AI product strategy, valuation metrics, and key risks. Include relevant charts."
-
-**Academic deep dive:**
-> "Research the current state of nuclear fusion energy. Cover recent breakthroughs (NIF, ITER, private companies like Commonwealth Fusion), technical challenges remaining, timeline to commercial viability, and investment landscape."
-
-## Tips for Better Research
-
-- **Be specific**: "AI market" is vague. "Enterprise AI automation market in healthcare" is better.
-- **Specify timeframe**: "Recent" is ambiguous. "2025-2026" or "last 6 months" is clearer.
-- **Define scope**: "Compare everything about X and Y" leads to bloat. "Compare X and Y on pricing, features, and market positioning" is focused.
-- **Request structure**: "Include executive summary, key findings, and recommendations" helps organize output.
-- **Mention output format**: "Deliver as PDF" or "Create interactive HTML dashboard" gets you the right format.

package/skills/research/methodology/academic-mentor-guide/SKILL.md

@@ -1,169 +0,0 @@
----
-name: academic-mentor-guide
-description: "AI research advisor for graduate students navigating academic careers"
-metadata:
-  openclaw:
-    emoji: "🎓"
-    category: "research"
-    subcategory: "methodology"
-    keywords: ["academic mentoring", "graduate research", "PhD guidance", "research strategy", "academic career", "thesis advising"]
-    source: "https://github.com/AcademicSkills/academic-mentor-guide"
----
-
-# Academic Mentor Guide
-
-An AI-powered research advisor that provides strategic guidance for graduate students and early-career researchers navigating the complexities of academic research. Covers research question development, methodology selection, publication strategy, time management, advisor relationships, and career planning.
-
-## Overview
-
-Graduate school presents a unique set of challenges that extend far beyond coursework: identifying a viable research niche, developing original contributions, managing the advisor-advisee relationship, navigating peer review, and building a professional network. Many students receive excellent technical training but lack strategic guidance on how to position their work, when to pivot, and how to build a sustainable research program. This skill fills that gap by providing structured, evidence-based advice on the strategic dimensions of academic research.
-
-The skill is designed as a conversational advisor. Present your situation, dilemma, or question, and receive tailored advice drawing from best practices in research methodology, academic career development, and graduate education scholarship. It covers all stages from prospective student through early-career faculty.
-
-## Research Question Development
-
-### From Interest to Researchable Question
-
-```
-Stage 1: Broad Interest
-  "I'm interested in natural language processing for healthcare"
-
-Stage 2: Focused Topic
-  "Clinical note understanding using large language models"
-
-Stage 3: Research Gap
-  "Existing LLMs hallucinate medical facts in clinical summaries"
-
-Stage 4: Specific Question
-  "How does retrieval-augmented generation reduce factual
-   hallucinations in clinical note summarization compared to
-   standard fine-tuning?"
-
-Stage 5: Testable Hypothesis
-  H1: "RAG-augmented summarization produces fewer factual errors
-       (as measured by clinical expert annotation) than fine-tuned
-       models without retrieval, on the MIMIC-III discharge
-       summary dataset."
-```
-
-### Question Quality Checklist
-
-A strong research question should be:
-
-| Criterion | Test | Example Pass/Fail |
-|-----------|------|------------------|
-| **Specific** | Can you explain it in one sentence? | Pass: "Does X affect Y in context Z?" |
-| **Novel** | Does it add something new? | Fail: "Can neural networks classify images?" |
-| **Feasible** | Can you answer it with available resources? | Fail if you need data you cannot access |
-| **Significant** | Would the answer matter to the field? | Pass: fills an identified gap in literature |
-| **Measurable** | Can you define success criteria? | Pass: specific metrics and benchmarks identified |
-
-## Methodology Selection
-
-### Matching Method to Question
-
-| Question Type | Suggested Methods | Considerations |
-|--------------|-------------------|----------------|
-| "Does X cause Y?" | RCT, quasi-experiment, IV, DiD | Causation requires design, not statistics |
-| "How does X relate to Y?" | Correlation, regression, SEM | Cannot infer causation from observational data |
-| "What is the experience of...?" | Interviews, phenomenology | Qualitative rigor: saturation, reflexivity |
-| "How much of X exists?" | Survey, secondary data analysis | Sampling design determines generalizability |
-| "Does method A outperform B?" | Benchmark experiments, ablation | Control for confounds; use multiple metrics |
-| "What themes emerge from...?" | Grounded theory, thematic analysis | Systematic coding, inter-rater reliability |
-
-### Mixed Methods Considerations
-
-When a single method cannot fully address your question, consider:
-
-1. **Sequential explanatory**: Quantitative first, then qualitative to explain results.
-2. **Sequential exploratory**: Qualitative first, then quantitative to test emergent themes.
-3. **Convergent parallel**: Both simultaneously, then integrate at interpretation.
-
-Key rule: each method must be rigorous on its own terms. Mixed methods is not a shortcut to avoid depth in either tradition.
-
-## Publication Strategy
-
-### Planning Your Publication Pipeline
-
-```
-Year 1-2 of PhD:
-  - Systematic/scoping review paper (establishes expertise, builds lit knowledge)
-  - Workshop paper or short paper (low-stakes venue to get feedback)
-
-Year 2-3:
-  - Main contribution paper #1 (core of dissertation)
-  - Tool/dataset paper if applicable (broadens impact)
-
-Year 3-4:
-  - Main contribution paper #2 (extends or deepens #1)
-  - Invited talks or poster presentations at top conferences
-
-Year 4-5 (if applicable):
-  - Synthesis/position paper (demonstrates thought leadership)
-  - Dissertation defense (integrates everything)
-```
-
-### Venue Selection Framework
-
-| Criterion | Conference | Journal |
-|-----------|-----------|---------|
-| Speed of dissemination | Fast (3-6 months) | Slow (6-18 months) |
-| Networking | Presentations, hallway conversations | Less direct interaction |
-| Page limits | Usually strict (8-12 pages) | Flexible (often 20-40 pages) |
-| Revision opportunity | Usually accept/reject | Revise and resubmit common |
-| Impact factor | Varies by field | Often higher for journals |
-| Best for | Incremental results, new ideas | Complete, mature studies |
-
-## Time Management for Researchers
-
-### The Weekly Research Schedule
-
-Protect blocks of uninterrupted time for deep research work:
-
-- **Monday-Wednesday**: 3-4 hour morning blocks for writing and analysis (most cognitively demanding tasks).
-- **Thursday**: Meetings, literature reading, email, administrative tasks.
-- **Friday**: Experimentation, coding, data collection.
-- **Weekends**: Optional reading; avoid burnout by setting boundaries.
-
-### The 80/20 Rule for PhD Progress
-
-- Spend 80% of research time on your core contribution (the work that becomes your dissertation).
-- Spend 20% on exploratory reading, skill building, and collaborative side projects.
-- Audit your time monthly: if the ratio has flipped, recalibrate.
-
-## Navigating the Advisor Relationship
-
-### Communication Best Practices
-
-1. **Regular updates**: Send a brief weekly email summarizing progress, blockers, and next steps.
-2. **Come prepared**: Every meeting should have a written agenda with specific questions.
-3. **Manage expectations**: If a deadline is at risk, communicate early, not at the last minute.
-4. **Seek feedback explicitly**: "Could you review Section 3? I'm unsure about the framing of the contribution."
-5. **Document agreements**: Follow up meetings with an email summarizing decisions and action items.
-
-### When Things Go Wrong
-
-- **Advisor is unresponsive**: Escalate gradually: email reminder after 1 week, in-person conversation after 2 weeks, committee member consultation after a month.
-- **Disagreement on direction**: Present your reasoning in writing with evidence. If the impasse continues, involve a committee member as mediator.
-- **Advisor leaves institution**: This is more common than students expect. Document your progress, know your program's policies, and have a backup advisor relationship.
-
-## Career Decision Framework
-
-### Academic vs. Industry vs. Alternative Careers
-
-Consider these factors honestly:
-
-| Factor | Academia | Industry Research | Industry Applied |
-|--------|---------|-------------------|-----------------|
-| Autonomy | High (after tenure) | Moderate | Lower |
-| Compensation | Lower | Higher | Higher |
-| Geographic flexibility | Very low | Moderate | High |
-| Publication freedom | High | Varies | Limited |
-| Job security | Low until tenure | Moderate | Moderate |
-| Work-life balance | Self-managed (often poor) | Structured | Structured |
-
-## References
-
-- Dericks, G., et al. (2019). PhD Supervision: Understanding the Factors That Lead to Success. *Studies in Higher Education*.
-- Lovitts, B. E. (2005). Being a Good Course-Taker Is Not Enough. *Studies in Higher Education*, 30(2), 137-154.
-- Bolker, J. (1998). *Writing Your Dissertation in Fifteen Minutes a Day*. Henry Holt.

package/skills/research/methodology/deep-innovator-guide/SKILL.md

@@ -1,242 +0,0 @@
----
-name: deep-innovator-guide
-description: "AI research for idea generation and scientific discovery methods"
-metadata:
-  openclaw:
-    emoji: "💡"
-    category: "research"
-    subcategory: "methodology"
-    keywords: ["idea generation", "scientific discovery", "innovation", "research ideation", "creative research", "hypothesis generation"]
-    source: "https://github.com/AcademicSkills/deep-innovator-guide"
----
-
-# Deep Innovator Guide
-
-A skill for systematic idea generation and scientific discovery in academic research. Combines structured creativity techniques with computational methods to help researchers identify novel research directions, generate testable hypotheses, and find unexpected connections across disciplines.
-
-## Overview
-
-Scientific innovation rarely happens through pure serendipity. Research on scientific breakthroughs consistently shows that novel ideas emerge at the intersection of existing knowledge domains, from systematic exploration of anomalies, or through the creative recombination of known concepts. This skill provides structured methods for each of these innovation pathways, making the ideation process more systematic without sacrificing creativity.
-
-The skill is designed for researchers at any career stage who feel stuck in incremental thinking, who want to explore new research directions, or who need to generate a portfolio of potential research questions for a grant proposal or thesis prospectus. It draws on techniques from design thinking, TRIZ (Theory of Inventive Problem Solving), bibliometric innovation detection, and computational creativity research.
-
-## Structured Ideation Methods
-
-### Method 1: Cross-Domain Analogy Mining
-
-The most frequently cited source of scientific breakthroughs is the transfer of ideas from one domain to another. This method systematizes that process.
-
-```
-Step 1: Define your problem abstractly
-  Concrete: "How can we detect early-stage Alzheimer's from speech patterns?"
-  Abstract: "How to detect gradual degradation of a complex system from
-             its output signals?"
-
-Step 2: Identify analogous domains
-  - Manufacturing: detecting machine wear from vibration patterns
-  - Ecology: detecting ecosystem decline from biodiversity metrics
-  - Software: detecting code quality degradation from commit patterns
-  - Music: detecting performer fatigue from playing characteristics
-
-Step 3: Study solutions in analogous domains
-  - Manufacturing uses wavelet analysis on vibration frequency shifts
-  - Ecology uses species abundance distribution curves
-  - Software uses entropy measures on code change patterns
-
-Step 4: Transfer and adapt
-  Idea: Apply wavelet decomposition to speech prosody features to detect
-        frequency-domain shifts that precede clinical Alzheimer's diagnosis.
-```
-
-### Method 2: Contradiction Resolution (TRIZ-Inspired)
-
-```python
-def identify_research_contradictions(domain_knowledge: dict) -> list:
-    """
-    Identify contradictions in current knowledge that could lead to
-    innovative research questions.
-
-    Types of contradictions:
-    1. Technical: Improving A makes B worse
-    2. Physical: System needs property X and not-X simultaneously
-    3. Empirical: Study A finds effect, Study B finds opposite
-    """
-    contradictions = []
-
-    # Type 1: Technical contradictions
-    # "Increasing model accuracy requires more data, but more data
-    #  increases privacy risk"
-    # Resolution principle: SEPARATION (separate the contradictory
-    #  requirements in time, space, or condition)
-    # Idea: Federated learning separates data access from model training
-
-    # Type 2: Empirical contradictions
-    # "Drug X shows efficacy in lab but not in clinical trials"
-    # Resolution principle: INVESTIGATE THE BOUNDARY
-    # Idea: What differs between lab and clinical contexts? (dosage,
-    #  population heterogeneity, adherence, bioavailability)
-
-    return contradictions
-```
-
-### Method 3: Gap Matrix Analysis
-
-Create a matrix crossing two relevant dimensions to identify unexplored combinations:
-
-| | Method A | Method B | Method C | Method D |
-|---|---------|---------|---------|---------|
-| **Population 1** | Studied | Studied | GAP | Studied |
-| **Population 2** | Studied | GAP | GAP | Studied |
-| **Population 3** | GAP | GAP | GAP | GAP |
-| **Population 4** | Studied | GAP | Studied | GAP |
-
-Each GAP cell represents a potential novel study. Prioritize GAP cells by:
-- Theoretical interest (would the finding be surprising or informative?)
-- Feasibility (can you access this population and apply this method?)
-- Impact (would filling this gap change practice or theory?)
-
-## Computational Innovation Detection
-
-### Bibliometric Novelty Signals
-
-```python
-def detect_novelty_signals(papers: list) -> dict:
-    """
-    Detect signals of emerging innovation in a set of papers.
-    """
-    signals = {}
-
-    # Signal 1: New keyword combinations
-    # Papers that combine keywords from traditionally separate fields
-    # indicate cross-disciplinary innovation
-    signals['novel_keyword_pairs'] = find_new_keyword_cooccurrences(papers)
-
-    # Signal 2: Citation bridge papers
-    # Papers cited by two communities that rarely cite each other
-    signals['bridge_papers'] = find_citation_bridges(papers)
-
-    # Signal 3: Accelerating citation curves
-    # Recently published papers gaining citations faster than average
-    signals['sleeping_beauties'] = find_acceleration_anomalies(papers)
-
-    # Signal 4: Method migration
-    # Methods first published in Domain A appearing in Domain B
-    signals['method_transfers'] = find_method_migrations(papers)
-
-    # Signal 5: Terminology shifts
-    # Old terms being replaced by new ones (signals paradigm change)
-    signals['terminology_evolution'] = find_term_replacements(papers)
-
-    return signals
-```
-
-### Hypothesis Generation from Data
-
-```
-Observation-Driven Hypothesis Generation:
-
-1. ANOMALY DETECTION
-   Look for unexpected patterns in existing data:
-   - Results that deviate from theoretical predictions
-   - Subgroups that behave differently from the main sample
-   - Variables that correlate when they "should not"
-   Question template: "Why does [anomaly] occur in [context]?"
-
-2. BOUNDARY CONDITIONS
-   Test where known effects break down:
-   - At what sample size does effect X disappear?
-   - In which populations does intervention Y not work?
-   - Under what conditions does relationship Z reverse?
-   Question template: "Under what conditions does [known effect] not hold?"
-
-3. MECHANISM GAPS
-   Identify well-documented effects without known mechanisms:
-   - "We know X causes Y but not how"
-   - "The mediating pathway is unknown"
-   Question template: "What mechanism explains the effect of [X] on [Y]?"
-
-4. SCALE TRANSITIONS
-   Phenomena may behave differently at different scales:
-   - Lab findings that may not scale to real-world settings
-   - Individual-level effects that may not apply at group level
-   Question template: "Does [finding at scale A] hold at [scale B]?"
-```
-
-## Innovation Portfolio Management
-
-### Balancing Risk and Impact
-
-Maintain a portfolio of research ideas at different risk-reward levels:
-
-| Category | Risk | Potential Impact | Portfolio % | Examples |
-|----------|------|-----------------|------------|---------|
-| **Incremental** | Low | Moderate | 40-50% | Replicate with new data, extend to new population |
-| **Substantial** | Medium | High | 30-40% | New method applied to existing problem |
-| **Transformative** | High | Very High | 10-20% | New framework, cross-disciplinary breakthrough |
-| **Exploratory** | Very High | Unknown | 5-10% | Speculative, curiosity-driven |
-
-### Idea Evaluation Rubric
-
-Score each idea on a 1-5 scale for each criterion:
-
-| Criterion | Question | Weight |
-|-----------|----------|--------|
-| Novelty | Has this been done before? | 25% |
-| Feasibility | Can I do this with my resources? | 25% |
-| Significance | Would the answer change the field? | 25% |
-| Fundability | Would an agency fund this? | 15% |
-| Personal fit | Does this align with my expertise and interests? | 10% |
-
-```python
-def evaluate_research_idea(scores: dict, weights: dict = None) -> dict:
-    """
-    Evaluate a research idea using the weighted scoring rubric.
-    """
-    if weights is None:
-        weights = {
-            'novelty': 0.25, 'feasibility': 0.25, 'significance': 0.25,
-            'fundability': 0.15, 'personal_fit': 0.10
-        }
-
-    weighted_score = sum(scores[k] * weights[k] for k in scores)
-    max_score = 5.0
-
-    return {
-        'scores': scores,
-        'weighted_total': round(weighted_score, 2),
-        'max_possible': max_score,
-        'percentage': round(weighted_score / max_score * 100, 1),
-        'recommendation': (
-            'Strong pursue' if weighted_score >= 4.0 else
-            'Worth developing' if weighted_score >= 3.0 else
-            'Needs refinement' if weighted_score >= 2.0 else
-            'Consider alternatives'
-        )
-    }
-```
-
-## Collaborative Ideation Sessions
-
-### Structured Brainstorming for Research Teams
-
-1. **Individual divergence** (10 min): Each team member silently generates 5-10 ideas.
-2. **Round-robin sharing** (15 min): Each person shares one idea at a time, no criticism allowed.
-3. **Affinity clustering** (10 min): Group similar ideas together on a whiteboard.
-4. **Build and combine** (15 min): Look for ways to merge or strengthen ideas across clusters.
-5. **Evaluation** (10 min): Apply the scoring rubric to the top 5 ideas.
-6. **Selection** (5 min): Choose 1-2 ideas to develop further with feasibility analysis.
-
-## Best Practices
-
-- Schedule regular ideation time (monthly) rather than waiting for inspiration to strike.
-- Keep an "idea notebook" where you record fleeting thoughts before they are lost.
-- Read outside your field deliberately. Subscribe to journals in adjacent disciplines.
-- When you find an anomaly in your data, resist the urge to explain it away immediately. Investigate it.
-- Test ideas cheaply before committing resources. A quick literature search or back-of-the-envelope calculation can eliminate weak ideas early.
-- Discuss ideas with people outside your specialty. They bring fresh perspectives and ask different questions.
-
-## References
-
-- Uzzi, B., et al. (2013). Atypical Combinations and Scientific Impact. *Science*, 342(6157), 468-472.
-- Foster, J. G., et al. (2015). Tradition and Innovation in Scientists' Research Strategies. *American Sociological Review*, 80(5), 875-908.
-- Wang, D., Song, C., & Barabasi, A.-L. (2013). Quantifying Long-Term Scientific Impact. *Science*, 342(6154), 127-132.