Skip to content

K-Dense-AI/scientific-agent-skills / skills/literature-review

literature-review

Verified

Conducts systematic, scoping, and narrative literature reviews using PubMed, arXiv, bioRxiv, Semantic Scholar, and other appropriate sources. Use for research synthesis, reproducible literature searches, screening, citation checking, or preparing Markdown and PDF reviews. Tracks search coverage, records versus studies, and evidence limitations; supports meta-analysis planning but does not supply a meta-analysis engine.

assets/review_template.md

View on GitHub
# [Literature Review Title]

**Authors**: [Author Names and Affiliations]
**Date**: [Date]
**Review Type**: [Narrative / Systematic / Scoping / Meta-Analysis / Umbrella Review]
**Review Protocol**: [PROSPERO ID if registered, or state "Not registered"]
**PRISMA Compliance**: [Yes/No/Partial - specify which guidelines]

---

## Abstract

**Background**: [Context and rationale]  
**Objectives**: [Primary and secondary objectives]  
**Methods**: [Databases, dates, selection criteria, quality assessment]  
**Results**: [n studies included; key findings by theme]  
**Conclusions**: [Main conclusions and implications]  
**Registration**: [PROSPERO ID or "Not registered"]  
**Keywords**: [5-8 keywords]

---

## 1. Introduction

### 1.1 Background and Context

[Provide background information on the topic. Establish why this literature review is important and timely. Discuss the broader context and current state of knowledge.]

### 1.2 Scope and Objectives

[Clearly define the scope of the review and state the specific objectives. What questions will this review address?]

**Primary Research Questions:**
1. [Research question 1]
2. [Research question 2]
3. [Research question 3]

### 1.3 Significance

[Explain the significance of this review. Why is it important to synthesize this literature now? What gaps does it fill?]

---

## 2. Methodology

### 2.1 Protocol and Registration

**Protocol**: [PROSPERO ID / OSF link / Not registered]  
**Deviations**: [Document any protocol deviations]  
**PRISMA**: [Checklist in Appendix B]

### 2.2 Search Strategy

**Databases:** [PubMed, Scopus, Web of Science, bioRxiv, etc.]  
**Supplementary:** [Citation chaining, grey literature, trial registries]

**Search String Example:**
```
("CRISPR"[Title/Abstract] OR "Cas9"[Title/Abstract]) AND
("disease"[MeSH Terms]) AND ("2015/01/01"[Publication Date] : "2024/12/31"[Publication Date])
```

**Dates:** [YYYY-MM-DD to YYYY-MM-DD] | **Executed:** [Date]  
**Validation:** [Key papers used to test search strategy]

### 2.3 Tools and Software

**Screening:** [Rayyan, Covidence, ASReview]  
**Analysis:** [VOSviewer, R, Python]  
**Citation Management:** [Zotero, Mendeley, EndNote]  
**AI Tools:** [Any AI-assisted tools used; document validation approach]

### 2.4 Inclusion and Exclusion Criteria

**Inclusion Criteria:**
- [Criterion 1: e.g., Published between 2015-2024]
- [Criterion 2: e.g., Peer-reviewed articles and preprints]
- [Criterion 3: e.g., English language]
- [Criterion 4: e.g., Human or animal studies]
- [Criterion 5: e.g., Original research or systematic reviews]

**Exclusion Criteria:**
- [Criterion 1: e.g., Case reports with n<5]
- [Criterion 2: e.g., Conference abstracts without full text]
- [Criterion 3: e.g., Editorials and commentaries]
- [Criterion 4: e.g., Duplicate publications]
- [Criterion 5: e.g., Retracted articles]
- [Criterion 6: e.g., Studies with unavailable full text after author contact]

### 2.5 Study Selection

**Reviewers:** [n independent reviewers] | **Conflict resolution:** [Method]  
**Inter-rater reliability:** [Cohen's kappa = X]

**PRISMA Flow:**
```
Records identified (by source): n=[X]
Removed before screening (duplicates/other reasons): n=[D]
Records screened: n=[X-D]; records excluded: n=[E]
Reports sought: n=[S]; reports not retrieved: n=[U]
Reports assessed: n=[S-U]; reports excluded with reasons: n=[F]
Reports included: n=[S-U-F]; studies included: n=[N from study-report map]
```

**Exclusion reasons:** [List with counts; distinguish reports not retrieved]
**Study-report map:** [Study ID -> all reports, versions, trial IDs, and outcome sources]

### 2.6 Data Extraction

**Method:** [Standardized form (Appendix E); pilot-tested on n studies]  
**Extractors:** [n independent] | **Verification:** [Double-checked]

**Items:** Study ID, design, population, interventions/exposures, outcomes, statistics, funding, COI, bias domains

**Missing data:** [Author contact protocol]

### 2.7 Quality Assessment

**Tool:** [Cochrane RoB 2 / ROBINS-I / Newcastle-Ottawa / AMSTAR 2 / JBI]\
**Method:** [2 independent reviewers; third for conflicts]  
**Rating:** [Use tool-specific domain and overall judgments; RoB 2: low/some concerns/high]\
**Publication bias:** [Funnel plots, Egger's test - if meta-analysis]

### 2.8 Synthesis and Analysis

**Approach:** [Narrative / Meta-analysis / Both]  
**Statistics** (if meta-analysis): Effect measures, heterogeneity (I², τ²), sensitivity analyses, subgroups  
**Software:** [RevMan, R, Stata]  
**Certainty:** [GRADE per outcome/body of evidence; risk of bias, inconsistency, indirectness, imprecision, publication bias]\

---

## 3. Results

### 3.1 Study Selection

**Summary:** [Records screened; reports sought/not retrieved/assessed/excluded/included; N distinct studies (M in meta-analysis)]\
**Study types:** [RCTs: n=X, Observational: n=Y, Reviews: n=Z]  
**Years:** [Range; peak year]  
**Geography:** [Countries represented]  
**Source:** [Peer-reviewed: n=X, Preprints: n=Y]

### 3.2 Bibliometric Overview

[Optional: Trends, journal distribution, author networks, citations, keywords - if analyzed with VOSviewer or similar]

### 3.3 Study Characteristics

| Study | Year | Design | Sample Size | Key Methods | Main Findings | Quality |
|-------|------|--------|-------------|-------------|---------------|---------|
| First Author et al. | 2023 | [Type] | n=[X] | [Methods] | [Brief findings] | [Tool-specific judgment] |

**Risk of bias:** [Report tool-specific outcome/domain judgments and denominators]

### 3.4 Thematic Synthesis

[Organize by themes, NOT study-by-study. Synthesize across studies to identify consensus, controversies, and gaps.]

#### 3.4.1 Theme 1: [Title]

**Findings:** [Synthesis of key findings from multiple studies]  
**Supporting studies:** [X, Y, Z]  
**Contradictory evidence:** [If any]  
**Certainty:** [GRADE rating if applicable]

### 3.5 Methodological Approaches

**Common methods:** [Method 1 (n studies), Method 2 (n studies)]  
**Emerging techniques:** [New approaches observed]  
**Methodological quality:** [Overall assessment]

### 3.6 Meta-Analysis Results

[Include only if conducting meta-analysis]

**Effect estimates:** [Primary/secondary outcomes with 95% CI, p-values]  
**Heterogeneity:** [I²=X%, τ²=Y, interpretation]  
**Subgroups & sensitivity:** [Key findings from analyses]  
**Publication bias:** [Funnel plot, Egger's p=X]  
**Forest plots:** [Include for primary outcomes]

### 3.7 Knowledge Gaps

**Knowledge:** [Unanswered research questions]  
**Methodological:** [Study design/measurement issues]  
**Translational:** [Research-to-practice gaps]  
**Populations:** [Underrepresented groups/contexts]

---

## 4. Discussion

### 4.1 Main Findings

[Synthesize key findings by research question]

**Principal findings:** [Top 3-5 takeaways]  
**Consensus:** [Where studies agree]  
**Controversy:** [Conflicting results]

### 4.2 Interpretation and Implications

**Context:** [How findings advance/challenge current understanding]  
**Mechanisms:** [Potential explanations for observed patterns]

**Implications for:**
- **Practice:** [Actionable recommendations]
- **Policy:** [If relevant]
- **Research:** [Theoretical, methodological, priority directions]

### 4.3 Strengths and Limitations

**Strengths:** [Comprehensive search, rigorous methods, large evidence base, transparency]

**Limitations:**
- Search/selection: [Language bias, database coverage, grey literature, publication bias]
- Methodological: [Heterogeneity, study quality]
- Temporal: [Rapid evolution, search cutoff date]

**Impact:** [How limitations affect conclusions]

### 4.4 Comparison with Previous Reviews

[If relevant: How does this review update/differ from prior reviews?]

### 4.5 Future Research

**Priority questions:**
1. [Question] - Rationale, suggested approach, expected impact
2. [Question] - Rationale, suggested approach, expected impact
3. [Question] - Rationale, suggested approach, expected impact

**Recommendations:** [Methodological improvements, understudied populations, emerging technologies]

---

## 5. Conclusions

[Concise conclusions addressing research questions]

1. [Conclusion directly addressing primary research question]
2. [Key finding conclusion]
3. [Gap/future direction conclusion]

**Evidence certainty:** [High/Moderate/Low/Very Low]  
**Translation readiness:** [Ready / Needs more research / Preliminary]

---

## 6. Declarations

### Author Contributions
[CRediT taxonomy: Author 1 - Conceptualization, Methodology, Writing; Author 2 - Analysis, Review; etc.]

### Funding
[Grant details with numbers] OR [No funding received]

### Conflicts of Interest
[Author-specific declarations] OR [None]

### Data Availability
**Protocol:** [PROSPERO/OSF ID or "Not registered"]  
**Data/Code:** [Repository URL/DOI or "Available upon request"]  
**Materials:** [Search strategies (Appendix A), PRISMA checklist (Appendix B), extraction form (Appendix E)]

### Acknowledgments
[Contributors not meeting authorship criteria, librarians, patient involvement]

---

## 7. References

[Use consistent style: APA / Nature / Vancouver]

**Format examples:**

APA: Author, A. A., & Author, B. B. (Year). Title. *Journal*, *volume*(issue), pages. https://doi.org/xx.xxxx

Nature: Author, A. A. & Author, B. B. Title. *J. Name* **volume**, pages (year).

Vancouver: Author AA, Author BB. Title. J Abbrev. Year;volume(issue):pages. doi:xx.xxxx

1. [First reference]
2. [Second reference]
3. [Continue...]

---

## 8. Appendices

### Appendix A: Search Strings

**PubMed** (Date: YYYY-MM-DD; Results: n)
```
[Complete search string with operators and MeSH terms]
```

[Repeat for each database: Scopus, Web of Science, bioRxiv, etc.]

### Appendix B: PRISMA Checklist

[Summary below is not the complete checklist. Attach the official PRISMA 2020 checklist, including all applicable items/subitems, or the appropriate extension.]

| Section | Item | Reported? | Page |
|---------|------|-----------|------|
| Title | Identify as systematic review | Yes/No | # |
| Abstract | Structured summary | Yes/No | # |
| Methods | Eligibility, sources, search, selection, data, quality | Yes/No | # |
| Results | Selection, characteristics, risk of bias, syntheses | Yes/No | # |
| Discussion | Interpretation, limitations, conclusions | Yes/No | # |
| Other | Registration, support, conflicts, availability | Yes/No | # |

### Appendix C: Excluded Studies

| Study | Year | Reason | Category |
|-------|------|--------|----------|
| Author et al. | Year | [Reason] | [Wrong population/outcome/design/etc.] |

**Summary:** Wrong population (n=X), Wrong outcome (n=Y), etc.

### Appendix D: Quality Assessment

**Tool:** [Cochrane RoB 2 / ROBINS-I / Newcastle-Ottawa / etc.]\

| Study | Domain 1 | Domain 2 | Domain 3 | Overall |
|-------|----------|----------|----------|---------|
| Study 1 | Low | Low | Some concerns | Some concerns |
| Study 2 | [Score] | [Score] | [Score] | [Overall] |

### Appendix E: Data Extraction Form

```
STUDY: Author______ Year______ DOI______
DESIGN: □RCT □Cohort □Case-Control □Cross-sectional □Other______
POPULATION: n=_____ Age_____ Setting_____
INTERVENTION/EXPOSURE: _____
OUTCOMES: Primary_____ Secondary_____
RESULTS: Effect size_____ 95%CI_____ p=_____
RISK OF BIAS: Tool______ Outcome______ Judgment______ Rationale______
FUNDING/COI: _____
```

### Appendix F: Meta-Analysis Details

[Only if meta-analysis performed]

**Software:** [R 4.x.x with meta/metafor packages / RevMan / Stata]  
**Model:** [Random-effects; justification]  
**Code:** [Link to repository]  
**Sensitivity analyses:** [Details]

### Appendix G: Author Contacts

| Study | Contact Date | Response | Data Received |
|-------|--------------|----------|---------------|
| Author et al. | YYYY-MM-DD | Yes/No | Yes/No/Partial |

---

## 9. Supplementary Materials

[If applicable]

**Tables:** S1 (Full study characteristics), S2 (Quality scores), S3 (Subgroups), S4 (Sensitivity)  
**Figures:** S1 (PRISMA diagram), S2 (Risk of bias), S3 (Funnel plot), S4 (Forest plots), S5 (Networks)  
**Data:** S1 (Extraction file), S2 (Search results), S3 (Analysis code), S4 (PRISMA checklist)  
**Repository:** [OSF/GitHub/Zenodo URL with DOI]

---

## Review Metadata

**Registration:** [Registry] ID: [Number] (Date: YYYY-MM-DD)  
**Search dates:** Initial: [Date]; Updated: [Date]  
**Version:** [1.0] | **Last updated:** [Date]

**Quality checks:**
- [ ] DOI registration checked; identity, claim support, and correction/retraction status manually reviewed
- [ ] PRISMA checklist completed
- [ ] Search reproducible
- [ ] Independent data verification
- [ ] Code peer-reviewed
- [ ] All authors approved

---

## Usage Notes

**Review type adaptations:**
- Systematic Review: Use all sections
- Meta-Analysis: Include sections 3.6, Appendix F
- Narrative Review: May omit some methodology detail
- Scoping Review: Follow PRISMA-ScR, may omit quality assessment

**Key principles:**
1. Remove all [bracketed placeholders]
2. Follow PRISMA 2020 guidelines
3. Pre-register when feasible (PROSPERO/OSF)
4. Use thematic synthesis, not study-by-study
5. Be transparent and reproducible
6. Verify reference identity and claim support, including items without DOIs
7. Make data/code openly available

**Common pitfalls to avoid:**
- Don't list studies - synthesize them
- Don't cherry-pick results
- Don't ignore limitations
- Don't overstate conclusions
- Don't skip publication bias assessment

**Resources:**
- PRISMA 2020: https://www.prisma-statement.org/
- PROSPERO: https://www.crd.york.ac.uk/prospero/
- Cochrane Handbook: https://training.cochrane.org/handbook
- GRADE: https://www.gradeworkinggroup.org/

**DELETE THIS SECTION FROM YOUR FINAL REVIEW**

---
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 323kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 323kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 323kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 198kUpdated Claude CodeCodexVerifiedInstall