Use this skill when conducting a literature review or surveying academic research on a topic. Trigger phrases: 'do a literature review', 'survey the research on', 'what does the literature say about', 'review academic papers on'. Do NOT use for primary research data collection, statistical data analysis, or generating new original knowledge.
This skill guides you through synthesizing existing academic and scholarly research into a coherent literature review. A good literature review doesn't just summarize sources one by one—it identifies themes, debates, gaps, and trajectories across a body of work. Whether you need a two-page background section for a thesis or a standalone systematic review, this skill provides the structure to move from a pile of sources to a compelling scholarly narrative that demonstrates command of a field.
fact-checker skill instead)summarizer skill instead)citation-formatter skill instead)| Task | Approach |
|---|---|
| Scope | Define a clear research question before searching; broad topics need narrowing |
| Organization | Thematic > chronological > methodological—choose one primary structure |
| Synthesis | Compare and contrast sources; avoid back-to-back summaries |
| Gaps | Explicitly name what the literature has NOT addressed |
| Transitions | Use signpost phrases to guide the reader between themes |
| Voice | Maintain academic objectivity; attribute claims to authors |
| Length | Background section: 500–1,500 words; standalone review: 3,000–8,000 words |
| Sources | Prioritize peer-reviewed journals; note when grey literature is used |
Define the scope and research question. Before reviewing anything, articulate the precise question the review will answer. A vague topic like "climate change" yields an unmanageable review; a focused question like "How do carbon pricing mechanisms affect household energy consumption in OECD countries since 2010?" is workable. Set inclusion/exclusion criteria: time range, geographic focus, study types (empirical, theoretical, meta-analyses), and disciplines.
Identify and organize your sources. List all sources to be reviewed. Group them loosely into categories based on theme, method, or argument before writing anything. Tools like a simple spreadsheet work well: columns for author, year, key findings, methods, and how each source relates to your research question. This prevents the most common literature review failure—a laundry list of summaries.
Identify major themes, debates, and trajectories. Read across your sources and ask: What do most scholars agree on? Where do they disagree, and why? Has the field shifted over time? Are there methodological divides? Document these patterns—they become the backbone of your thematic sections.
Choose an organizational structure. Most literature reviews use one of three structures:
Write the introduction to the review. Open by establishing why this topic matters and what the review will cover. State the organizing framework ("This review is organized thematically around three core debates…"). Define key terms if the field uses contested terminology.
Write each thematic section using synthesis, not summary. Each section should open with the theme's key claim or debate, then draw on multiple sources together. Instead of: "Smith (2019) found X. Jones (2021) found Y," write: "Studies consistently find X (Smith, 2019; Jones, 2021; Patel, 2022), though the magnitude varies by context (Lee, 2020)." Attribute claims and highlight where sources agree, disagree, or build on each other.
Identify and articulate the research gap. The literature review's payoff is naming what the field has not addressed—the gap your research (or the reader's) fills. Make this explicit: "Despite extensive work on X, little attention has been paid to Y in the context of Z."
Write the conclusion. Summarize the major themes without simply restating every point. Restate the gap and explain how it motivates the work that follows (or what future research should address). If this is a standalone review, offer a synthesis argument—what does the totality of the evidence suggest?
Review for synthesis, not summary. Do a final pass asking: Does every paragraph discuss multiple sources in relation to each other? Are there any paragraphs that summarize only one paper in isolation? Revise those to integrate them into broader themes.
Input: "Write a literature review background section for a proposal studying the effect of mindfulness-based stress reduction (MBSR) on burnout in healthcare workers. I have about 12 sources spanning 2015–2023."
Output:
Healthcare worker burnout has reached crisis levels, with surveys consistently reporting that 40–60% of physicians and nurses experience significant burnout symptoms (Panagioti et al., 2018; Shanafelt et al., 2022). Characterized by emotional exhaustion, depersonalization, and reduced personal accomplishment (Maslach & Leiter, 2016), burnout carries serious consequences for both providers and patients, including increased medical errors, higher turnover, and poorer patient satisfaction (Hall et al., 2016).
The Case for Mindfulness-Based Interventions
Mindfulness-based stress reduction (MBSR), originally developed by Kabat-Zinn (1990), has emerged as a leading candidate for organizational wellness interventions. Meta-analytic evidence supports its efficacy in reducing general stress and anxiety across populations (Khoury et al., 2015). In healthcare settings specifically, randomized controlled trials have demonstrated meaningful reductions in self-reported emotional exhaustion following 8-week MBSR programs (Goodman & Schorling, 2012; West et al., 2016). A systematic review by Lomas et al. (2019) found moderate effect sizes (d = 0.53) for mindfulness interventions on burnout subscales across 23 studies, though the authors noted significant heterogeneity in both intervention design and outcome measurement.
Contested Findings and Methodological Limitations
Despite this promise, the literature reveals important debates. Several studies report that effects are largest immediately post-intervention but attenuate at 6- and 12-month follow-ups, raising questions about sustainability (Asuero et al., 2014; Fortney et al., 2013). Others argue that individual-level interventions inadequately address the systemic causes of burnout—workload, staffing, and institutional culture—and may inadvertently shift responsibility to workers rather than organizations (Panagioti et al., 2018; Shanafelt & Noseworthy, 2017). Methodologically, most studies rely on self-report measures, lack active control conditions, and draw from volunteer samples, limiting causal inference (West et al., 2016).
Remaining Gaps
Three gaps motivate the proposed study. First, most existing trials enroll physicians almost exclusively; nurses and allied health workers, who face distinct stressors and have different workplace flexibility, are underrepresented (Lomas et al., 2019). Second, few studies examine the dose-response relationship—whether abbreviated programs (4-week or digital adaptations) yield comparable benefits. Third, no study to date has examined whether pre-existing trait mindfulness moderates treatment response, which would allow for more targeted intervention delivery. The proposed study addresses all three gaps through a multi-professional sample, a 4- vs. 8-week design comparison, and baseline mindfulness assessment.
Input: "I need a short literature review on whether social media use is associated with adolescent depression. Organize it thematically. Cover both sides of the debate."
Output:
The question of whether social media harms adolescent mental health has generated one of the most active—and contentious—debates in developmental psychology over the past decade. While popular discourse often treats the link as established fact, the scholarly literature presents a far more ambiguous picture, divided along lines of methodology, measurement, and theoretical framing.
Evidence Supporting an Association
A subset of studies reports statistically significant associations between heavy social media use and elevated depressive symptoms in adolescents, particularly among girls (Twenge et al., 2018; Kelly et al., 2019). Twenge and colleagues' large-scale analyses of U.S. survey data found that adolescents spending five or more hours daily on social media were 66% more likely to have at least one suicide risk factor compared to those with no screen time, prompting widespread alarm. Longitudinal work by Coyne et al. (2020) identified modest but significant predictive relationships between social media use and depressive symptoms over 6-year follow-ups in some subgroups. Proposed mechanisms include social comparison, cyberbullying exposure, sleep displacement, and passive consumption of curated content (Vannucci et al., 2017).
Evidence Against a Causal Relationship
However, a substantial body of research challenges these findings on both empirical and theoretical grounds. Orben and Przybylski's (2019) landmark reanalysis—using specification curve analysis across three large datasets—found that social media's association with well-being was comparable in magnitude to wearing glasses or eating potatoes, and accounted for less than 0.5% of variance in well-being outcomes. They argue that prior alarming findings reflect researcher degrees of freedom rather than robust effects. Several preregistered, longitudinal studies find no significant prospective association between social media use and depression after controlling for baseline mental health (Coyne et al., 2020; Vuorre et al., 2021). Critics further note that existing work conflates passive scrolling, active communication, and content creation—behaviors with potentially opposite effects.
Moderators and the Complexity of "Use"
Emerging consensus holds that the association, if any, is highly conditional. Platform type, content consumed, and social context appear to matter more than raw time spent. Fardouly and colleagues (2018) find that appearance-focused Instagram use predicts body dissatisfaction, while social communication use does not. Qualitative research with adolescents consistently reveals that they distinguish between social media as a source of belonging versus a source of comparison and exclusion (Pater & Mynatt, 2017). Gender emerges as a robust moderator, with girls showing stronger negative associations in most studies, possibly reflecting heightened social comparison vulnerability.
Synthesis and Remaining Questions
The literature does not currently support a simple claim that social media causes depression in adolescents. What it does suggest is that specific patterns of use—particularly passive, appearance-focused consumption among girls—may be associated with elevated risk, while other forms of use may be neutral or even protective. Future research must move beyond screen time as a monolithic variable and adopt experience-sampling methods that capture real-time emotional states alongside behavioral data.