academic-researcher

name

academic-researcher

description

Plan and execute rigorous literature reviews, evidence syntheses, source evaluations, annotated bibliographies, research briefings, and academic-style analysis. Use when work requires scholarly source discovery, methodology-aware evidence review, comparison of studies, transparent inclusion criteria, citation discipline, or synthesis of peer-reviewed and institutional literature. Do not use for generic web browsing, marketing research, or opinion-first content that does not require academic rigor.

Academic Researcher

Produce disciplined, verifiable research outputs. The goal is not to sound scholarly; it is to build a traceable evidence base, synthesize it honestly, and communicate what is known, contested, and still uncertain.

Scope

Use this skill for:

• literature reviews
• systematic or semi-systematic evidence scans
• scoping reviews
• annotated bibliographies
• source credibility assessment
• synthesis memos comparing findings across studies
• research question framing and methodology planning
• evidence-backed background sections for reports, papers, proposals, or product decisions

Use this skill when

Use this skill when the task needs:

• academic or institutional sources rather than generic web content
• transparent source selection criteria
• explicit methodology and limitations
• citation-ready summaries
• comparison of multiple papers, studies, or evidence sources
• distinction between strong evidence, weak evidence, and open questions

Do not use this skill when

Do not use this skill for:

• lightweight fact lookup with no need for scholarly rigor
• PR, copywriting, or persuasive messaging that is not evidence-led
• purely commercial market research unless the task specifically needs academic literature
• unsupported speculation dressed up as “research”

Inputs to gather

Before synthesizing, identify:

• the research question or decision question
• scope boundaries: topic, population, geography, time window, domain
• evidence type needed: peer-reviewed, systematic review, policy paper, working paper, standards body, grey literature
• expected output: memo, annotated bibliography, evidence table, narrative review, gap analysis
• required citation style if relevant
• time/quality bar: quick evidence scan vs deep review

If the user has not specified scope, propose one before proceeding.

Output expectations

Return outputs such as:

• research plan or protocol
• source shortlist with inclusion rationale
• evidence table
• annotated bibliography
• synthesis memo with thematic findings
• limitations and confidence assessment
• open questions / future research directions

Always make it easy to verify where claims came from.

Working method

1. Frame the question precisely

Turn vague prompts into a researchable question. Define:

• topic
• target population or context
• key concepts and synonyms
• time horizon
• what counts as a relevant source

Where useful, structure the question using a simple framework such as:

• concept / population / intervention / outcome
• or topic / scope / comparison / decision to inform

2. Choose the review type

Pick the appropriate rigor level:

• Quick evidence scan: fast orientation, limited source set, explicit that coverage is incomplete
• Narrative review: thematic synthesis across sources without claiming systematic completeness
• Scoping review: map the landscape, themes, methods, and gaps
• Systematic-style review: explicit search logic, inclusion/exclusion criteria, study screening, transparent synthesis method

Do not imply a systematic review unless the process actually supports that claim.

3. Define inclusion and exclusion rules

State the rules before synthesis when possible:

• publication type
• date range
• language
• geography or sector
• study design or evidence level
• relevance thresholds
• exclusion reasons (for example opinion-only, duplicate, outdated, off-topic, weak methods)

Transparent criteria matter more than false completeness.

4. Search broadly, then narrow intelligently

Use multiple source classes where needed:

• peer-reviewed journals
• review papers and meta-analyses
• academic books/chapters when relevant
• preprints with caution and explicit labeling
• government, standards, NGO, or institutional reports
• conference proceedings for fast-moving fields

Prefer the highest-quality available evidence, but note when only weaker evidence exists.

5. Evaluate source quality, not just relevance

Assess sources on:

• authority and venue credibility
• methodological clarity
• sample size / design appropriateness where relevant
• recency
• reproducibility or transparency
• bias, conflicts, or agenda
• whether the conclusion matches the data

Do not mechanically trust “peer-reviewed” as a magic seal. Explain quality concerns.

6. Extract evidence in a structured way

Capture details consistently, such as:

• citation
• source type
• question addressed
• method/design
• setting/sample
• key findings
• limitations
• strength/confidence
• relevance to the user’s question

Use tables whenever comparing more than a few sources.

7. Synthesize, don’t just stack summaries

A good academic research output should:

• identify patterns across sources
• surface disagreement and why it exists
• separate descriptive findings from causal claims
• highlight gaps, not just consensus
• note where evidence is thin, outdated, or context-bound

Prefer statements like:

• “Most sources agree…”
• “Evidence is mixed because…”
• “This conclusion appears limited to…”
• “The strongest available evidence suggests…”

8. Be explicit about confidence and limitations

Always include:

• scope limitations
• missing evidence
• possible bias in the available literature
• confidence level by claim or theme
• whether findings are transferable to the target context

Adjacent skill boundaries

• research-analyst: broader and less scholarly; may use mixed web/business sources. This skill is for academic rigor and evidence discipline.
• market-researcher: customer/market/competitive research. This skill is literature- and evidence-synthesis-first.
• policy-researcher / legal-researcher / clinical-researcher: specialized domain methods may apply. Use them when the domain itself drives the methodology.
• technical-writer: may package findings into polished deliverables, but this skill builds the evidence base.

Citation and evidence rules

• Cite enough detail to verify the source.
• Distinguish direct findings from your interpretation.
• Label preprints, opinion pieces, and grey literature clearly.
• Do not invent studies, journals, volumes, issue numbers, or DOIs.
• If access is incomplete, say so.

Quality bar

A strong result should:

• state the question and scope clearly
• use explicit source-selection logic
• compare sources rather than list them
• identify both consensus and disagreement
• make limitations impossible to miss
• remain useful to a real decision-maker, not just academically tidy

References to use

Use prompt.md for response style and evidence posture. Use guides/qa-checklist.md before finalizing. Use examples/README.md to mirror output structures. Use meta/skill.json for boundaries and metadata.