Synthesize a Research Corpus into a Traceable Literature Review

Required inputs

Have these details available before following the usage instructions.

• Research question, review purpose, target audience, and intended use of the synthesis.
• Source corpus or excerpts with enough citation details to identify each source.
• Any inclusion scope, time range, domain boundaries, priority constructs, populations, interventions, outcomes, or comparison criteria.
• Preferred synthesis format, such as narrative review, thematic synthesis, evidence matrix, executive memo, or gap analysis.
• Known methodological standards, appraisal criteria, citation style, and any sources that must be treated cautiously or excluded.

How to use this Skill

When to use:
- A researcher, analyst, student, or team has already gathered a defined set of sources and needs a structured synthesis.
- A review must compare themes, findings, methods, populations, measures, contexts, or contradictions across studies.
- A decision memo, dissertation chapter, policy brief, or research report needs transparent evidence and uncertainty rather than a simple summary.
- The user needs an evidence matrix or synthesis draft that can be reviewed by a supervisor, editor, methodologist, or subject-matter expert.

When not to use:
- Do not use when the task is only to find new sources; use a source-discovery or source-triage capability first.
- Do not use for systematic-review screening protocol design unless eligibility and extraction workflow are the main task.
- Do not use to fabricate missing study details, infer unavailable methods, or cite sources not supplied or verifiably identified.
- Do not use as final academic, clinical, legal, or policy authority without qualified human review.

Instructions:
1. Open AMO-P-000041, supply its required inputs, and run the complete source Prompt in Claude. Review its evidence register and limitations before using the result.
2. Create a source inventory and evidence matrix that separates source-reported findings, methods, sample/context details, limitations, and reviewer interpretation.
3. Group findings into themes only when supported by multiple sources or clearly identify single-source observations.
4. Compare agreements, contradictions, heterogeneity, study quality, bias risks, and contextual limits before drawing synthesis-level conclusions.
5. Mark supplied facts, assumptions, inference, missing information, and uncertainty distinctly, especially for causal or generalizable claims.
6. Avoid citation laundering: do not let a review article stand in for primary evidence when the original source is required but unavailable.
7. Produce conclusions with confidence levels or evidence strength labels tied to the supplied corpus, and list unresolved questions or evidence gaps for follow-up review.

Expected output:
A literature synthesis package including source inventory, evidence matrix, thematic synthesis, areas of agreement and conflict, source-quality and limitation notes, uncertainty register, research gaps, and draft review text or memo sections as requested.

Constraints and boundaries:
- Use only supplied or explicitly verified source information; do not invent bibliographic details, sample sizes, methods, findings, or quotations.
- Require evidence for factual and causal claims, and identify when evidence is indirect, weak, outdated, conflicting, or outside scope.
- Keep interpretive recommendations separate from authorization or policy decisions.
- Require human review before publication, submission, clinical use, legal use, policy adoption, or other consequential reliance.
- Apply the supplied appraisal method, or obtain confirmation of an appropriate method when it affects conclusions. Do not invent confidence scores. Preserve abstract-only versus full-text access, link companion reports to avoid double-counting studies, retain opposing/null findings, and do not imply searches or source-status checks occurred without evidence.

Based on an Amo.ng Prompt

Literature Review Synthesis Prompt

Open the linked Prompt to read the instructions this Skill is based on.

Completion criteria

Complete when:
- Every material claim in the synthesis is traceable to one or more listed sources or is clearly labeled as interpretation or uncertainty.
- Contradictory evidence and methodological limitations are visible, not smoothed into a single unsupported conclusion.
- The evidence matrix includes enough study-level detail to audit how themes were derived.
- Unsupported gaps, missing full texts, inaccessible data, and uncertain citations are explicitly flagged.
- The final synthesis matches the stated research question and does not overgeneralize beyond the supplied corpus.