Fact-checking
AuraScore 79/100

Meta-Analytic Literature Synthesis Fact-Check Brief

Validate causal assertions, effect size aggregations, and citation integrity in research syntheses.

Use this template when validating complex meta-analyses, evidence syntheses, or multi-study literature reviews. It identifies citation bias, effect size inflation, and unwarranted causal extrapolation across scientific corpora.

Template

Role: Senior Meta-Science Auditor specializing in evidence synthesis and scientific reproducibility.

Context

  • Synthesis document under review: {{synthesis_document}}
  • Focal causal relationships: {{focal_causal_claims}}
  • Source publication corpus: {{primary_citations_corpus}}
  • Systematic inclusion/exclusion rules: {{inclusion_criteria}}
  • Declared author affiliations: {{conflict_of_interest_records}}
  • Aggregate statistical parameters: {{meta_analysis_metrics}}

Task

Construct a meta-analytic fact-checking brief that assesses whether {{synthesis_document}} accurately synthesizes the empirical evidence from {{primary_citations_corpus}}, uncovering citation misattributions, selective pooling, and unsupported causal generalizations.

Method

  1. Map every assertion in {{focal_causal_claims}} to its underlying citations within {{primary_citations_corpus}}.
  2. Review individual primary studies to determine if cited findings are reported accurately or taken out of context.
  3. Verify that the synthesized studies strictly satisfy {{inclusion_criteria}} and that non-qualifying papers were omitted.
  4. Recalculate combined effect sizes, heterogeneity scores (I²), and publication bias metrics from {{meta_analysis_metrics}}.
  5. Detect instances of citation distortion, such as transforming correlational evidence into affirmative causal claims.
  6. Evaluate whether study inclusion aligns with potential biases detailed in {{conflict_of_interest_records}}.
  7. Assess funnel plot asymmetry and file-drawer effects that could distort the aggregated findings.
  8. Rate the synthesis on established scientific grading systems (e.g., GRADE criteria) to produce actionable verdicts.

Constraints

  • MUST NOT accept secondary summaries; every citation check MUST reference primary study methodologies.
  • MUST explicitly differentiate between statistical correlation and proven causal mechanisms.
  • Highlight any discrepancy greater than 5% in pooled effect size calculations.
  • Confine evaluations to peer-reviewed evidentiary standards.

Output format

  1. Evidence Synthesis Authenticity Scorecard (Table with Metrics, Expected, Calculated, Divergence)
  2. Citation-by-Citation Verification Log (Columns: Citation, Claim in Review, True Finding, Distortion Risk, Verdict)
  3. Causal Inference Validity Assessment (max 250 words)
  4. Synthesis Vulnerability & Bias Summary (max 200 words)

Self-review

  • Did I verify primary sources directly rather than relying on abstract summaries?
  • Are recalculations for all pooled metrics in {{meta_analysis_metrics}} fully documented?
  • Have all causal extrapolations in {{focal_causal_claims}} been challenged against the underlying study designs?
AuraScore breakdown
79/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering10/12 · Adequate

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness3/5 · Adequate

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

research-analysis
research-fact-checking
complex-reasoning-analysis-math
meta-science
evidence-synthesis
fact-checking