Fact-checking
AuraScore 81/100

Empirical Contradiction Cross-Examination Script

Author a structured forensic interview and cross-examination script to evaluate conflicting empirical findings across complex research syntheses.

Use this prompt when preparing investigative scientific panels, peer review hearings, or factual arbitrations between conflicting research teams. It creates a timed, staged cross-examination script targeting empirical discrepancies.

Template

Role: Principal Forensic Research Synthesist and Scientific Cross-Examiner.

Context

  • Target Research Domain: {{research_domain}}
  • Contradictory Literature Corpus: {{primary_paper_corpus}}
  • Core Contested Findings: {{contested_findings}}
  • Governing Epistemic Criteria: {{epistemic_criteria}}
  • Presiding Panel Chair: {{panel_chair_name}}
  • Investigation Rigor Level: {{audit_depth_level}}

Task

Construct a comprehensive, timed forensic cross-examination script for {{panel_chair_name}} to systematically interrogate conflicting empirical claims, detect statistical anomalies, and establish factual ground truth within {{research_domain}}.

Method

  1. Dissect {{primary_paper_corpus}} into competing evidentiary matrices, pairing conflicting causal assertions against each other.
  2. Map {{contested_findings}} directly to specific methodological failure modes, such as p-hacking, selection bias, or unadjusted confounding variables.
  3. Structure sequential interrogation phases ranging from baseline methodology verification to intense factual cross-examination.
  4. Formulate precise, closed-ended questions targeting sample provenance, data exclusion criteria, and mathematical transformation anomalies.
  5. Design contingency interrogation branches anticipating evasive, ambiguous, or mathematically implausible responses from audited researchers.
  6. Incorporate evidentiary exhibit cues referencing contradictory tables, confidence intervals, and methodological appendices.
  7. Synthesize expected evidentiary outcomes into a structured post-session verdict protocol based on {{epistemic_criteria}}.

Constraints

  • MUST structure script strictly with speaker tags, stage directions, timed blocks, and explicit exhibit citations.
  • MUST align all interrogation lines to the standards defined in {{audit_depth_level}}.
  • MUST NOT permit leading questions without corroborating baseline evidence cited in {{primary_paper_corpus}}.
  • Every contradictory finding MUST have at least three targeted follow-up questions exposing potential empirical divergence.

Output format

  1. Session Overview & Epistemic Framework: Objective, panel rules, and burden of proof standards.
  2. Interrogation Timeline & Exhibit Index: Timed phases, evidence numbers, and reference tables.
  3. Staged Examination Script: Full verbatim dialogue including Chair lines, probe branches, and exhibit presentations (minimum 4 distinct phases).
  4. Admissibility & Verdict Scorecard: Rubric evaluating claim validity, replicability, and credibility.

Self-review

  • Confirm all 6 context variables are present and contextualized in the script text.
  • Verify that cross-examination questions directly challenge the specific claims in {{contested_findings}}.
  • Check that branch protocols account for both affirmative admissions and evasive answers.
  • Ensure script dialogue maintains an objective, forensic, and scientifically rigorous tone.
AuraScore breakdown
81/100Provisional
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 engineering12/12 · Strong

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
fact-checking
cross-examination
research-synthesis