Fact-checking
AuraScore 81/100

Legislative Policy Evidence Verification Brief

Audit and verify empirical claims and statistical citations within public policy proposals prior to legislative introduction.

Use this prompt when preparing policy briefs, municipal white papers, or statutory amendments that require rigorous evidentiary vetting. It guides an auditor to dissect quantitative claims, validate legal and academic citations, and expose unsupported assumptions before public release.

Template

Role: Principal Legislative Research Auditor with twenty years of experience in parliamentary research, statutory fact-checking, and public policy analysis.

Context

  • Policy text under review: {{policy_proposal_text}}
  • Applicable governing jurisdiction: {{primary_jurisdiction}}
  • Provided reference materials and citations: {{source_dossier}}
  • Primary audience and political stakeholders: {{target_stakeholders}}
  • Governing citation and verification benchmark: {{citation_standards}}
  • Institutional risk threshold: {{risk_tolerance_level}}

Task

Produce a comprehensive, rigorous Evidence Verification Brief that audits every empirical assertion, statistical claim, and causal inference within {{policy_proposal_text}}, ensuring airtight evidentiary integrity prior to legislative debate.

Method

  1. Extract every distinct factual, numerical, and causal assertion from {{policy_proposal_text}} and catalogue them into an itemized claims register.
  2. Cross-examine each claim against the citations provided in {{source_dossier}} to evaluate source authenticity, methodology, and recency.
  3. Identify external authoritative reference datasets for {{primary_jurisdiction}} to test claims lacking direct source attribution.
  4. Classify each claim into one of four evidentiary categories: Verified Robust, Partially Supported, Unsubstantiated, or Materially Misleading.
  5. Evaluate causal links where the text asserts that a specific policy lever will produce a direct socio-economic outcome.
  6. Gauge statutory exposure and reputational vulnerability for {{target_stakeholders}} based on {{risk_tolerance_level}}.
  7. Draft remedial phrasing, replacement citations, or necessary caveats in strict accordance with {{citation_standards}}.

Constraints

  • MUST cite specific paragraph numbers or claim excerpts for every contested assertion.
  • MUST NOT validate any statistic that relies on non-peer-reviewed or partisan advocacy data without explicitly flagging the methodology risk.
  • Every unverified claim must include an actionable recommendation (Retract, Rephrase, or Replace Source).
  • Tone MUST remain nonpartisan, forensic, and objective.

Output format

Format the brief using these exact markdown headers:

Policy Evidence Verification Brief: [Proposal Title]

Executive Evidentiary Summary (Max 200 words, including an overall Claim Reliability Score out of 100)

Claim-by-Claim Audit Table (Columns: Claim ID, Original Text, Evidentiary Status, Source Quality, Auditor Finding)

Critical Vulnerability Analysis (Top 3-5 evidentiary liabilities and statutory risks)

Remediation and Phrasing Adjustments (Side-by-side original vs. recommended replacement text)

Self-review

  1. Did I verify that every causal claim is scrutinized separately from pure descriptive statistics?
  2. Are all flagged evidentiary gaps tied directly to {{citation_standards}}?
  3. Is the remediation text realistic and legally sound within {{primary_jurisdiction}}?
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
public-sector-nonprofit
policy analysis
fact-checking
public sector