SuperAPI AI Directory

evaluate-evidence

Writing & research

View repository

Updated 10-16 06:26

About this skill

Follow these steps:

1. Identify the claim precisely

State what is being asserted with enough specificity to be testable. Separate bundled claims into distinct assertions.

2. Inventory the evidence

Catalog what evidence is actually offered:

  • Direct or indirect support
  • Primary or secondary source
  • Quantitative or qualitative data

3. Assess evidence quality

  • Source reliability: expertise and potential bias
  • Data quality: sample size and methodology
  • Currency: how recent
  • Convergence: do multiple independent sources agree
  • Directness: does it support the claim directly or require inferential leaps

4. Look for counter-evidence

Actively seek contradicting evidence, alternative explanations for the same observations, and relevant evidence that may have been omitted.

5. Rate the support level

  • Strong: multiple high-quality direct sources converge
  • Moderate: solid evidence with some gaps or conflicts
  • Weak: limited, indirect, or questionable evidence
  • Unsupported: speculation lacking substantive evidence

6. Note confidence boundaries

Identify what assumptions must hold for the claim to be valid, and what additional evidence would meaningfully strengthen or weaken it.