surfaceanalysisweaknesses
Analysisscalar · returns json

WEAKNESSES

Identify logical weaknesses and gaps in an argument

Per-row — runs once for each row.

analysisllmtext

Syntax

WEAKNESSES({{ argument }})
WEAKNESSES({{ argument }}, '{{ severity }}')
{{ argument }} WEAKNESSES

Arguments

nametypedescription
argumentVARCHAR
severity(optional)VARCHARMinimum severity: 'minor', 'moderate', 'major', or 'all'

About

Identify logical weaknesses, gaps, and vulnerabilities in an argument. Finds where reasoning breaks down - logical fallacies, unsupported leaps, missing evidence, scope problems, and other flaws that undermine validity. Perfect for: - Peer review (what's wrong with this paper?) - Due diligence (where could this proposal fail?) - Self-critique (what am I missing?) - Debate prep (where will they attack?)

Examples

Identifies sampling bias / hasty generalization

SELECT
  weaknesses (
    'Everyone I know loves this restaurant, so it must be the best in the city'
  )

Identifies post hoc fallacy / correlation vs causation

SELECT
  weaknesses (
    'Sales increased after we changed the logo, proving the rebrand worked'
  )

Nearby rabbit holes

same domain
Climb back to The Looking Glass