From cognition
Use when someone presents a chain of reasoning or a story explaining why something happened, and the logic may have a structural flaw or the narrative may be too tidy. Covers logical fallacies (false dichotomy, post hoc, straw man) and narrative fallacy — the tendency to construct a coherent story over random events. Triggers on: "is this argument valid", "check my logic", "논리 오류", "이 주장이 맞나", "왜 이렇게 됐는지 설명해줘", "이 논리에 문제가 있나". Best for: evaluating arguments before acting on them, auditing explanatory narratives, debate preparation. Not for: cognitive bias diagnosis (use bias-auditor), confidence calibration (use epistemic-reasoner).
How this skill is triggered — by the user, by Claude, or both
Slash command
/cognition:fallacy-detectorThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
**Use when:**
Use when:
Not for:
Step 1 — State the argument or story. Confirm the core claim being analyzed.
Step 2 — Scan for logical fallacies. Flag only what's actually present.
| Category | Fallacies |
|---|---|
| Structural | False dichotomy, slippery slope, circular reasoning, straw man |
| Authority/Social | Ad hominem, appeal to authority, bandwagon |
| Causation | Post hoc ergo propter hoc, cum hoc, hasty generalization |
Step 3 — Scan for narrative fallacy. Separate scan — check for:
Step 4 — Deliver the analysis. If no fallacies are present, say so.
논리적 오류 / Logical Fallacies Detected:
[Fallacy name]: [Where it appears in the argument, specifically]
Why this matters: [What it invalidates or weakens in the conclusion]
서사 오류 / Narrative Fallacy Assessment:
[Present or absent. If present: which element of the story appears constructed rather than accurate]
핵심 문제 / Core Issue:
[The single most damaging error — the one that most undermines the argument's validity]
| Claude | You |
|---|---|
| Scans systematically across fallacy categories | Provide the argument or explanatory story |
| Names only fallacies that are actually present | Confirm whether the identified pattern applies |
| Explains why each fallacy undermines the conclusion | Decide whether to revise, discard, or defend the argument |
| Runs a separate narrative fallacy scan on explanatory stories | Update your reasoning based on the structural findings |
Key constraint: Naming a fallacy without explaining which step is unjustified and why is taxonomy, not analysis.
bias-auditor — for cognitive biases rather than argument structureepistemic-reasoner — for calibrating confidence to evidenceassumption-extractor — for surfacing hidden premises beneath argumentsnpx claudepluginhub newkayak12/claude-skills --plugin cognitionProduces a complete logic report on any argument, plan, or reasoning — validates premises, tests inference, detects fallacies, and surfaces hidden assumptions.
Analyzes claims by mapping arguments, auditing evidence quality, detecting logical fallacies and biases, and issuing verdicts. For evaluating research or technical arguments.
Analyzes documents for logically invalid inferences and logical fallacies, running three independent passes with consolidated severity-ranked results.