---
name: bias-truth-judgment-belief-bias
description: "Use Belief Bias as a focused cognitive-bias lens. Explain the pattern, check whether it plausibly fits the situation, surface an alternative non-bias explanation, and return a practical counter-check without diagnosing people or claiming causation."
---

# Belief Bias

Do you only find arguments convincing when their conclusion aligns with your beliefs? That’s Belief Bias – the tendency to judge an argument’s strength based on how believable its conclusion seems, rather than its actual logic.

## When to use

- The user names Belief Bias or asks whether this pattern may matter in a real situation.
- The agent needs a focused lens for reviewing a decision, claim, estimate, memory, judgment or interaction.
- The user wants a practical check rather than a label applied to a person.

## Workflow

1. Restate the relevant situation or claim in neutral terms before applying the bias label.
2. Explain Belief Bias in plain language using the canonical Cognitive Biases entry as the source of truth.
3. Identify the specific observation that makes this lens plausible. If no concrete observation exists, say that the fit is weak or unknown.
4. Give at least one ordinary alternative explanation that does not require Belief Bias, such as incentives, constraints, missing information, chance, measurement error or a different cognitive mechanism.
5. Check the canonical page for qualification, evidence status and related concepts when source access is available. Preserve uncertainty instead of upgrading the claim.
6. Suggest one or two practical counter-checks that could reduce the risk or distinguish this explanation from alternatives.
7. End with what would make the Belief Bias interpretation more likely, less likely, or still unresolved.

## Required output

- Plain-language explanation
- Observed signal or reason the lens was considered
- Alternative non-bias explanation
- Evidence / uncertainty boundary
- Practical counter-check
- What would change the assessment

## Evidence and safety boundaries

- Treat Belief Bias as a candidate lens, not a diagnosis, personality judgment, intent claim or proof of causation.
- Do not infer intelligence, character, mental health or motives from the presence of a cognitive-bias pattern.
- Do not present a generated, disputed, limited or unreviewed claim as settled science. Preserve the status shown by the canonical project material.
- Do not force this bias to fit when another explanation is simpler or better supported.
- For medical, legal, financial or mental-health decisions, keep the skill educational and preserve the need for appropriate professional judgment.

## Canonical bias

- [Belief Bias](https://cognitive-biases.github.io/biases/truth-judgment-belief-bias/)
- [Public bias dataset](https://cognitive-biases.github.io/data/biases.json)
- Category: Truth Judgment

## Portability

This is a free, instruction-only Agent Skill. It requires no secrets, no executable scripts and no network access to run. If source-access tools are available, use the linked Cognitive Biases material to preserve the current wording, review state and uncertainty.

Licence: CC-BY-NC-SA-4.0.